DOE/ER-0446P

uman
enome

y

QH

447

.H85

\u I ■

1989-90

Program
Report

It'^^^^^^^Bm

United

States

Department

of

Energy

Office
of

Energy
Researcli

Office

of

Health

and

Environmental

Research

JVIarch 1990

Please address queries on this
publication to:

Dr. Benjamin J. Barnhart

Manager, Human Genome Program

Office of Health and Environmental Research

U.S. Department of Energy

ER-72 GTN

Washington, DC 20585

(301 ) .353-5037, FTS 233-.5037

FAX: (30 1 ) 353-505 1 , FTS 233-505 1

Human Genome Management
Information Sy.stem

Oak Ridge National Laboratory

P.O. Box 2008

Oak Ridge. TN 3783 1 -6050

(615) 576-6669. FTS 626-6669

FAX: (615)574-9888. FTS 624-9888

m.

.-4-

This report has been reproduced directly from the best available copy.

Available from the National Technical Information Service, U.S. Department of Commerce, Springfield, Virginia 22161.

Price: Printed Copy AGS
Microfiche AOl

Codes are used for pricing all publications. The code is determined by the number of pages in the publication. Information

pertaining to the pricing codes can be found in the current issues of the following publications, which are generally avail-

ble in most libraries: Energy Research Abstracts, (ERA); Government Reports Announcements and Index (GRA and I);

Scientific and Technical Abstract Reports (STAR); and publication, NTIS-PR-360 available from (NTIS) at the above address.

o

(3~

DOE/ER-0446P

Human
Genome

1989-90
Program Report

O:

m-
11

i CO

i i-n
i nj

I D

I °

i □
; m
! □

Date Published: March 1990

J

D
>

U.S. Department of Energy

Office of Energy Research

Office of Health and Environmental Research

Washington, D.C. 20585

Major Events in the Development of the DOE
Human Genome Program

Santa Fe Meeting

Human Genome Initiative announced

Pilot projects pursued at national laboratories:

• Computer modeling and optimization of library ordering strategies;

• Database management support;

• Advanced instrumentation and vectors for ordering and sequencing;

• Isolation of centromere, telomere, and chromosome-specific clones; and

• Production of large DNA insert libraries of chromosomes

HERAC Report on the DOE Human Genome Initiative

Interagency workshops on resources and informatics initiated

Designation of Lawrence Berkeley Laboratory and Los Alamos National
Laboratory as Centers

Initiative management and program plans finalized

Special peer review panels for research proposals
1988 primary R&D awards
Small Business Innovative Research (SBIR) awards
DOE/NIH Memorandum of Understanding

Human Genome Coordinating Committee formed

Human Genome Management
Information System initiated

1989 primary R&D awards

SIBR awards

DOE/NIH Five-Year Plan Workshop

Program plan update for 1990

1990 program announcements

DOE/NIH working groups for informatics,
mapping, and ethical issues

Contractor-Grantee Workshop
Large Insert Cloning Workshop
Human X Chromosome Workshop

DOE/NIH Five-Year Plan for
the Human Genome Project

1990 primary R&D awards
1986 1987 1988 1989 1990

Preface

This nation's Human Genome Project is the first broadly based organized endeavor in
the biological sciences. Conceived in 1986. the program was initiated in 1987 as the
Human Genome Program of the U.S. Department of Energy (DOE) Office of Health
and Environmental Research (OHER). Since that time, it has grown significantly; in
1988 the National Institutes of Health also niitiated a human genome program. An
ambitious undertaking spanning the disciplines of biology, chemistry, physics,
engineering, mathematics, and information science, the national project has a well-
defined, long-range endpoint: to decipher the genetic code in the DNA of the entire
human genome. Having gathered support in Congress, the Executive Office, and the
scientific and commercial sectors, the project of mapping and sequencing the genome
has the momentum to make major advances in the knowledge and technologies that are
needed to understand the complexities of human cellular processes in a manner never
before possible. These advances will impact biological principles as well as the practice
of medicine, the growing biotechnology industry, and society.

The project is unusual in that few existing strategies and technologies can be used to
achieve its goals. Indeed, the driving force within this endeavor is the development and
implementation of innovative, cost-effective methods and technologies for mapping,
sequencing, and interpreting the genome. As the.se developments take place, advances
in data analysis and database management will make map and sequence information
accessible.

This document is a status report on DOE OHER's Human Genome Program and
includes a brief background to this agency's initiative, as well as an explanation of the
program's projected focus over the next 15 years. Of special interest are the section on
research highlights, the narratives on major genome research efforts conducted at three
of DOE's national laboratories, and the abstracts of work in progress. Figures and
captions provided by investigators give additional detailed information.

Achievements were reported at the first DOE OHER workshop for grantees and
contractors of the Human Genome Program on November 4 and 5. 1989, in Santa Fe,
New Mexico. The presentations demonstrated that progress is being made in physical
mapping, DNA sequencing, and infonnatics. DOE plans to convene these workshops on
a continuing basis every 18 to 24 months. In the interim, this report and future revisions,
together with the bimonthly newsletter and special technical overviews will provide
both the interested scientist and layperson with information on developments in this
rapidly moving, multidisciplinary project.

Benjamin J. Bamhart, Manager

Human Genome Program

Office of Health and Environmental Research

Office of Energy Research

U.S. Department of Energy

111

Acknowledgements

The Office of Health and Environmental Research gratefully acknowledges the
contributions made by genome research grantees and contractors in submitting
abstracts, photographs, captions, and narratives, Charles Cantor and Sylvia Spengler
of Lawrence Berkeley Laboratory provided the material adapted for the "Primer on
Molecular Genetics"' in Appendix A. "The Alta Summit, December 1984." by Robert
Mullan Cook-Deegan was reprinted with the pennission of Gt'iiiniiics. which is
published by Academic Press. Inc, The glossary in Appendix C was adapted from
the glossary in the April 19S8 U,S. Congress, Office of Technology Assessment
publication: Mappini; Our Genes — The Genome Projects: How Bii;'.' How fast?
Finally, the Human Genome Management Infonnation System at Oak Ridge
National Laboratory (operated by Martin Marietta Energy Systems. Inc.. for the
U.S. Department of Energy under contract DE-AC05-84OR21400) is recognized for
collecting and organizing the information, preparing the manuscript, and implementing
the design and production of this publication.

IV

Contents

Introduction to the DOE Human Genome Program 1

Development of the Human Genome Initiative 2

The HERAC Recommendation 3

OHER Mission 8

Management of the Human (Jenome Program 14

Program Management Structure 14

Program Management Task Group 14

Human Genome Coordinating Committee 15

Interagency Coordination 17

Joint DOE/NIH Activities 18

International Human Genome Organisation 19

Resource Allocation 20

Continuing Implementation ". 21

Short-Term Focus (1-5 Years) 21

Mid-Term Focus (5-10 Years) 21

Long-Temi Focus ( 10-15 Years) 21

Research Highlights 24

Research Facility Narratives 32

Lawrence Berkeley Laboratory 32

Lawrence Livermore National Laboratory 36

Los Alamos National Laboratory 42

Human Genome
1989-90 Program Report

Contents

Abstracts of DOE-Funded Research 49

Project Categories and Principal Investigators 30

Resource Development 52

Physical Mapping 62

Mapping Instrumentation 77

Sequencing Technologies 85

Infomiatics 96

Small Business Innovative Research (SBIR) — Phase I
(1989 Awards) 105

Small Business Innovative Research (SBIR) — Phase II
(1988. 1989 Awards) Ill

Index to Project Investigators 1 15

Appendices 1 19

A. Primer on Molecular Genetics 121

B. "The Alta Summit. December 1984"

(reprinted from Genomics) 139

C. Glossary 145

Acronym List (Inside back cover)

Introduction to the DOE
Human Genome Program

The structural characterization of genes and the elucidation of their encoded
functions have become a cornerstone of modem health research, biology, and
biotechnology. A genome program is an organized effort to characterize all the
genetic material — DNA — of an organism. The human genome encodes 50.000 to
100.000 genes on its 24 distinct chromosomes, but only some 2000 genes are now
available for study in the forni of identified cloned DNAs. To accelerate effective
access to all human genes and eventually togenerate reference DNA sequences of the
chromosomes, the U.S. Department of Energy (DOE) began a Human Genome
Initiative in 1986. Intensive studies of the needs and promise of genomics research then
ensued. The value of a broad, supportive infrastructure for genomics has now been
recognized in the United States and abroad. Genome projects on several important
organisms are now planned or are progressing. Two federal agencies are now
supporting an expanded national human genome project, and the coordination of
human genome research has become international in scope.

Sequencing Technologies

Resonance ionization mass spectrometer. Investigators are
shown with the resonance ionization mass spectrometer that is
used to measure stable isotopes of a variety of elements that
are attached to DNA. Since 50 or more stable isotopes are
available for such DNA labeling, a multiplex procedure is
available in which either the single radioisotope or the four
fluorescent labels are replaced. When performing sequence
analysis by using four stable isotopes simultaneously on the
four dideoxynucleotide-terminated DNA fragments, the gel
lanes needed for electrophoresis can be reduced from the
usual four lanes to one lane. Many such elements can be used
simultaneously in the same electrophoresis lane since the
resonance ionization mass spectrometer will sort out the
elements and all their isotopes. Even greater multiplexing
would occur when Church probes are used to locate DNA
fragments, since all hybridization could be performed
simultaneously. Studies are under way to label DNA with a
variety of elements that have multiple stable isotopes and to
detect the DNA so labeled. These studies are being carried out
at the Oak Ridge National Laboratory in collaboration with
Atom Sciences. Inc., where this mass spectrometer has been
developed to its present state. (Photograph provided by Bruce
Jacobson, Oak Ridge National Laboratory, and Heinrich
Arlinghaus, Atom Sciences, Inc.)

Introduction

The Office of Health and Environmental Research (OHER) manages the Human
Genome Program within DOE. as a focused program of resource and technology
development. The general objective is to advance, as economically and efficiently as
possible, the U.S. effort in human genome research. Other OHER responsibilities, such
as assessing the effects of energy by-products on our population and environment, will
benefit from applications of the new biological and computational resources and the
innovative technologies developed within the genome program. Moreover, the resources
and technologies being developed are of broad and continuing value: many facets of
biomedical research, modem agriculture, and the growing biotechnology industries will
be advanced.

Knowledge of principles guiding the three-dimensional structure-function relationships
of cellular macromolecules is essential for the interpretation and application of the
linear DNA sequence that is being elucidated in the Human Genome Program. Special
facilities in DOE laboratories are being used (see table on p. 3) to determine the three-
dimensional structure of macromolecules. DOE is planning for the greatly increased
demand that will be placed on these facilities as a result of the Human Genome
Program.

Development of the Human Genome Initiative

In 1984 DOE and the International Commission for Protection Against Environmental
Mutagens and Carcinogens cosponsored a workshop in Alta, Utah. A specific charge to
the participants was to evaluate the present state of and project future directions for
mutation detection and characterization. The growing roles of novel DNA technologies
in diagnostics were highlighted. There, in the excitement of the meeting, some core
techniques of current genomic analysis were conceived (see "The Alta Summit.
December 1984," in Appendix B). These new approaches increasingly included gene
cloning and sequencing. Although the isolation of genes from libraries of clones had
been an integral component of biomedical research for many years, the one-gene-at-a-
time procedures being employed were wasteful of scientists" time and research
resources.

The small genomes of several viruses were the targets of the first genome projects in the
1960s. These projects initiated the development of some of the current important
techniques in molecular genetics. (A molecular genetics primer is included as Appendix
A of this report. ) With the advent of molecular cloning techniques in the 1 970s. a
library of manageable cloned DNAs could be produced for any species. Genome studies
on many viruses, the bacterium Escherichia coli, two yeast species, and a nematode (a
minute worm) were subsequently implemented. With the skills already demonstrated, a
human genome program could then be considered. However, it would be a task far more
vast than any previously implemented in biological research.

In 1985 DOE began to consider whether its expertise with high-technology projects
could facilitate and sustain a human genome program. To as.sess the desirability and

feasibility of ordering and sequencing clones representing the entire human genome.
DOE sponsored in March 1986 an international meeting in Santa Fe. New Mexico. With
virtual unanimit\ the participating experts concluded that this objective was meritorious
and obtainable and that it would be an outstanding achievement in modern biology.

The HERAC Recommendation

Further guidance was sought from DOE"s Health and Environmental Research Advisory
Committee (HERAC). which provided its report on the Human Genome Initiative in
April 1987. This report urged DOE and the nation to commit to a large, multi-
disciplinary, technological undertaking to order and sequence the human genome.
This effort would first require significant innovation in the general capability to
manipulate DNA. Also required would be major new analytical methods for ordering
and sequencing DNA segments, theoretical developments in computer science and
mathematical biology, and great expansion in the ability to store and manipulate the
information and interface it with other large and diverse genetic databases. The report
further recommended that DOE have a leadership role because of its demonstrated
expertise in managing complex, long-term multidisciplinary projects, involving
both the development of new technologies and coordination of efforts among industries,
universities, and its own laboratories.

The role of the Office of Health and
Environmental Research (OHER) in
its mission to understand the health
effects of radiation and other by-
products of energy production was
noted in the report. This mission
requires fundamental knowledge
of the effects of damage to the
genome, and it has already led
to a number of research and
technological developments that are
integral components of the human
genome mapping and sequencing
project: DOE computer and data
management expertise initiated and
supports the GenBank" DNA
sequence repository (cosponsored
by the National Institutes of
Health); the chromosome-sorting
facilities essential to the Genome
Initiative were developed and are
maintained at DOE laboratories:
and within the National Laboratory
Gene Library Project, libraries of

Major DOE Facilities and Resources Relevant
to Molecular Biology Research

Center for X-Ray Optics

GenBank" Data Sequence Repository

High Flux Beam Reactor

Los Alamos Neutron Scattering Center

Molecular Sciences Research Center

National Flow Cytometry Resource

National Laboratory Gene Library Project

Protein Structure Data Bank

National Synchrotron Light Source

Scanning Transmission Electron Microscope Resource BNL

Scanning Tunneling Microscopy LLNL, ORNL

Stanford Synchrotron Radiation Laboratory Stanford

LBL
LANL

BNL
LANL

PNL

LANL

LANL, LLNL

BNL

BNL

Developing Facilities:

Advanced Photon Source
Advanced Light Source

ANL
LBL

Introduction

cloned sequences from single human chromosomes are produced for the research
community. Thus, the Human Genome Initiative was a natural outgrowth of ongoing
DOE-supported research.

OHER responded to the Santa Fe meeting and HERAC reports by implementing three
major objectives that are being pursued concurrently:

1. The generation of refined physical maps of the chromosomes, including the ordering
of representative libraries of DNA clones.

2. The development of requisite supportive strategies, chromosomal resources, and
instrumentation, which includes development and testing of advanced sequencing
technologies.

3. The expansion of communication networks and computational and database
capacities and the development of advanced algorithms for managing and
interpreting the clone ordering and sequence data.

A small number of genes or other selected regions of interest will be sequenced during
the generation of physical maps; however, the transition from mapping to an intensive
genome-sequencing effort awaits development of more accurate, rapid, and economical
technologies that are needed to commence large-scale sequencing. Once suitable new
technologies are implemented, contiguous segments of DNA will be decoded into a
reference .sequence of the human genome. These segments will be derived from ordered
clones and DNA fragments that are identified or mapped to particular locations on
chromosomes by sequence-tagged sites (STSs) or other methodologies. Emerging
methodologies will be tested and validated in pilot sequencing projects prior to
incorporating any single protocol as the primary method.

As implementation of the OHER program began with a small number of pilot projects,
other government agencies, scientific societies, and commercial organizations initiated
their own studies of associated policy and strategy issues and presented their
recommendations. The most prominent reports are those of the National Research
Council (NRC) and the Congressional Office of Technology Assessment (OTA). While
broadly in accordance with the earlier HERAC recommendations, the NRC and OTA
reports further recommended that several nonhuman species also be included in the
national effort and that the physical mapping of chromosomes be complemented by
genetic mapping.

The OHER human genome program remains focused on physical map construction and
development of advanced sequencing methodologies and technologies. Because many
of the resources and technologies being developed have broad applicability, they
contribute substantially to OHER programmatic objectives in the fields of radiation
biology, chemical toxicology, molecular epidemiology, and the ecological and
environmental biosciences and aid the developing genome programs of other agencies
as well.

Mapping Instrumentation

DNA-Protein-Binding Assay — Use of gel retardation for recognition of
promoter sequences that are necessary for the polymerase enzyme to
synthesize DNA for sequencing studies. Increasing quantities of T7 RNA
polymerase were incubated withi pET-1 DNA (contains a strong T7 promoter) and
pAT153 DNA (no promoter) for 10 min at room temperature; samples were then
electrophioresed in a 1% agarose gel for 2 hr at 2.5 V/cm. As the ratio of
polymerase molecules to DNA increased, the quantity of pET-1 DNA band
decreased, and a new band with lower mobility appeared. Without requiring
quantitative loading of the gel. the ratio of fluorescence in the pET-1 band to that
in the control pAT153 band permits quantitation of the fraction of the pET-1
molecules bound to polymerase. This is a useful technique for finding specific
promoter sequences; once found, these promoter sequences can be attached to
DNA fragments of choice (e.g.. fragments the researcher is interested in
sequencing). (Photograph provided by Betsy Sutherland, Brookhaven National
Laboratory.)

Informatics

Data access through interactive workstations. One of the great challenges
of the human genome project is how to integrate and provide access to the
growing mass of genomic data. One solution is development of sophisticated
worl^stations that would provide a uniform user interface with all map and
sequence databases. With the prototype developed at Lawrence Berkeley
Laboratory's Human Genome Center, the user examines data at increasing
resolution by "enlarging " selected regions of successive displays. The three
illustrations shown here display (a) the full complement of chromosomes, (b) a
single chromosome with locations of human disease genes, and (c) the
nucleotide sequence for a selected region within that chromosome. Within this
region, the order of the nucleotide bases is displayed by the following colors: A
(chartreuse), C (orange), G (light blue), and T (pink). The icons at the bottom or
side of figures a-c indicate access to other levels of information about each
chromosome including staining, gene mapping, morbidity (disease), and
sequence. In figures a and b, the dark blue bands indicate the characteristic
Giemsa staining patterns; the chromosome centromeres are pink; and the yellow
areas on the chromosomes represent the heterochromatic regions (C-banding
pattern). Less characterized areas are light blue. (Photographs a-c provided by
the Human Genome Center, Lawrence Berkeley Laboratory.)

Informatics

GnomeView Workstation. The mosaic shown in the bottom photograph
illustrates the versatility of the X-window system that is part of the GnomeView
Interface currently in use at the U.S. Department of Energy's Pacific Northwest
Laboratory. Shown on the same screen are simultaneous views of chromo-
somes at various magnifications (upper screen), restriction maps (windows on
lower left), two magnification levels of a sequence from GenBank' (windows on
lower right), and a GenBank" file information header (text window, lower
screen). The X-window system, coupled with the network model database
system of the GnomeView Interface, allows easy access and simultaneous
viewing of information from all levels of the human genome hierarchy.
(Photograph provided by Richard Douthart, Pacific Northwest Laboratory.)

j^b)Morbid anatomy: Chromosome 1^
I — Colorectal cancer ^ '
— _|| — Miller-Dieker lissencephaly syndroi

I p von Recklinghausen neurofibromat

I [Galaciokinase deficiency

Growth hormone deficiency
niig type lA; Kowarski type

I i I Ehlers-Danlos syndrome type Vn Al

Osteogenesis imperfecta (2 or more forms)
; Marfan syndrome, atypical

BlluTJwaiifflRra^^BnfflfflBn

Glanzmann thrombasthenia
Pompe disease

Adult a^id^maHaoo Aafinian

^yelop

Acanth
Apolipc
Pieman

■

' ',

( |irx>iii>»oitir .;

■ '

a

1

el*

1

1

•a:

■
1

4

■4. ■..
1.4

■

It,™ B •, X

n..'

B

^^—

ir: .

^9 ^^^^^1

■

■1

■

I Itrvtitioaonie II

OHER Mission

The Office of Health and Environmental Research (OHER) has research and
development responsibilities that are mandated by 1946 and 1954 legislative
acts (see "Enabling Legislation" on p. 10). some of which have been carried
forward from DOE's predecessor agencies, the Atomic Energy Commission (AEC) and
the Energy Research and Development Administration (ERDA). The first national
support for genetics research was provided by AEC: further responsibilities were
authorized in 1974 and 1977. Although the initial focus was on radiation effects, the
objectives were later broadened to include the health consequences of all energy
technologies and their by-products. Long-range goals are to address applications of the
resources and technologies developed in the genome program to the Department's
interests in genetic damage from exposures to ionizing radiation and chemicals. An
extensive program of OHER-sponsored research on genome structure, maintenance,
damage, and repair continues at the national laboratories and universities.

A major concern today is human exposure to background environmental factors and
how the body responds to such factors. In the environment there are unavoidable
genome-damaging agents from which we are at risk. Among them are natural radiation
sources, which include components of sunlight, cosmic rays from space, and the radon
released from the Earth. There are both inorganic and organic chemicals that can cause
DNA damage. Some of these chemicals are natural to the environment, while others are
generated by human commerce and energy-related processes.

-'cts,,

CH6S9J PC

»AC-

016S3
016580
01658)
016SB2

I

6S51
iS97

016SI9
DI6&q]
016S98

0i6Sa D.

016S79
016596

S.30

CtSM

Ol6S7i

016SI01

D>65«
DI6S9I
0I6SIU

Physical Mapping

Diagram of human chromosome 16 showing the
G-banding (Giemsa-stalning) pattern. On the left of the
figure are the points at which chromosome breaks have
been defined. A correlation has been found between the
occurrence of these breaks and other chromosome
anomalies, such as translocations, deletions, or fragile sites
(the latter designated by prefix FRA). Mouse/human somatic
hybrids (designated by prefix CY) have been constructed by
transferring a portion of chromosome 1 6 to a mouse cell line.
On the right of the figure are names of cloned DNA
fragments (probes) from human chromosome 16. The
fragments have been mapped to the defined regions of this
chromosome by Southern blot analysis of DNA from the
somatic cell hybrids and by in situ hybridization. The DNA
probes are either anonymous cloned fragments of DNA or
cloned genes. (Figure provided by Grant Sutherland, North
Adelaid Children's Hospital, Australia.)

Normal biological activities also contribute to the risk of genetic damage. A body's own
cells produce some potentially damaging molecules in the course of normal metabolic
processes; some of these molecules are produced in considerable abundance during
defensive actions against microbes and during detoxification of harmful environmental
substances. The genome replication system itself sometimes errs during cell prolifer-
ation. Even DNA is not completely stable chemically: its nomial methyl-cytosine
constituent has a low but measurable rate of spontaneous mutagenic change.

Life has thus evolved under a continuous low-level infliction of genomic damage and
mutation. Under this pressure, systems that reverse or ameliorate many types of DNA
damage have evolved, so that a wide range of repair mechanisms exists within cells of
all species. Several of the human genes contributing to DNA repair processes are being
characterized now, and others await detection and molecular cloning. Repair gene
deficiencies are manifested as cellular sensitivity to low-level DNA damage and in
diseases such as cancer. Humans exhibit genetic diversity in capacity for DNA repair in
response to ubiquitous DNA-damaging agents.

In recognition of this diversity, a major goal of the current OHER health effects and
general life sciences program areas has been formulated: the development of capacities
to diagnose individual susceptibility to genome damage imposed by energy-related
factors. Some major components of this OHER research are:

• molecular cloning and characterization of DNA repair genes;

• improvement of methodologies and development of new resources for use in
quantitating and characterizing mutations (molecular epidemiology); and. most
recently,

• focused resource and technology development needed to map and sequence the
human aenome — the Human Genome Program.

OHER Mission

Enabling Legislation

The Atomic Energy Act of 1946 (P.L. 79-585) provided the
initial charter for a comprehensive program of research and
development related to the utilization of fissionable and
radioactive materials for medical, biological, and health
purposes.

The Atomic Energy Act of 1954 (P.L. 83-703) further
authorized the AEC "to conduct research on the biologic
effects of ionizing radiation."

The Energy Reorganization Act of 1974 (P.L. 93-438) provided
that responsibilities of the Energy Research and Development
Administration (ERDA) shall include "engaging in and
supporting environmental, biomedical, physical and safety
research related to the development of energy resources and
utilization technologies."

The Federal Non-nuclear Energy Research and Development
Act of 1974 (P.L. 93-577) authorized ERDA to conduct a
comprehensive non-nuclear energy research, development,
and demonstration program to include the environmental and
social consequences of the various technologies.

The DOE Organization Act of 1977 (P.L. 95-91) mandated the
Department "to assure incorporation of national environmental
protection goals in the formulation and implementation of
energy programs; and to advance the goal of restoring,
protecting, and enhancing environmental quality, and assuring
public health and safety," and to conduct "a comprehensive
program of research and development on the environmental
effects of energy technology and program."

10

Physical Mapping

Researchers comparing photographs of gels in which restriction fragments
have been separated. The technology being developed includes a reliable
method for producing a partial digest of DNA in agarose. (Photograph provided by
Michael McClelland, California Institute of Biological Research.)

11

Sequencing Technologies

Application of flow cytometry to DNA sequencing. The small yellow spot In
the center of this multiple-exposure photograph shows the fluorescence from
approximately 1 000 molecules of the laser dye — rhodamine 6G. The apparatus
shown is a modified flow cytometer with the green argon laser beam traversing
from left to right: the flow cuvette is vertical in the center. The fluorescence
collection lens can be seen in the background. An apparatus similar to the one
shown in the photograph is being developed to sequence DNA by detection of
single, fluorescent molecules. (Photograph provided by James Jett, Los Alamos
National Laboratory.)

12

11 I

Management of the Human Genome Program

Program Management Structure

The highly multidisciplinary, focused, and long-temi character of the Human
Genome Program is novel to biological research. An infrastructure connecting
biomedical research, technology development, computer .sciences, data and
physical repositories, and supporting agencies has thus become essential. The Health
and Environmental Research Advisory Committee (HERAC) provides policy, strategy,
and scientific guidance to the program.

Program Management Task Group

Within DOE, the management structure recommended by HERAC is that the Human
Genome Program Manager and Management Task Group work within the Office of
Health and Environmental Research to coordinate:

• Independent scientific boards that
prospective and retrospective eval

DOE Human Genome Program
Management and Coordination

HERAC

Review
panels

Office of

Energy

Research

Other

funding

agencies

Office of Health and

Environmental

Research

Human Genome Program
f\/lanagement Task Group

Information
system

Coordinating
Committee

Task groups

Projects at universities,

national laboratories,

and industrial institutions

provide peer review of research proposals; both
nations are utilized.

• Administration of awards,
collaboration with all concerned
agencies and organizations,
organization of periodic workshops,
and responses to the needs of the
developing program.

• The support services provided by
the Human Genome Management
Information System (HGMIS) at
Oak Ridge National Laboratory. As
a DOE management tool to facilitate
communications among manage-
ment and research personnel and to
update the public on genome
research. HGMIS publishes
newsletters and technical and other
program reports and maintains an
electronic bulletin board that carries
current infomiation and announce-
ments. The on-line bulletin board
and publications are available to all
persons interested in the genome
project.

14

Human Genome Coordinating Committee

Another component of the DOE management structure is the Human Genome
Coordinating Committee (HGCC), which was chartered by HERAC. The committee,
originally named the Human Genome Steering Committee, was formed in October
1988. The HGCC membership comprises and represents DOE genome program
research participants. Members of the Human Genome Program Management Task
Group (ex-officio members of the HGCC) and observers from other government and
private agencies participate in the regularly scheduled meetings of the HGCC. whose
responsibilities include:

• assisting OHER with the overall coordination of DOE-funded genome research,

• facilitating the development and dissemination of novel genome technologies.

• ensuring proper management of data and samples.

• interacting with other national and
international efforts.

' communicating the program to the
press and public, and

' establishing task groups to
analyze specific issues such as
ethics, informatics, resource
sharing, cost of resource
distribution, and use of
chromosome-tlow-sorting
facilities.

Human Genome Coordinating Committee

Chairman: Charles R. Cantor, Director, Human Genome
Center, Lawrence Berkeley Laboratory

Anthony V. Carrano, Director, Human Genome Project,
Lawrence Livermore National Laboratory

C. Thomas Caskey, Director, Institute for Molecular
Genetics, Baylor College of Medicine

Leroy E. Hood, Director, Center for Integrated Protein and
Nucleic Acid Chemistry and Biological Computation, and
Director, Cancer Center, California Institute of
Technology

Robert K. Moyzis, Director, Center for Human Genome
Studies, Los Alamos National Laboratory

HGCC Executive Officer: Sylvia J. Spengler, Lawrence
Berkeley Laboratory

15

Management of the
Human Genome
Program

Physical Mapping

Human DNA fragments obtained
from rodent somatic cell hybrid
background separated on agarose
gels. A primer designed to recognize
human Alu sequences is used for
rapid amplification of regions of
fiuman DNA in rodent/fnuman somatic
cell hiybrids^ Rodent/human hybrid
cells are constructed and used in
human genome studies because they
contain manageable amounts of
human DNA in which genome
regions of interest can be manip-
ulated and characterized. The TC-65
oligonucleotide primer was designed
to recognize the human, but not the
rodent, Alu sequences and provides
specific amplification of human DNA
between regions of these ubiquitous
Alu sequences when polymerase
chain reaction (PCR) methods are
used. [Alu sequences are about
300 bp long and repeated thousands
of times in the human genome.) The
specificity of the TC-65 primer is
demonstrated in the figure: DNA
fragments of total human genome
(lane 2) and fragments of different
rodent/human hybrid cell lines have
been amplified and separated by gel
electrophoresis (lanes 3-12). Note
the abundance of bands (white) of
DNA fragments in lanes 2-12 and the lack of fragments in lanes 13 and 14, where pure rodent genome samples were
electrophoresed. No human Alu repeat sequences are found in the rodent genomes, and the rodent Alu equivalent
sequences are not amplified; this TC-65 phmer/PCR method is thereby validated. Lane 1 contains standard DNA fragments
of known size for determining sizes of DNA fragments in the other lanes.

This method is useful for rapid comparison of hybrid cell lines' DNA content and overlap and can also be used in
preparation of nucleic acid probes from cloned human DNAs — especially for clones in yeast artificial chromosome (YAC)
vectors. [Photograph was first published in Proc. Natl. Acad. Sci. USA 86. 6686-6690 (1989). Photograph provided by
David L Nelson and C. Thomas Caskey, Baylor College of Medicine.]

TC-65

16

Interagency Coordination

The U.S. agencies engaged in genome research meet fomially under the auspices
of the White House Office of Scientific and Teclinology Policy. The
Department of Agriculture is initiating a genomics program. The National
Science Foundation has computational and informatics programs supportive of
genomics in addition to individual awards in genetics and molecular biology. The
Howard Hughes Medical Institute, a private foundation, contributes substantially to the
genome effort through its support of biomedical research and related infrastructure. The
National Institutes of Health (NIH) started its own genome program in 1988. The NIH
program complements the DOE program by supporting predoctoral and postdoctoral
training in molecular genetics, studying model organisms, emphasizing genetic
diseases, and preparing human genetic maps requiring family studies.

Interagency Coordination of Genome Research

White House Office of Science
and Technology Policy (OSTP)

Federal Coordinating Council

on Science, Engineering and

Technology (FCCSET)

National Science
Foundation

National Institutes
of Health

Standing Committee
on the Life Sciences

Subcommittee

on the Human

Genome

Department
of Energy

Department
of Agriculture

Howard Hughes
fvledical Institute

17

Interagency Coordination

Management of the
Human Genome
Program

Joint DOE/NIH Activities

A 1988 Memorandum of Understanding specifies procedures for coordinating DOE and
NIH efforts and establishes a joint advisory committee and an interagency working
group. NIH observers attend the quarterly meetings of the DOE Human Genome
Coordinating Committee, and DOE observers attend meetings of the Program Advisory
Committee (PAC) to the NIH National Center for Human Genome Research. In August
and October, DOE and NIH representatives met infomially as a joint planning group to
begin fomiulation of a coordinated multiyear research plan, which was presented in
December to the HERAC and PAC subcommittees, who then completed the plan. This
national plan was presented to the U.S. Congress in early 1990.

Several important workshops have been cosponsored by DOE and NIH; they include:

• Workshop on Repositories, Data Management, and Quality Assurance for the
National Gene Library and Genome Ordering Projects (August 1987).

• Workshop on Data Management for Physical Mapping (cosponsored with the
Howard Hughes Medical Institute) (May 1988),

i

ff'w^^>^.

Informatics

Human Genome Management Information System
(HGMIS) staff. HGMIS staff are located at the Oak
Ridge National Laboratory in the Biomedical and
Environmental Information Analysis Section of the
Health and Safety Research Division. Members of the
Graphics Division and the Publications Division assist
with manuscript design and preparation. HGMIS
welcomes contributions and suggestions from
genome researchers. (Photographs provided by
HGMIS, Oak Ridge National Laboratory.)

18

• Workshop on Nomenclature for Physical Mapping of Complex Genomes
(cosponsored with Howard Hughes Medical Institute) (April 1989).

• Large Insert Cloning Workshop (December 1989), and

• Workshops on Chromosomes 16 and X (June and December 1989).

The Joint Informatics Task Force, comprised of experts appointed by the NIH PAC
and the DOE HGCC. has constructed a document that makes recommendations for the
present and future computing needs of researchers involved in the Human Genome
Project. Another joint DOE/NIH working group has been formed to address ethical,
social, and legal issues associated with the Human Genome Project. A third joint
working group, on chromosome mapping, is being fonned.

International Human Genome Organisation

The international Human Genome Organisation (HUGO) has been formed to assist with
coordination of national efforts, facilitate exchange of research resources, encourage
public debate, and provide information and advice on the implications of human
genome research. Conceived in April 1988. HUGO is incorporated in Switzerland and
in the United States. New members from among participants in genome research are
elected in a manner similar to that of the European Molecular Biology Organisation and
in some ways parallel to the U.S. National Academy of Sciences. Its 42 founding
members represented 17 countries and included 3 members of the DOE HGCC. Within
its first year. 219 members were elected, including 12 participants in DOE-funded
genome projects. The election of 20 new members in December 1989 increased the
membership to 239. HUGO'S officers for 1990 are Sir Walter Bodmer (United
Kingdom). President; Charles R. Cantor (United States), Vice President, North
America; Kenichi Matsubara (Japan), Vice President, Asia; and Andrei D. Mirzabekov
(Soviet Union), Vice President, Eastern Europe.

19

Resource Allocation

Management of the
Human Genome
Program

87

88 89 90

FISCAL YEAR

The reports of HERAC and the National Research Council on the Human
Genome both recommended that national funding for the Human Genome
Initiative increase to reach a sustaining yearly level of $200 million. The
expenditures within the DOE program have been $5.5 million in FY 1987, $10.7 million
m FY 1988. $17.5 million in FY 1989, and $25.9 million in FY 1990. The presidential
budget for the DOE Human Genome Program in FY 1991 is $46.0 million, as shown in
the figure. Major administrative categories are;

Resource and Technology Development

• The National Laboratory Gene Library Project

• Instrumentation and biological support for physical mapping

• Physical mapping through clone ordering and macro-restriction analyses

• Sequencing technology, including automation and robotics

• Data management, analysis, and networking,
including GenBank -'

Training

• Predoctoral and postdoctoral fellowships at national
laboratories

Supporting Activities

• Human Genome Coordinating Committee

• Task groups

• Human Genome Management Information System

• Workshops

• Ethical and societal issues

• Support for national and international meetings

• Improvements to national laboratory resources and
facilities

• Technology transfer activities

• Publications

Expenditures and FY 1991 presidential budget for the
DOE Human Genome Program

20

Continuing Implementation

The 1987 HER AC report on the Human Genome Initiative provided the broad
guidehnes tor OHER's Human Genome Program. Refined management and
program plans were prepared in 1988. With the experience and progress now
achieved and with participation of the Human Genome Coordinating Committee,
program guidelines for DOE Human Genome Program implementation have been
updated:

Short-Term Focus (1-5 Years):

• Improve by an order of magnitude the efficiency and cost-effectiveness of mapping
and sequencing technologies.

• Rapidly develop a database system for current single chromosome projects.

• Complete orderings of monochromosomal clone libraries already initiated.

• Initiate physical mapping of additional chromosomes.

• Improve and implement methods for infonnation and materials dissemination.

• Develop a long-term human genome database system.

• Continue small-scale DNA sequencing as an adjunct to physical mapping, and as a
test bed for improved sequencing concepts and technologies.

• Encourage increased private involvement in all areas of genomics,

Mid-Term Focus (5-10 Years):

• Accelerate DNA sequencing as more efficient systems are validated.

• Continue development of algorithms for interpreting sequence information.

• Utilize accumulating genome knowledge to improve assessments of individual
susceptibility to genetic damage, from both unavoidable environmental agents and,
especially, energy by-products.

• Utilize accumulating genome knowledge to identify the more biologically
significant damage sites in chromatin.

• Elucidate the structure, function, and interaction of the body's macromolecules by
complementing DNA sequence information with the national laboratories" unique
technologies for structural biology studies.

Long-Term Focus (10-15 Years):

• Complete large-scale DNA sequencing and apply interpretative algorithms.

• Emphasize applications of genome knowledge to prospective and retrospective
analysis of individual exposures to low levels of energy-related agents.

21

Physical Mapping

Localization of unique cosmid clones delineates physical landmarks on
chromosomes, (a) A powerful approach for constructing pfiysical maps is to
use fragments of fiuman DNA cloned in cosmid vectors in in situ hybridization
experiments with the chromosomes being mapped. These fragments can be
localized to specific regions (accurate to within 1 '-i of the chromosome length)
on human metaphase chromosomes by using computer-controlled confocal
laser microscopy to detect fluorescence hybridization between the fragments
and complementary regions on chromosomal DNA. The researcher in the
background is operating the microscope to produce an image on the distant
monitor. The researcher in the foreground is performing computer analysis on
the digitized hybridization data, (b) Shown in this photo is the in situ hybridization
of an anonymous fluorescently labeled cosmid to the long arm of human
chromosome 1 1 . The chromosomes are stained with propidium iodide (red),
and cosmid hybridization is indicated by yellow fluorescence from fluorescein.
The red dye on the chromosome is uniform, except at the location of in situ
hybridization as indicated by line /on the graphic. [Photograph b first published
by the American Association for the Advancement of Science in P. Lichter et al.,
"High-Resolution Mapping of Human Chromosome 11 by in Situ Hybridization
with Cosmid Clones," Science 247, 64-69 (Jan. 5, 1990). Photographs a and b
provided by Glen Evans, The Salk Institute, and Peter Lichter. Yale University
Medical School.!

22

23

1989-90 Research Highlights

Tlie first research and development projects supported by the Human Genome
Initiative were pilot projects in the national laboratories and in academia.
Subsequent projects have been initiated after evaluation by special peer review
panels in 1988 and 1989. Abstracts of all current projects are included in this report and
are supplemented by research narratives from national laboratories and by special
reports in the Appendices. There have been numerous incremental contributions to the
resource and technology development, in addition to significant progress toward the
major goals. Some highlights of the total DOE program include:

The construction of libraries made up of DNA clones with large-capacity phage/
cosmids containing human DNA is progressing within the National Laboratory Gene
Library Project. These libraries will represent the 24 distinct chromosomes (one
chromosome representing each of the 22 autosome pairs plus the X and Y
chromosomes) and. even now, are an extremely valuable resource for physical mapping
projects. Libraries representing chromosomes 4, 5. 8. II, 17, 21 , and 22 are being
offered for evaluation and cooperative use in 1990.

>^W ^1

Tppn

Wt ^H

m W ^m m

sJm

■ii- muuMi^^^J

''':-';:.V^^K ^^^^^^^^^HHR^^^^^^^^I

Limr^^^H

^^^^^B

l_«^" ^H

HHj

Physical Mapping

Cross-protection against Not\ sites in E. coli.

Physical map construction is more efficient when
large DNA fragments are utilized. The restriction
endonuclease Not I cleaves the sequence 5'-
GCGGCCGC-3'; however, If DNA is first
methylated at "^CGCG with M»FnuD II or M»eep I,
a subset of the Not I sites cannot be cleaved; the
specificity of Not I Is effectively doubled. Shown In
the figure is a pulsed-field gel of E. coli RRI
genomic DNA treated as follows. Lanes 2 and 5:
Methylated at '^CGCG by M»FnuD II, then cut with
Not I. Lane 3: Unmethylated. cut with Not I.
Lanes 4 and 6: Methylated at "^CGCG by M»eep I,
then cut with Not I. Lane 1: Bacteriophage lambda
concatemer ladder, molecular weight marker,
48,502 bp per step. (Photograph provided by
Michael McClelland, California Institute of
Biological Research.)

24

Physical Mapping

Physical map of contigs on chromosome 11. Researchers at The Salk
Institute and Yale University Medical School have generated a series of
overlapping sets of cosmids. or contigs, that vary from 2 to 27 clones. The
position and relative order of many of the contigs have been determined by using
fluorescence in situ hybridization on metaphase chromosome spreads, in
particular, the relative order of contigs containing known cloned genes,
anonymous DNA markers, or Hpa-ll-tiny-fragment (HTF) islands (possibly
indicating the location of as yet undescribed genes) are indicated for comparison
to the ideogram of chromosome 1 1 . The position of hybridization is determined by
the fractional length from the 11 p telomere (FLpter) rather than using cytogenetic
banding. Known genes that have been mapped on the long arm include those
encoding the neural cell adhesion molecule (NCAf\/l), the Thy-1 antigen, the
ApoAl cluster, the CDS cluster, porphyrinogen deaminase (PBG), the ETS1
oncogene, and the signal recognition particle receptor (SRPR), as well as others.
[Figure first published by the American Association for the Advancement of
Science in P. Lichter et al., "High-Resolution IVIapping of Human Chromosome 1 1
by in Situ Hybridization with Cosmid Clones," Science 247, 64-69 (Jan. 5, 1990).
Figure provided by Glen Evans, The Salk Institute, and Peter Lichter, Yale
University Medical School.]

Fractional
length

I J3-1

I J2-2

I 13~4A, 18-10B

I PTH
I J4-17

I HRAS

I J7.1.HBBC

I J5-3

I J10-3. J2.2
I J10-17
I 4.4B
FSHB

I 5-2A
I J1.2

I piaL

I PVGM

I ZC7 I XB
I XBll

I 9 27 ,

I I 66

I 3 16

I 23 20

I NCAM

, ,- , I '^^' I AP0AI.4 13,9 4. ZA7

I ^'-^ I THYl I tD3D, PBGD. XB1. XH5

111 I ZD8 I

XBIO.ZDS. ZD7

I SRPR

I ETSI.23,2

I XB2

I 9«
8 5

25

Research Highlights

Physical Mapping

Automated robotic system used to
prepare DNA cosmid clones, (a) The

investigator is using a robot to prepare
sample solutions that contain cloned DNA
fragments for loading onto gels for
electrophoresis. Electrophoresis is used to
determine the size of cloned DNA fragments
prior to employing the fragments in
hybridization techniques, (b) Close-up view
of the eight-channel pipette arm that
prepares and loads samples onto gels.
{Photographs provided by Glen Evans,
The Salk Institute.)

26

The ordering of DNA clones has been initiated for clironiosomes 5, II, 16, 17, 19,21,
22. and X. Efficacy and speed have been demonstrated in these projects for three
distinct ordering strategies.

The 1988 identification of the basic repeat sequence of the human telomere at

LANL has been followed by the cloning of telomeric regions of several chromosomes.
Thus the "end points" of the physical mapping tasks are becoming well defined and are
providing orientation for mapping activities. The telomeric sequence has been found to
be conserved across vertebrate species.

Novel computer software now makes possible the direct entry of raw experimental
results into a database, subsequent data analysis, and future transmission of results to
other laboratories and data repositories. The.se systems are simplifying the requirements
for recording and processing map and sequence data.

Broader problems in the area of human genome Informatics have been addressed in
a series of workshops cosponsored by concerned federal agencies. To pursue these
issues further, the Joint Infomiatics Task Force (JITF) has now been formed. Recently,
guidelines have been published to eliminate ambiguities in clone names and thus
provide for unique naming or identification.

Very fast computer boards for sequence search and comparison tasks have been
demonstrated and are being commercialized.

Improvements in protocols for construction of yeast artificial chromosomes

(YACs) have culminated with the production of YAC libraries whose human DNA
inserts have an average size of 410,000 bp.

For chromosome mapping through pulsed-field gel electrophoresis, the number of
useful cleavage sites has been increased by protocols for modifying DNAs in vitro.

The reliability of a core DNA-sequencing step of the Sanger strategy has been
substantially increased, through genetic engineering of DNA polymerase (of
bacteriophage T7) and modification of polymerase reaction conditions.

A scheme for rational combination of random and directed-sequencing runs on

cosmids provides for much more economical use of expensive primers.

The processing of DNA fragment autoradiographs into assembled sequence data is
now being accelerated by automatic film readers coupled with computerized analysis.

A novel scheme for sequencing by hybridization (SBH) crucially depends on a
capacity to distinguish short segments of perfectly base-paired DNA from segments
with even a single base-pair mismatch. Both the effective theory and practice for such
discrimination have now been demonstrated.

27

Research Highlights

Instrumentation and methodologies, being developed to use multiple stable isotopes
as DNA labels for mapping and sequencing tasks, will increase the speed of sequencing.

Chemiluminescent techniques for displaying DNA fragments are now achieving
.sensitivities of radioisotopic labels, with promise for reduction in safety hazards and
hazardous waste disposal costs.

The first nondestructive images of naked DNA have been achieved through scanning
tunneling micro.scopy and provide promise for single-molecule DNA sequencing.

A multiplex walking strategy has been applied to an 1 100-member library
representing chromosome segment 1 Iq. Automated restriction mapping is now being
utilized to confirm/reject the presence of overlaps between cosmid pairs with
homologies. Some 300 contigs have thus far been constructed.

An automated fluorescence-based method for clone fingerprinting has been
developed, validated, and coupled to software used for contig assembly, data storage,
and graphical display of map infomiation. These procedures are being successfully
applied toward the development of a cosmid and YAC contig map of human
chromosome 19.

A new method of genome mapping using human-specific repetitive sequences and the
polymerase chain reaction {Alii PCR) has been developed. This method is used for the
isolation of regionally localized DNA fragments and for the rapid and efficient
characterization of cloned DNAs, particularly those in YAC vectors.

28

Resource Development

Computerized robotics used to speed repetitive tasks of mapping and sequencing DNA. Application of robotics in
liuman genome research requires expertise in and interaction among a variety of disciplines, including molecular biology,
engineering, and computing science. Hewlett-Packard, Inc., has provided the Human Genome Center at Lawrence
Berkeley Laboratory (LBL) with a computer-dnven robot for handling and processing biological samples. The robot
consists of an active arm (a) capable of accurate and precise movement and of being programmed to change hands
during procedures; eight pumps for dispensing and sampling very small volumes with comparable accuracy and precision
(not in view); a spectrophotometer (b) for color analysis of the samples; rack towers and incubator hotels (c) that hold
either unlidded plates (d) or racks of pipette tips (e); a hand tree (f) that holds tools for gripping (g) or pipetting with either
a large single mandrill (h) or with 1 6 channels (i); a rake to scrape off used tips; and a blank hand for future
customization. The control pole (j) is capable of five degrees of freedom; rotation, height, grip, reach, and wrist twist
(disabled). The staff of the LBL divisions of Engineering, Computing Science, and ^ylolecular Biology are working with the
engineers of Hewlett-Packard to modify existing hardware, as well as to develop new software. Initial applications
developed in this effort will speed the use of second- and third-generation robots in commercial, medical, and forensic
laboratories. (Photograph provided by the Human Genome Center, Lawrence Berkeley Laboratory.)

29

Physical Mapping

Automated cosmid fingerprinting and contig assembly. Chromosome-19-
speclfic cosmids are digested with restriction enzymes, and thie fragments are
labeled with fluorochromes. The Beckman Instruments, Inc., Biomek «' robotic
system processes sets of 48 cosmids per experiment. Throughput is increased,
because of the capability to load the restriction fragments from three cosmids (each
labeled with a different fluorochrome) plus size standards (a fourth fluorochrome) in
each lane of a denaturing polyacrylamide gel. The labeled restriction fragments are
detected, and the fluorescence is digitized as the fragments migrate past a laser
beam in an Applied Biosystems 370 automated DNA sequencer. Fragment data
acquisition may be monitored during electrophoresis to ensure that operation is
proceeding normally.

In the large photograph, the fragment peaks from each of three cosmids are
labeled in blue, green, and yellow. The size standard is in red. Several lanes of the
gel are depicted on the top of this figure, and a historical view of the data from one
lane is shown in the lower plot. The restriction fragment mobility data for each
cosmid are analyzed by a suite of software programs developed at Lawrence
Livermore National Laboratory. Fluorescence signals are processed to remove
noise, to identify peaks (representing restriction fragments), and to calculate
fragment lengths by comparison to the size standards.

The inset photograph demonstrates fragment size comparisons for all pairs of
cosmids to determine whether they share a significant number of fragments of the
same size (to within 1-1 .5 bases). A single statistic is calculated that estimates the
strength of the overlap. A best-overlap solution is determined and presented
graphically for inspection, manipulation, and detailed query of the underlying
database. The cosmids, represented by the warmer-color (red and yellow) bars,
are those that exhibit the most overlap. (Photographs provided by Anthony
Carrano, Lawrence Livermore National Laboratory.)

30

Research Facility Narratives

Lawrence Berkeley Laboratory

Introduction

In September 1987, Lawrence Berkeley Laboratory (LBL) and Los Alamos
National Laboratory (LANL) were designated as Human Genome Centers. LBL's
response was to initiate an effort that would incorporate at one site all of the
elements necessary for successful execution of the project and provide an environment
in which integration of emerging concepts, methods, and techniques was immediate.
Interdisciplinary efforts are a hallmark of LBL and of the other national laboratories.
The unique aspect of the LBL Human Genome Center — the breadth of its activities — is
made possible by the juxtaposition of the great variety of LBL talent in several areas
(i.e., instrumentation, materials science, and computing technologies) with the large,
outstanding biological research communities m the Berkeley and neighboring Bay Area
institutions.

The Center's current activities are concentrated in four areas:

• construction of a physical map of the human genome,

• automation of existing physical mapping methods and development of new ones,

• enhancement of existing technologies for handling and sequencing DNA, and

• improvement of methods for interpreting and analyzing maps and sequence data.

Current efforts focus on chromosome 21 — with 50 Mbp, the smallest human
chromosome. Three principles guided the development of this research agenda. The
first was the realization that new methods, techniques, and instrumentation must be
developed to complete the genome project, and that the most effective way to do this
would be to work in close physical and intellectual contact with pilot-scale mapping
and sequencing efforts. The second principle was that new methods are more easily
implemented on relatively small-scale projects. The third guiding principle was the
projection that the program would be characterized by the use of newer and more
powerful techniques for automated sample handling and biochemical analysis that
would be needed for the increasingly larger data-producing projects.

Thus, development of improved data analysis and management methods was thought to
be necessary both to handle the data generated at the LBL Center and to merge and
reconcile these results with those from the many other laboratories involved in genome
mapping and sequencing. The scientific direction of the Center is reviewed annually by
an eight-member advisory committee, whose members include two Nobel laureates and
six members of the U.S. National Academy of Sciences. The Center is involved directly
with the University of California at Berkeley in a graduate training program in
biotechnology; approximately half of this program's faculty are associated with the
Center at LBL.

The unique features of LBL's current research program include the development of
totally new DNA-handling procedures and physical mapping methods and the use of
yeasts both as a source and as a testing ground of new techniques.

32

Accomplishments

Restriction maps of large regions of chromosome 21 completed. By using (a) single-
copy probes with i<nown regional locations to assign fragments and (b) partial digests,
chromosome-specific Not 1 linking probes, and polymorphism among cell lines to
assign neighboring bands, LBL researchers have identified about 40 Mb of DNA from
chromosome 21 and linked Not I fragments in several regions of the c/ ami; these
fragments include a continuous section that starts at the telomere and extends for
8.5 Mb.

Physical Mapping

A pulsed-fleld gel electrophoresis test bed. The basic DNA separation technique of modern molecular biology is gel
electrophoresis — in particular, pulsed-field gel (PFG) electrophoresis. In the PFG lest bed shown here (left), conditions
at any point in the gel can be monitored by a probe mounted on a computer-controlled platform (right). Parameters such
as current, gel temperature, and pH can now be monitored to optimize conditions and to assure reproducibility among
different PFG sessions. The test bed allows active computer control of electrode potentials, as well as the capability for
programming complex pulse cycles. The results of continuing studies with this and other test beds will be increased
resolution and shorter separation times, especially for DNA fragments of more than 5 Mbp. (Photographs provided by
the Human Genome Center, Lawrence Berkeley Laboratory.)

33

Research Facility
Narratives: LBL

Restriction map of the entire S. pombe genome completed.

Second-generation mapping strategy designed — to be implemented using the
polymerase chain reaction (PCR) and automated-sequencing procedures. LBL's
strategy is to use DNA sequences troni the ends of the fragments as unique "I. D." tags.
Matching these DNA sequences with similarly sized DNA sequences from linking
clones (clones that contain DNA from the ends of adjacent large fragments) will
facilitate map construction. The advantages of this protocol over others is the speed
with which results are generated, the precision of ordering, the simplicity of data
analysis, and the fact that the mapping process generates sequence data as well.

Computer-controlled pulsed-field gel (PF(;) electrophoresis apparatus constructed
and used to discover new pulse shapes that speed separations five- to tenfold. The

test apparatus consists of a 24-node computer-controlled power supply that is capable of
independently programming the pulse sequence of individual electrodes. Computer
calculations are used to define the electrode voltage distribution required to generate
specific electric field profiles within the gel. During the run, a three-dimensional
precision manipulator allows computer-controlled scanning of the gel to monitor
parameters such as voltage, ionic strength. pH. and temperature.

Interactive computer processing of gel images developed so that collection of over-
and underexposed areas eliminates the need for repeating exposure times. This
program provides fully automatic analysis of the separated DNA, if desired; however,
the operator has the option to substitute his/her own judgement at every step. The data
sets of separated DNA are subsequently used for algorithmic comparisons between
lanes and gels, as well as in mapping algorithms, and are stored in a laboratory
database.

New computational algorithms for map assembly developed. Dynamic programming
algorithm.s — traditionally, the technique of choice for matching problems — are likely to
be impracticable for problems of the size generated by human genomic research. For
this purpose. LBL has worked to extend the use of suffix trees: the objective is to
develop computational methods that are faster and more practical than current ones.

New model for map assembly data management systems (based on extended-
entity-relationship model) designed. LBL researchers are using a database schema
design tool (SDT) developed at LBL to provide a powert'ul and easy-to-use interface for
biologists and to increase the productivity of the database design process.

Mapping the telomere of human chromosome 4, which contains the gene for
Huntington's disease, initiated. Probes have been developed to recognize a telomeric

34

area of chromosome 4. Various mapping techniques have been used to begin mapping
around this area.

Method under development for testing single fluorescent molecules in flowing
streams. Based on a new theory for optimizing fluorescence detection, an instrument
has been developed for measuring single molecules of phycoerythrin. This single
molecule detection (SMD) system can measure concentrations three orders of
magnitude lower than conventional fluorescence detection systems. The SMD system is
now being applied to the optimization of fluorescence detection of DNA fragments on
sequencing gels.

Future Directions

• Develop methods for microsurgical dissection of single DNA molecules and PCR of
fragments for direct mapping and sequencing.

• Implement a chromosome-2 1 database pilot project to facilitate cooperation among
different groups and obtain user feedback on desirable or necessary features.

• Complete an ordered library of chromosome 21 by second-generation methods.

• Extend the size range of PFG to allow much higher resolution in the I- to 10-Mbp
range and separations of even larger DNAs.

• Increase the sensitivity of imaging techniques for chromosome in situ hybridization.

• Develop applications for robotics in laboratory procedures such as screening
libraries for linking clones and PCR analysis of sliced gel lanes.

• Develop imaging techniques for direct sequence reading by scanning tunneling
microscopy (STM) or atomic force microscopy ( AFM).

• Incorporate direct-imaging plate technology into automated protocols.

• Develop efficient, versatile algorithms for searching and matching strings in
databases.

• Extend data thesaurus techniques of consistent naming and prototypes to store,
index, search, and retrieve genetic map information.

• Automate genetic stock centers to conduct pilot studies for acquisition, evaluation,
maintenance, and distribution of clones and associated data.

For more information on the LBL Human Genome Center, please contact
Charles R. Cantor, Director, at (415) 486-6800 or FTS 451-6800.

35

Lawrence Livermore National Laboratory

Research Facility
Narratives

Introduction

The human genome project at Lawrence Livermore National Laboratory
(LLNL) is a multidisciplinary effort. It draws upon the Livermore matrix
organization to bring together a team of chemists, molecular biologists,
physicists, mathematicians, computer scientists, and engineers in an interactive research
environment. The broad goals of the project are to:

• develop biological and physical resources useful for genome research.

• model and evaluate DNA mapping and sequencing strategies,

• combine these resources and strategies in an optimal way to construct ordered-clone
maps and DNA sequences of human chromosomes, and

• use the map and sequence information to study genome organization and variation.

Livermore's entry into genomics research was facilitated by existing scientific interest,
expertise, and research in molecular biology, cytogenetics, mutagenesis, and instru-
mentation development, as well as by participation m the National Laboratory Gene
Library Project. In addition, the programs at Livermore have contributed substantially to
the identification and characterization of human DNA repair genes and specifically to
the three DNA repair genes on chromosome 19. It is not unexpected then that LLNL's
initial interest focused on this particular chromosome. Because Livermore"s program is
multidisciplinary. a variety of scientific talent can be drawn upon to meet its needs.
Livemiore's role in the DOE Human Genome Program is one of technology devel-
opment, validation, and application to ongoing and new programs in structural biology
and mutagenesis. The present human genome effort at Livermore involves several
interactive research components that have as a common goal the construction of
ordered-clone maps of the human genome. These component tasks include:

The National Laboratory Gene Library Project. This project is a joint effort with
Los Alamos National Laboratory and has as its goal the construction of human-
chromosome-specific libraries in lambda, cosmid, and yeast artificial chromosome
(YAC) vectors for use in physical mapping and other studies. The project draws upon
experience in flow instrumentation and chromosome sorting at these two national
laboratories.

Resource Development and Management. The group working on this project is
responsible for several tasks, including the construction of specialized cloning vectors
and recombinant libraries for application to physical map construction; the development
of new biochemical and biophysical techniques for mapping and sequencing; and the
management and distribution of Livermore"s material resources such as cell lines,
probes, library arrays, filters, and DNA.

36

Mathematics and Computations. This project's staff is responsible for mathematical
modeling of mapping strategies, the development of data analysis algorithms, data
processing, and the construction and maintenance of interactive relational databases for
internal and external access.

Map Construction. Using material resources, techniques, and computational methods,
the group assigned to this task is assembling ordered-clone maps. This task involves the
application of methods for clone fingerprinting, large-fragment electrophoresis,
multiplex walking strategies, and in situ hybridization to chromosomes.

groo

Minor ^

Sequencing Technologies

Imaging by scanning tunneling microscopy (STM). In a collaborative effort, Lawrence Livermore National
Laboratory and Lawrence Berkeley Laboratory scientists have used STM to image native, unstained DNA molecules
under conditions of normal atmospheric pressure. The image here, as illustrated in the schematic depiction beside
it, is sufficiently resolved to show the major and minor grooves of the DNA double helix. Some researchers have
even suggested that, with further development, STM imaging may someday be an important technology used to
sequence DNA. [Photograph first published by the American Association for the Advancement of Science in Thomas
Beebe, Jr., et al., "Direct Observation of Native DNA Structures with the Scanning Tunneling Microscope, " Science
243, 370-372 (Jan. 20, 1989). Photograph and figure provided by the Human Genome Center, Lawrence Berkeley
Laboratory.]

37

Research Facility
Narratives: LLNL

Instrumentation and Automation. The group responsible for this task develops
instrumentation that will facilitate the handling of DNA and cloned materials so that
human involvement in highly repetitive tasks can be minimized.

High-Resolution Imaging. In this project, two novel approaches to DNA sequencing
are being developed using scanning tunneling microscopy (STM) and high-resolution
imaging with X-ray lasers.

The research groups involved in these component tasks are highly interactive. Individ-
ual staff members often have responsibilities in more than one group. Collaborating
with other research groups throughout the world. Livennore coordinates its research
activities with others who are either involved directly in the human genome effort or
who have mutual scientific interests.

Accomplishments

In the past two years, Livermore has made excellent progress in the construction of
chromosome-specific libraries, in the development and application of new biochemical
and mathematical approaches for constructing ordered-clone maps, in the automation of
fingerprinting chemistries, and in high-resolution imaging of DNA.

As part of the National Laboratory Gene Library Project, small insert libraries of each
human chromosome were constructed using material purified by flow sorting. These
libraries have been deposited with the American Type Culture Collection for worldwide
distribution. Construction of large-insert partial-digest libraries in both lambda and
cosmid vectors has begun: currently, five such libraries are being characterized for
chromosomes 11, 19, 21, 22, and Y.

In the area of resource development, new or modified existing vectors have been
constructed to:

• facilitate cloning small amounts of DNA in cosmids,

• clone Not I linking probes in lambda and plasmids, and

• clone large fragments of DNA as YACs,

Several of the cosmid vectors have been transferred to the U.S. commercial sector. They
were used to construct cosmid libraries of flow-sorted chromosomes. The plasmid and
lambda vectors were used to create a small Not I linking library of chromosome 19. A
library of chromosome 19 in YACs is currently being expanded. A probe repository for
chromosome 19 has been established, and the collection is growing: these probes have
been used to screen Livermore's chromosome-19 libraries. Individual cosmid clones
and filters containing DNA from individual clones of chromosome 19 are being
distributed to the scientific community as part of a collaborative effort to map this
chromosome.

38

Physical Mapping

Chromosome-19 contig map
as of October 1989. Of the

2200 cosmids processed for
fingerprint analysis, over 900
of thiem are in contigs with an
average contig length of about
3 cosmids. Selected cosmid
contigs are shown in this figure
with their map location as
determined by fluorescence in
situ hybridization. Each cosmid
is designated by its clone
number; bold numbers repre-
sent the "minimal" covering
set for that contig. Contigs
associated with known genes
are in ovals, and the gene
designation is indicated above
each oval. The graph at the
bottom of the figure shows the
rate of contig formation.
(Figure provided by Anthony
Carrano, Lawrence Livermore
National Laboratory.)

-T 1 1 1 1 I 1 1 I

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
Number of cosmids analyzed

39

Research Facility
Narratives: LLNL

To construct a set of cosmid contigs (overlapping clones) for chromosome 19, LLNL
has developed a semiautomated fluorescence-based strategy for fingerprinting each
clone. For this procedure, researchers use a robotic system to attach fluorophores to the
ends of restriction fragments from each cosmid clone. Fragment lengths are determined
using a commercially available laser scanning device to acquire data in real time from a
polyacrylamide gel. The data collection is multiplexed in the sense that up to four
different fluorophores (i.e.. four clones) can be run in each gel lane. In this
conflguration, 48 cosmids can be analyzed per gel run.

Livemiore has developed software that will process the acquired fingerprint signals,
convert the information to a fragment length for each cosmid, use the fragment length
data to compute a statistical measure of overlap between cosmids. and allow users to
display data graphically and browse among the cosmid contigs. The flngerprinting
procedure, from fluorochrome labeling through data processing and contig
visualization, is now automated.

About three thousand cosmids have been processed to date. Researchers at LLNL have
established over 330 cosmid contigs for chromosome 19 — estimated to span 30% of the
chromo.some or about 20 Mbp. In addition, they established 6 cosmid contigs that span
approximately 600 kb of chromosome 14. Several of the chromosome- 19 contigs
represent known gene loci; the others are located throughout the chromosome. A
number of these contigs have been validated by restriction fragment digests or by in situ
hybridization to metaphase chromosomes.

In a study parallel to contig map construction, Livermore is developing a large-fragment
restriction map of chromosome 19: this approach will eventually All the gaps between
contigs. LLNL researchers have discovered that two of the DNA repair genes on this
chromosome lie within 260 kb of each other. In addition, they have devised a technique
to isolate region-specific probes based upon polymerase chain reaction (PCR)
amplification of DNA located between human A/w-repetitive sequences. These probes
are being used to identify those cosmids (from Livermore "s chromosome- 19 library)
that span a specific region of the chromosome.

In collaboration. Lawrence Berkeley Laboratory and LLNL have used scanning
tunneling microscopy to produce the highest resolution image of DNA to date. Further
work at Livermore is focusing on the application of this technique to DNA sequencing.
Advances have been made in DNA deposition techniques, construction of a new
computer-controlled scanning tunneling microscope, refinement of the electronics to
minimize noise, and development of new computer programs to improve visualization
and analysis of imaged DNA.

40

Future Directions

• Livemiore's highest priority is to complete, to the extent possible, an ordered-clone
map of chromosome 19. This map will likely be a composite linear array of co.smid,
lambda, and YAC clones. Other goals are to correlate this physical map with the
genetic map and assist the .scientific community in the localization and i.solation of
all the genes from chromosome 19. Livermore researchers will use state-of-the-art
sequencing technology to sequence selected high-interest regions of the
chromosome. Once the technology for map construction for a large portion of
chromosome 19 has been validated, they will scale their efforts to other
chromosomes.

• At some point in the human genome project, emphasis will shift from mapping to
sequencing. Livermore plans to use large fragments such as cosmids or YACs as
templates to explore rapid DNA sequencing methods that can be automated. The
STM and X-ray imaging technologies at Livemiore will be utilized in th-?
sequencing effort, if appropriate.

• Because automation is an essential element of the physical mapping process.
Livermore will continue to explore new processes and instruments to reduce the
need for human involvement in the highly repetitive tasks. For example, LLNL will
complete the development of a prototype cosmid DNA extractor and evaluate its
utility. A number of other instruments for clone manipulation and biochemical
processes will also be considered for automation.

• To assist in the completion of the ordered clone maps, Livermore's interaction with
the scientific community is critical and will continue to be given high priority. An
LLNL facility will serve as a resource laboratory for clones and for map
information on chromosomes of interest. Ultimately, map and sequence infomiation
developed at Livermore will be used to study the global architecture of the
chromosome and to evaluate somatic and genetic variation, spontaneous and
induced, in man.

For more information on human genome research at LLNL, please contact
Anthony V. Carrano at (415) 422-5698 or FTS 532-5698.

Los Alamos National Laboratory

Research Facility
Narratives

Introduction

I

n several of the key technologies necessary for the DOE Human Genome
Program. Los Alamos National Laboratory's (LANL'sj experience and
capabilities include:

• a strong core of molecular biology expertise,

• flow cytometry for sorting chromosomes and for single molecule detection.

• organization of databases (GenBank" ). and

• computer analysis of nucleic acid sequences.

LANL has built upon this experience to develop resources, technologies, and strategies
for mapping and sequencing the human genome and for organizing and analyzing the
resulting data.

Some of the activities of the Human Genome Program require centralization and close
coordination. Among these are ( 1 ) the provision of arrayed cosmid or yeast artificial
chromosome (YAC) libraries from sorted chromosomes that will serve as reference
material for physical mapping and (2) the assembly of the physical mapping and
sequencing information into a computer database. LANL is continuing to play a leading
role in providing these critical resources and in utilizing the findings and materials from
the genome program for basic research.

LANL will collaborate with private industry, both to utilize the skills and resources of
the private sector for strengthening the Los Alamos genome program and to assure the
effective transfer of technologies to the U.S. commercial sector.

Accomplishments

LANL investigators identified and cloned the human telomere sequence, TTAGGG,
repeated several hundred times at the end of each human chromosome. Significant
features of this discovery for the Human Genome Program include the following;

• The sequence provides definitive ends to the map of each chromosome and useful
starting points for physical maps.

• The identification of the human telomere allowed 100,000-200.000 nucleotide
human telomeric DNA sequences to be cloned in YAC vectors (in collaboration
with M. Olson, Washington University). These sequences have been used to
construct contig maps of the ends of several human chromosomes.

Physical mapping of human chromosome 16 is well under way at LANL. where a new
approach has been developed to identify overlapping cosmid clones by exploiting the

42

high density of repetitive sequences in complex genomes. Individual clones are
fingeqjrinted using a combination of restriction enzyme digestions followed hv
hybridization with selected classes of repetitive sequences. Along with information on
the lengths of all restriction fragments, the occurrence of repeat sequences is acquired
by image capture. With this fingerprinting data, overlapping cosmid clones are
identified. Cosmid clones obtained from tlow-sorted chromosome 16 were used to
identify 2261 individual clones arrayed in 389 contig sets — approximately one-half of
chromosome 16. The approach of "nucleating" at specific regions in the human genome
and exploiting the high density of interspersed repetitive sequences in human DNA
allows ( 1 ) rapid progress in early contig mapping phases that have generated large
(>100 kb) contigs and (2) the production of a contig map with landmarks useful for
rapid integration of the genetic and physical maps.

Telomeric
Repeats

Single Multigene
Copy Families
Genes

Centromere
(C-Bands)

G-Bands
(R-Bands,
Q-Bands)

Telomere

^8000-Fold Expansion

I ■ I

44+-

■++

Long Tandem

Repeats

(Satellite

Repetitive

Sequences)

Interspersed Repeats

Long

Interspersed

Repeats

LI

(+ Others)

Short
Interspersed

Repeats

Alu Repeats

Midi- and

Mini-
Satellites

Telomeric
Repeats

Physical Mapping

Organization of human chromosomes. This illustration summarizes the major types of sequences and regions that have
been characterized on a human genome — or mammalian genome, in general. (Figure provided by C. E. Hildebrand, Los
Alamos National Laboratory.)

43

Research Facility
Narratives: LANL

In the National Laboratory Gene Library Project, libraries using the vector Charon 40
and cosmid libraries using the vector sCos 1 have been constructed for human
chromosomes 16, 5, 8, and 4. A lambda library has been made for the X chromosome.

Progress has been made in the detection of single fluorescent molecules in a flowing
liquid — an essential step in LANL's proposed system for sequencing single DNA
molecules at a rate of -10' bp/s. In this approach, the molecule is excited by a laser.
LANL has markedly enhanced the signal-to-noise ratio so that single molecules such as
fluorescein may be detected reliably.

A physical mapping database pilot project has been designed and is being used to
manage data accumulating in the physical mapping project at LANL. A relational
database has been established and is being managed with the Sybase data management
system. Every clone is given a unique identifier and an arbitrary number of charac-
teristics such as source, restriction fragments derived by various digests, restriction
map, probe hybridization, relation to other clones, and relation to genetic markers or
sequences.

A process has been established for identifying industrial partners. A workshop, attended
by 24 companies, was held at Los Alamos, and proposals from 8 of those companies
that responded to a request for proposal (RFP) are now under review.

Future Directions

• Establish, jointly with Lawrence Livermore National Laboratory (LLNL), a
resource to make available arrayed libraries of cosmid clones for all the human
chromosomes. The generation of YAC libraries from flow-sorted material will be
investigated.

• Continue physical mapping of chromosome 16 with cosmid clones. Strategies and
techniques for linking cosmid contigs will be developed, mostly with YAC clones;
mapping of additional chromosomes will be initiated. Clones will be distributed to
provide ties between physical and genetic linkage maps.

• Establish an integrated pilot program for sequencing of megabase regions generated
by physical mapping.

• Develop a system for sequencing single DNA molecules at a rate of -10' bp/s.

• Develop computational tools to support the Library Resource, clone characterization
for physical mapping, and assembly of the physical map from clone overlap
probabilities.

• Develop an integrated database for physical mapping and sequence information
(linked to the genetic mapping database) plus computation, communication, and
analysis tools to make them accessible at scientific workstations in molecular
biology laboratories.

44

• Investigate the structure and function of repetitive sequences in the human genome.

• Study the organization and function of chromatin.

• Develop analysis programs for detecting and characterizing functionally significant
patterns in genomic DNA.

• Collaborate with private companies to ensure the effective transfer of technology to
the commercial sector.

For more information on the LANL Center for Human Genome Studies, please contact
Robert K. Moyzis at (505) 667-3912 or FTS 843-3912.

45

Physical Mapping

Identification and cloning of tfie human telomere to define the ends of
the human genetic and physical maps. Telomeres are defined as the ends
of cfiromosomes. Tfiese specialized structures are involved in the replication
and stability of linear DNA molecules. Investigators at Los Alamos National
Laboratory (LANL) have identified and cloned the human telomere [Moyzis et
al., Proc. Natl. Acad. Sci. USA85. 6622-6626 (1988)]. Fluorescence in situ
hybridization has been used, in addition to other biochemical and biophysical
techniques, to localize this unusual sequence. (TTAGGG)„, to human
telomeres. Seen in the inset photograph as fluorescent yellow spots on red-
stained human chromosomes, this sequence is present at the telomeres of all
vertebrate species and. hence, must have arisen over 400 million years ago
[Meyne et al.. Proc. Natl. Acad. Sci. USA 86. 7049-7053 (1989)].

The ultimate proof that this repeating DNA sequence TTAGGG is the human
telomere is to show that the sequence functions as a unit in an artificial
chromosome. In collaboration with the staff of Fvlaynard Olson's laboratory at
Washington University, the LANL staff was able to construct yeast artificial
chromosomes (YACs) in which the natural human (TTAGGG)„ sequence
functioned as a telomere in yeast cells [Riethman et al., Proc. Natl. Acad. Sci.
86, 6240-6244 (1989)]. These results indicate that the yeast telomere
replication machinery can indeed recognize the human telomere, even though
the common ancestor of yeast and humans lived over one billion years ago.

In addition to demonstrating that the (TTAGGG)^ sequence functions as a
telomere, these YACs allowed large (100,000-200,000) nucleotide fragments
to be isolated from the ends of human chromosomes. Seen in the large
photograph is in situ hybridization of DNA from one of these YAC clones that
originated from the telomere of human chromosome arm 7q. DNA from such
YACs can be used to define the ends of the human genome genetic and
physical maps.

LANLs discovery of the human telomere is a significant milestone in efforts
to map the human genome. Prior to identifying the telomeric sequence,
investigators were without a reference point from which to orient DNA
mapping studies for identification of DNA markers that would be useful for the
analysis of disease genes. [The inset photograph was first published in Proc.
Natl. Acad. Sci. USA 85, 6622-6626 (1988). The large photograph was first
published in Proc. Natl. Acad. Sa. USA 86, 6240-6244 (1989). Photographs
provided by Robert Moyzis, Los Alamos National Laboratory.]

46

47

Abstracts of DOE-Funded Research

The abstracts in this section were contributed by the DOE Human Genome
Program grantees and contractors. The names of the principal investigators are
in hold; the address and phone number following each abstract title are those
of the principal investigator. If more than one principal investigator is listed with an
abstract, the address and phone number belong to the first. An index of project
categories and principal investigators is given at the beginning of this section. Listed
at the end is an index of all project investigators named in the abstracts.

49

Project Categories and Principal Investigators

Abstracts

Principal investigators of the research projects described by the abstracts in this section
are listed here under their respective subject categories.

Resource Development

Raghbir S. Athwal 52

Larry L. Deaven 53

Calvin Giddings 54

Richard P. Haugland 55

William C. Nierman and Donna R. Maglott 56

Charles C. Richardson 57

Carl W. Schmid 58

Marvin A. Van Dilla 59

Shennan M. Weissman 61

Physical Mapping

S. E. Antonarakis 62

David F. Barker 63

Charles R. Cantor 64

Anthony V. Carrano 66

C. Thomas Caskey 68

Glen A. Evans 69

Michael McClelland 71

Robert K. Moyzis 72

Robert K. Moyzis 73

Melvin 1. Simon 74

Cassandra L. Smith 75

Grant R. Sutherland 76

Mapping Instrumentation

Tony J. Beugelsdigk 77

Charles R. Cantor 78

Jack B. Davidson 80

James F. Hainfeld 81

Leonard S. Lerman 82

Betsy M. Sutherland 83

E. S. Yeung 84

50

Sequencing Technologies

Rodney L. Balhorn and Wigberl Siekhaus 85

Douglas E. Berg 86

George M. Church 87

Radomir Crkvenjakov 88

John J. Dunn 89

T. L. Ferrell and R. J. Wannack 90

Raymond F. Gesteland 91

K. Bruce Jacobson 92

Joseph M. Jaklevic 93

James H. Jett 94

Richard A. Mathies 95

Informatics

Christian Burks and David C. Tomey 96

Charles R. Cantor 98

Richard J. Douthart 99

Christopher A. Fields 100

Leroy Hood 101

Betty K. Mansfield and John S. Wassom 102

RossOverbeek 103

Karl Sirotkin 104

Small Business Innovative Research (SBIR) —
Phase I (1989 Awards)

Norman G. Anderson 105

Heinrich F. Arlinghaus 106

Jeffrey M. Stiegman 107

Charles D. Stormon 108

George M. Storti 109

JohnC. Voyta 110

Small Business Innovative Research (SBIR) —
Phase II (1988, 1989 Awards)

Edward M. Davis 1 1 1

Gunter A. Hofmann 112

Ronald A. McKean 113

John West 114

Index to Project Investigators 1 15

51

Resource Development

Abstracts

Monochromosomal Hybrids for the Analysis
of the Human Genome

Raghbir S. Athvval

Department of Microbiology and Molecular Genetics. University of Medicine and
Dentistry of New Jersey. New Jersey Medical School. Newark. NJ 07103-2757
(201)456-4484

In this research project we have proposed to develop rodent/human hybrid cell lines,
each containing a single different human chromosome. The human chromosomes will
be marked with Ecagpl and stably maintained by selection in the hybrid cells.

This experimental approach to producing the proposed cell lines involves the following:
Using a retroviral vector, we will first transfer a cloned selectable marker. Ecogpt (an
E. coli gene for xanthine-guanine phosphoribosyltransferase: XGPRT), to normal
diploid human cells. The transferred gene will integrate at random into multiple sites in
the recipient cell genome. Clonal cell lines from independent transgenotes will each
carry the selectable marker integrated into a different site and perhaps a different
chromosome. The chromosome carrying the selectable marker will then be transferred
further to mouse cells by microcell fusion. In addition, we will use directed integration
of Ecogpt into the chromosome present in rodent cells, not otherwise marked with a
selectable marker. This will allow us to complete the bank of proposed cell lines.

Since the human chromosome will be marked with a selectable marker, it can be
transferred to any other cell line of interest for complementation analysis/ Clones of
each cell line, containing varying sized segments of the same chromosome produced by
selection for the retention or loss of the selectable marker following X-irradiation or by
metaphase chromosome transfer, will facilitate physical mapping and determination of
gene order on a chromosome.

52

Human Recombinant DNA Library

Larry L. Deaven, Robert K. Moyzis, Jon Longmire, and C. E. Hildebrand

Life Sciences Division. Los Alamos National Laboratory, Los Alamos, NM 87345

(505) 667-31 14, FTS 843-31 14

The goal of the National Laboratory Gene Library Project (NLGLP) is the production
of chromosome-specific human gene libraries and their distribution to the scientific
community ( 1 ) for studies of the molecular biology of genes and chromosomes,
(2) for the study and diagnosis of genetic disease, and (3) for the physical mapping
(ordering) of chromosomes. This is a cooperative project employing the flow-sorting
and molecular-cloning expertise at the Los Alamos National Laboratory (LAND and
the Lawrence Livermore National Laboratory. The specific aim of Phase I of the project
was the production of complete digest libraries from each of the human chromosomal
types purified by flow sorting; the average insert size expected was about 4 kb. The
bacteriophage lambda vector was Charon 21 A, which has both EcoR I and Hind III
insertion sites accommodating human DNA fragments 0-9. 1 kb in size. Each laboratory
has produced a complete set of chromosome-specific libraries: LANL with EcoR I and
LLNL with Hind III. The small insert libraries are deposited in a repository at the
American Type Culture Collection, Rockville, Maryland: over 2000 aliquots have been
distributed to over 500 laboratories worldwide.

The second phase of the project — the construction of partial digest libraries with larger
inserts in more advanced, recently developed lambda vectors (9-23 kb) and in cosmid
vectors (33—46 kb) — is under way. These large in.sert libraries have characteristics that
are better suited to basic studies of gene structure and function, organization of genes on
chromosomes, and ordering of cloned sequences. The Phase II strategy is to split the
genome between the two laboratories, with Livermore cloning 12 chromosomal types
(starting with 7, 1 1, 19, 21, 22, and Y) and Los Alamos cloning the other 12 (starting
with 4, 5, 8, 16. 17, and X). In this way, each chromosomal type will be cloned into
both lambda and cosmid vectors. Vectors currently in use include Charon 40 and
lambda GEMII (phage) and sCosl and Lawrist 5 (cosmid). Partial digest libraries
have been constructed in either phage or cosmid vectors for chromosomes X, Y, 16,
19, 21, and 22.

53

Abstracts:

Resource Development

Field-Flow Fractionation of Chromosomes

J. Calvin Giddings

Department of Chemistry, University of Utah, Salt Lai^e City, UT 841 12
(801)581-6683

Fieid-tlow fractionation (FFF). a powerful and versatile methodology of relatively
recent origins, is applicable to the separation of virtually all categories of macro-
molecules and particles. The object of this study is to apply state-of-the-art field-flow
fractionation methods to chromosomes, in an effort to separate and purify them from
one another and from background cellular debris. This research is focused primarily on
the utilization of sedimentation/steric FFF for this problem, but other FFF techniques,
including flow FFF. may be involved as well. Recent experiments involving particles in
the size range of chromosomes demonstrate the feasibility of working in the
chromosome-size range. In all likelihood, FFF methods have sufficient tlexibility to
circumvent any potential problems encountered in chromosome separation, such as
chromosome adsorption or disruption.

54

New Dyes for DNA Sequencing

Hee-Chol Kang. James E. Whiiaker. Peter C. Hewitt, and Richard P. Haugland

Molecular Probes, Inc.. Eugene, OR 97402

(415)486-5717

We have been actively synthesizing and evaluating sets of new tluorophores that can be
excited by the argon laser at 488 or 5 14 nm for possible use in DNA sequencing. The
objectives are to synthesize sets of four dyes whose emission spectra have relatively low
overlap, whose fluorescence when excited with the argon laser is brighter than currently
available tluorophores. and whose properties of ionic charge are uniform for minimum
interference with electrophoretic separations. Principal among the dyes prepared have
been tluoreNcein-rhodamine bifluorophores in which the energy absorbed by the
fluorescein is emitted almost totally at the rhodamine emission wavelength. Examples
of these dyes have been prepared where the energy transfer has been >98% efflcient
with pseudo-Stokes shifts of up to 100 nm. Several reactive versions of rhodamine and
of rhodol dyes have been prepared which fluoresce when excited by the argon laser and
whose emission is brighter than tetramethylrhodamine. The fourth class of new
fluorophores with potential for use in DNA sequencing are reactive, boron dipyrro-
methene difluoride (Bodipy"^' ) derivatives, which have been prepared in several
reactive fonns. Probes derived from this fluorophore have unusually narrow emission
band width and have high absorbance and quantum yield. The prospects for preparation
of new DNA sequencing dyes with higher detectability and spectral resolution will be
presented.

55

Abstracts:

Resource Development

Optimizing Procedures for a Human Genome Repository

William C. Nierman and Donna R. Maglott

American Type Culture Collection. Rockville. MD 20852
(310)231-5559

The cloned genes and DNA fragments identified during the human genome project
should be stored in a repository and made available to the research community. Such a
repository would also establish a set of reference clones to facilitate comparison of data
generated from different laboratories.

Repositories of well-characterized cloned human DNA fragments currently exist, but at
a much smaller scale than necessary for the human genome project. Procedures used in
these repositories cannot be expanded without modification. Methods must be improved
to automate DNA preparation: clone verification; data maintenance and analysis: and
sample storage, recovery, and distribution. Procedures reducing the amount of sample
needed for verification and storage must be perfected. The objective of this project is to
establish a pilot repository to evaluate such protocols and instrumentation. Initial
emphasis will be placed on automating clone verification by analyzing restriction
fragments on a DNA sequencing machine and comparing fragment sizes to those
already obtained by depositors. Methods will also be explored to use robotics for DNA
preparation: to manage information effectively; to verify clones for which there is no
restriction data: and to improve methods of sample storage, retrieval, and distribution.
These procedures will be tested through the development and operation of a pilot
repository using the contigs of lambda clones identified by Maynard Olson's laboratory
for the 5. cerevisiae genome, and chromosome- 1 6- and chromosome- 19-specific contigs
identified by the Los Alamos and Lawrence Livermore national laboratories.

56

DNA Sequence Analysis with Modified Bacteriophage T7
DNA Polymerase

Stanley Tabor, Hans E. Huber. John Rush, and Charles C. Richardson
Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical
School, Boston, MA 021 15
(617)732-1864

The 3' to 3' exonuclease activity of phage T7 DNA polymerase (gene 5 protein) can be
inactivated selectively by reactive oxygen species. The chemically modified enzyme is
highly processive in the presence of E. coli thioredoxin and discriminates against
dideoxynucleoside triphosphates (ddNTPs) only four- to sixfold. Consequently,
dideoxynucleotide-tenninated fragments have highly uniform radioactive intensity
throughout the range of a few to thousands of nucleotides in length. There is virtually
no background due to tenninations at pause sites or secondary-structure impediments in
the template. Chemically modified gene 5 protein, by virtue of having low exonuclease
activity, has enzymatic properties that distinguish it from native gene 5 protein. We
have exploited these properties to show by a chemical screen that modification of a
histidine residue reduces selectively the exonuclease activity. In vitro mutagenesis of
histidine 123, and of the neighboring residues, results in varying reduction of the
exonuclease activity. A deletion of 28 amino acids that encompasses His 123 eliminates
all exonuclease activity (< 10 " %). Incorporation of ddNTPs by T7 DNA polymerase
and E. coll DNA polymerase I is more efficient when Mn-* rather than Mg-* is used for
catalysis. SuKstituting Mn-* for Mg-* reduces the discrimination against ddNTPs
approximately 100-fold for DNA polymerase I and 4-fold for T7 DNA polymerase.
With T7 DNA polymerase and Mn'*, ddNMPs and dNMPs are incorporated at virtually
the same rate. Mn-* also reduces the discrimination against other analogs with
modifications in the furanose moiety, the base, and the phosphate hnkage. The lack of
discrimination against ddNTPs using the genetically modified T7 DNA polymerase and
Mn-* results in uniform terminations of DNA sequencing reactions, with the intensity of
adjacent bands on polyacrylamide gels varying in most instances by less than 10%. A
novel procedure that exploits the high uniformity of bands can be used for automated
DNA sequencing. A single reaction with a single labeled primer is carried out using
four different ratios of ddNTPs to dNTPs; after gel electrophoresis in a single lane, the
sequence at each position is determined by the relative intensity of each band.

For more information see the following articles by S. Tabor and C. C. Richardson:
Proc. Natl. Acad. Sci. USA 84, 4767^771 ( 1987), J. Biol. Chem. 264, 6447-6458
( 1989), and Proc. Natl. Acad. Sci. USA 86, 4076^080 (1989).

57

Abstracts:

Resource Development

Human Repetitive DNA Sequences for Use as Markers in
Mapping the Human Genome

Esther P. Leetlang. Gui-Lin Wang, Joe M. Gatevvood, and Carl W. Schmid
Department of Chemistry. University of CaHfomia, Davis. CA 95616
(916)752-3003

A library of repetitive human DNA sequences was constructed from renatured DNA
and subsequently screened with known repeats. Of the 460 clones examined. 267 did
not hybridize with any of the known repetitive DNAs. Following preliminary sequence
analysis and copy-number determination, ten of the clones were selected for further
study. The repetitive DNA clones were used as hybridization probes to isolate lambda
phage clones from a genomic library. The base sequence, copy number, genomic
arrangement, and evolutionary divergence of the new repeat families are now being
analyzed.

58

Gene Libraries for Each Human Chromosome:
Construction and Distribution

Marvin A. Van Dilla, Pieter de Jong, Barbara Trask. Anthony V. Carrano, Joe Gray,

Kathy Yokobata, and Ger J. van den Engh

Biomedical Sciences Division. Lawrence Livennore National Laboratory, Livermore,

CA 94550

(415) 422-5662, FTS 532-5662

Tiie goal of the National Laboratory Gene Library Project (NLGLP) is the production of
chromosome-specific human gene libraries and their distribution to the scientific
community for the diagnosis and study of genetic disease, determination of the structure
and function of genes, and for the physical mapping of chromosomes. This cooperative
project employs the flow-sorting and molecular-cloning expertise at the Los Alamos
and Lawrence Livermore national laboratories. The specific aim of Phase I of the
project was the production of complete digest libraries from each of the human
chromosomal types purified by flow sorting. The bacteriophage lambda vector used was
Charon 21 A, which has both EcoR 1 and Hind 111 insertion sites accommodating human
DN A fragments 0-9. 1 kb m size. Each laboratory has produced a complete set of
chromosome-specific libraries, LANL with EcoR I and LLNL with Hind 111. Library
purity ranges from nearly 100% for good chromosome preparations and favorably
placed peaks in the flow karyotype to about 50% for some early preparations from cell
lines with unfavorably placed peaks. The libraries are deposited in a repository at the
American Type Culture Collection ( ATCC). Rockville, Maryland; about 2400 aliquots
have been distributed to over 500 laboratories worldwide. All Livennore libraries have
been subcloned into the plasmid vector Bluescribe (Stratagene, La Jolla, California),
facilitating both the use of the DNA probes and the preparation of RNA end probes.

Phase 11, the construction of libraries with large inserts in lambda replacement vectors
(accept about 9-23 kb) and in cosmid vectors (accept about 33-46 kb), is under way.
These large insert libraries have characteristics that are better suited to basic studies of
gene structure and function, organization of genes on chromosomes, and ordering of
cloned sequences. The Phase II strategy is to split the genome cloning responsibility
between the two laboratories [i.e., Livermore will clone 12 chromosomal types (1. 2, 3,
7,9, 11, 12, 18, 19, 21.22, and Y), and Los Alamos will clone the other 12(4,5,6,8,
10, 13, 14, 15, 16. 17, 20, and X)]. In this way, each chromosomal type will be cloned
into both lambda and cosmid vectors. Livermore is using the lambda vector Charon 40
(accepts 10-25 kb inserts) and, more recently, lambda GEMl 1, which has about the
same acceptance range as Charon 40 but is particularly suited for the manipulations
required to efficiently clone, map, sequence, and "walk"" along contiguous segments of
genomic DNA. At Livennore, the cosmid vector is Lawrist 5 (accepts inserts of
34-46 kb). which has the same advantageous features for users as lambda GEMl 1 and
is double the insert size. The cloning procedures (more complicated than for Phase I)
have now been worked out, and we have constructed two large Charon 40 libraries for

59

Abstracts:

Resource Development

chromosome 19; large lambda GEM 1 1 libraries for chromosomes 11. 21. 22. and Y;
and Lawrist 5 libraries tor ail these chromosomal types except 1 1. Phase II libraries are
characterized as fully as resources allow both in-house and by a small number of
interested, high-quality external laboratories before release to ATCC. Currently, this
characterization by test labs is in progress for all but one of these libraries. The one
exception is the chromosome- 1 9 library in Charon 4(); positive characterization results
led to the release of this library to the ATCC in Auuust 1988.

60

New Approaches for Constructing Expression Maps
of Complex Genomes

Sherman M. Weissman, R. Kandpal. A. Swaroop. S. Parimoo, H. Arenstorf.

and D, Ward

Department of Human Genetics, Yale University, New Haven, CT 06510

(203)785-2677

The overall objective of this project is to develop and demonstrate methods for
preparing normalized cDNA libraries and using them for gene mapping and mutation
detection. A phage vector has been prepared that can be used efficiently to generate
single-stranded cDNA clones. A number of human sources have been used to prepare
the cDNA libraries. Biotin avidin selection methods have been developed for con-
venient preparation of subtracted cDNA libraries and are currently being evaluated for
their ability to generate selected cDNA libraries containing only those cDNA
complementary to selected segments of the human genome. In addition, polymerase
chain reaction methodology is being adapted to make chromosome jumping a much
more efficient general procedure for long-range genome mapping and to provide
improved methods for preparing selected cDNA libraries.

61

Physical Mapping

Abstracts

Human Chromosome 21: Linkage Mapping and Cloning
DNA in Yeast Artificial Chromosomes

S. E. Antonarakis, P. A. Hieter. and M. K. McCormick

Center for Medical Genetics. The Johns Hopkins University School of Medicine,

Baltimore, MD 21218-2608

(301)935-7872

The goal of our research is to contribute to the cloning of human chromosome 21 DNA
in yeast artificial chromosomes (YACs). Chromosome 21 is the smallest human
chromosome and contains about 1.4% of the human genome. The cloning of human
DNA in YACs (Burke et al.. Science 236, 806-812, 1987) allows large fragments of
DNA ( 100-1000 kb) to exist as additional chromosomes in S. cerevisiae. We used new
YAC cloning vectors that facilitate the manipulation and mapping of the resulting
YACs. DNA from cell line WA 17 (a mouse-human hybrid with chromosome 21 as the
only human material) and from flow-sorted chromosome 21 were used as the starting
material. Size-selected DNA from complete Ni>t I or partial EcoR I digestion was
ligated to the vectors, and yeast spheroplasts were transformed in the presence of
polyamines to eliminate a bias in favor of smaller DNA inserts. In our initial
experiments, YACs have been obtained from both DNA sources; the average size of
those from the WA-17 cell line was 410 kb.

Specific future research goals include mapping the YAC clones and scaling up the
experiments in order to obtain a large number of YACs, linking the YACs in
overlapping contigs. and constructing a macrorestriction map of chromosome 21.

62

Molecular Mapping of Chromosomes 17 and X

David F. Barker, Hunlington F. Willard. Pamela R. Fain, Arnold R. Oliphant,

and David E. Goldgar

Deparlment of Genetic Epidemiology, University of Utah Research Park,

Salt Lake City. UT 84108

(801)581-5070

The focus of this project is the construction of high-density genetic maps of
chromosomes 17 and X and the correlation of these maps with a set of overlapping
cloned DNA segments. We have isolated over 70 new restriction fragment length
polymorphisms (RFLPs) for chromosome 17 and over 75 for X. The set of available
chromosome- 17 probes is sufficient to construct a genetic map with an average density
of 1 to 2 cM and utilizes the CEPH (Centre d'Etude du Polymorphisme Humain) set of
reference linkage families. The set of X markers will permit the construction of a 2- to
4-cM map. Physical mapping of the chromosomes utilizes both naturally occurring
translocation break points and a series of selectively isolated "push-pull" hybrids that
provide a potentially unlimited series of physical break points from proximal Xp to
distal Xq. Physical localization of probes is also facilitated on the X chromosome by
studies of males with a variety of disease-associated small deletions, and on chromo-
some 17 by the existence of deletions associated with partial loss of 17p in some tumor
tissues and in the Miller-Dieker syndrome. With the combined application of the above
genetic and physical mapping methods, an initial ordering of clusters of DNA probes
along each chromosome will be established. The techniques of pulsed-field gel
fragment mapping and the isolation of overlapping clones in yeast artificial
chromosomes will then be applied to establish an ordered map of all probes and
fragments.

63

Abstracts:
Physical Mapping

Human Genome Center, Lawrence Berkeley Laboratory

Charles R. Cantor, C. Bustamante. M. Esposito, J. Gingrich, S. Levene, M. Maestre,
R. Mortimer. M. Saimeron, C. L. Smith, and M. Stoneking
Human Genome Center, Lawrence Berkeley Laboratory, Berkeley, CA 94720
(415)486-6800, (FTS) 451 -6Sn()

Researchers at the Human Genome Center at Lawrence Berkeley Laboratory (LBL) are
developing methodologies needed to complete a physical map and an ordered library of
the human genome. A top-down approach will be used by developing yeast artificial
chromosome libraries prepared both from total genomic DNA and from specific
physically isolated human chromosomes. The immediate goal is to integrate the
physical map, as it is developed, with the genetic map by defining the sites on the
physical map of cloned and genetically localized genes of specific significance to the
Office of Health and Environmental Research (OHER) mission. Several methods for
constructing the ordered map will be investigated, some of which include using junction
fragments, determining fragment overlap by restriction maps, and employing
recombination among artificial chromosomes.

Brief abstracts of the individual projects are listed below.

Optimization of Yeast Artificial Chromosome (YAC) Mapping (M. Esposito, J.
Gingrich, R. Mortimer, and C. L. Smith) — The use of larger DNA fragments means
that fewer fragments need to be ordered into a map: consequently, the initial major
focus has been to obtain clones containing large fragments of DNA from chromosome
21. The most promising avenue appears to be the use of YAC vectors. Different
strategies for producing YAC clones are currently being evaluated and optimized. Since
the starting material being used for these clones is a hybrid cell line consisting of human
chromosome 21 in a background of mouse chromosomes, a major part of the effort is
devoted to methods for identifying YAC clones that contain human DNA. These clones
are expected to comprise only 1-2% of the total YAC clones from this cell line. The
polymerase chain reaction is being evaluated as a new approach to this problem. This
identification strategy is based on the expectation that only those clones containing
human DNA would be amplified from a known human DNA sequence primer. A
method to link up YACs with overlapping sequence infonnation by using
recombination is under development, as are new methods for improved DNA
transfection into yeast.

New Mapping Methods

Sequence Matching (C. L. Smith and M. Stoneking) — To construct a map, a means of
uniquely identifying each DNA fragment is necessary; the strategy at LBL is to use
DNA sequences from the ends of the fragments. Knowing a small sequence (50-100
bp) from each end of a large DNA fragment will permit each unique fragment to be
identified. Furthermore, matching these DNA sequences with similarly sized DNA

64

sequences from linking clones (clones that contain the DNA from the ends of two
adjacent large fragments) will facilitate map construction. The advantages of this
protocol over existing ones is the speed of generating results, the precision of the
ordering, the simplicity of data analysis, and the fact that the mapping process generates
sequence data as well.

In addition to the traditional means of accomplishing the above tasks, the researchers
are also investigating ways of using amplification via the polymerase chain reaction
(PCR) to obtain DNA sequences from the ends of large fragments and from linking
clones. Ultimately, there are plans to generate linking clones directly via PCR and
thereby avoid some of the pitfalls of traditional cloning methods that make completing a
map difficult. The PCR-based sequencing strategies are also attractive because they can
be readily automated or adapted to existing automated technologies such as DNA
sequencers.

Direct Visualization of Chromosomes and DNA Fragments (S. Levene, M. Maestre,
M. Salmeron, and C. Bustamante) — Two other second-generation mapping protocols
are being investigated. First, hybridization of genes directly on chromosomes that are
visualized with confocal microscopy is used to develop physical maps of intemiediate
resolution. Second, further scanning tunneling microscope (STM) development would
produce STM and DNA handling techniques that would allow the nucleotide sequence
to be read directly from an isolated fragment of DNA.

65

Abstracts:
Physical Mapping

Physical Maps of Human Chromosomes: Methods
Development and Applications

Anthony V. Carrano. Elbert W. Branscomb, Pieter J. de Jong, Emilio Garcia,

Harvey W. Mohrenweiser, and Thomas Slezak

Biomedical Sciences Division, Lawrence Livemiore National Laboratory,

Livemiore, CA 94550

(415) 422-5698. FTS 532-5698

The initial goal of this project is to create physical maps of human chromosomes and to
correlate them with the genetic map. The physical maps will consist of overlapping
cloned DNA fragments (contigs) contained in phage, cosmid, and yeast vectors, all of
which span the chromosomes. The project is multidisciplinary, and its components are
synergistic. In the past two years, progress has been made in several areas. We
constructed new or modified existing vectors to ( 1 ) facilitate cloning small amounts of
DNA in cosmids, (2) clone Not I linking probes in lambda and plasmids, and (3) clone
large fragments of DNA as yeast artificial chromosomes ( YACs). Several of the cosmid
vectors have been transferred to industry. The cosmid vectors were used to construct
chromosome- 1 9-specific libraries from tlow-sorted chromosomes and from a
monochromosomal hybrid. About 10.000 cosmids (about sixfold redundancy) have
been arrayed in microtiler trays to form a reference library for chromosome 19. We used
the new plasmid and lambda vectors to create a No! 1 linking library of chromosome 19
and have initially isolated about 30 clones. We are currently expanding and character-
izing libraries of chromosome 19 in YAC and half-YAC vectors. To construct a set of
cosmid contigs for chromosome 19, we developed an automated fluorescence-based
strategy for fingerprinting each clone. For this procedure, a robotic system is used to
attach fluorophores to the ends of restriction fragments from each cosmid clone.
Fragment lengths are determined by using a commercially available laser scanning
device to acquire electrophoretic mobility data in real time. Up to four different
fluorophores (i.e., four clones) can be run in each gel lane. In the present configuration,
this permits us to analyze up to 48 cosmids per gel run. We developed software to
process the acquired fluorophore signals, convert the signal data to restriction fragment
lengths for each cosmid, and use the fragment length data to compute a statistical
measurement of overlap between cosmids. Several thousand cosmids have been
processed to date. We have established 6 cosmid contigs that span approximately 600
kb of chromosome 14 and have over 200 contigs for chromosome 19. Five of the
chromosome- 19 contigs represent known gene loci, and the others are located
throughout the chromosome. Contigs are validated by restriction fragment digests and/
or by in situ hybridization to metaphase chromosomes. By using large-fragment
analysis from pulsed-field gels to close the region of chromosome 19 containing three
DNA repair genes and the myotonic dystrophy locus, we discovered that two of the
DNA repair genes lie within 260 kb of each other. Finally, we devised a technique,
based upon PCR amplification of DNA, to isolate region-specific probes located
between human Alii repetitive sequences. These probes are being used to identify those

66

cosmids. from our chromosome- 19 library, that span a specific region of the
chromosome. As soon as we have processed about 8000 cosmids from chromosome 19
and have initiated the process of contig closure, we will begin to construct cosmid
contigs from another human chromosome, probably chromosome 3. The physical
mapping effort on this chromosome will be done in collaboration with several other
research groups. As the physical maps near completion, we will develop and exploit
new sequencing methods to study the molecular architecture of the chromosome and
new screening methods that will rapidly evaluate somatic variation and induced genetic
change in human populations.

67

Abstracts:
Physical Mapping

Mapping and Ordered Cloning of the Human X
Chromosome

C. Thomas Caskey. David L. Nelson, and David H. Ledbetter

Department of Molecular Genetics, Baylor College of Medicine, Houston. TX 77030

(713)798-4773

The ultimate goal of this project is the isolation of a complete set of overlapping DNA
clones comprising the human X chromosome. This goal will be achieved through
several means. A high-resolution pulsed-field gel map of regions of the chromosome
will be developed: the regions will begin in Xq28 and extend toward the centromere, in
order to assist in placement of clones as they become available. An extensive panel ot
somatic cell hybrids will be developed to assist in probe isolation and assignment.
Rapid isolation of novel X-region-.specific fragments will be achieved through the
development of a method based on the polymerase chain reaction human-specific AIn
primers for specific amplification of human sequences from somatic cell hybrid and
cloned DNAs will be utilized. Yeast artificial chromosomes retaining large X-region-
specific fragments will be utilized for regional isolation and overlap. Finally, a
computer database management system will be designed to assist in data handling for
these tasks. Initial efforts will focus on the Xq24-qter region with specific emphasis on
Xq28, a region with a large number of genetic disease loci.

68

Physical Map and Overlapping Cosmid Set for Human
Chromosome 11

(Jlen A. Evans, Kathy A. Lewis, Gary Hennanson, Kathryn C. Evans, Jun Zhao,

Kimball O. Pomeroy, Carlisle P. Landel, David McElligott. Mary Saleh, James

Eubanks, Daniel Kaufman, Ken D. Pischel, Shizhong Chen, Joseph Trotter, Reece Hart,

and Grai Andreason

Molecular Genetics Laboratory, The Salk Institute for Biological Studies, La Jolla,

CA 92037

(619)453-4100. ext. 279

The mission of The Salk Institute is to apply concepts and techniques of modem
biology to the solution of human medical problems. Human genome research at The
Salk Institute was initiated in 1988 with the support of the Department of Energy
(DOE), under the direction of Dr. Glen A. Evans. The recently established Center for
Human Genome Research, a closely integrated research group at The Salk Institute, is
working in collaboration with several DOE national laboratories within a highly focused
program to produce a continuous physical map and overlapping cosmid set for human
chromosome 1 1. Inherent in this approach is strong involvement in the development of
novel techniques and strategies for genome analysis.

In the past two years, researchers at The Salk histitute's Center for Human Genome
Research have made considerable progress in the development of new cloning
methodologies and techniques for genomic analysis and in using these approaches to
construct a physical map of human chromosome 1 1 . Extending over about 148 mb,
chromosome 1 1 represents about 4.2% of the human genome. We have constructed,
as a pilot project, an initial small set of cosmid clones in a specialized cosmid vector,
sCos- 1 , that allows the rapid detemiination of overlapping sequences in the collection
through the synthesis of directional RNA probes. Over 1000 cosmids mapping in the
region from 1 lql2 to 1 Iqter have been isolated from a cosmid library constructed from
a somatic cell hybrid containing a portion of human chromosome 1 1 in a mouse MEL
cell background. These cosmids have been organized in a 36-X-36 array on a
nitrocellulose filter: using a novel strategy of overlap determination — referred to as
multiplex mapping — that uses pools of clones, we detected 1099 pairs of linked clones
in the collection. These pairs have been assembled into 315 predicted cosmid contigs,
which are now undergoing analysis by restriction mapping.

Using a laboratory robot, we also devised techniques for automated preparation of
cosmid DNA and for restriction analysis. Each of the clones for four rare restriction
enzymes — Nor I, BssH II, Sfi I, and Sac II — have been mapped, and over 150 Not-l-
containing linking clones and 37 putative ///w/TI-tiny-fragment (HTF) islands have
been identified. This automated system is now being used to complete the restriction
mapping of the entire collection of cosmids.

69

Abstracts:
Physical Mapping

In collaboration with D. Ward and P. Lichter (Yale University Medical School) and
D. Housman and K. Call (Massachusetts Institute of Technology), we have used high-
resolution in situ hybridization of single cosmid clones to map selected contigs and
landmark clones to chromosomal locations. Produced in the pilot study, the resulting
collection of clones, containing over 609f of the 1 lql2 to I Iqter region, is contained in
a reference collection now undergoing analysis. To expand this pilot study to a larger
collection of clones representing a tenfold redundancy of the entire chromosome 1 1 , we
have used a fluorescence-activated cell sorter to purify human chromosome 1 1 from a
somatic cell hybrid, Jl, containing a single chromosome 1 1 in a CHO-Kl cell
background, and we are preparing a chromosome-1 1 -specific cosmid library in sCos-1.

This work represents the beginning of a large-scale mapping project to obtain,
reference, archive, and link cosmids spanning the entirety of human chromosome 11, a
project which may be complemented by studies using pulsed-field gel electrophoresis
and yeast artificial chromosomes. During the next year, the Center for Human Genome
Research will expand its program to include additional Salk Institute investigators
interested in gene recombination and amplification. YAC vectors and mapping
strategies, novel in vivo mapping techniques, and identification of recessive oncogenes.
The goals are to complete the chromosome-1 1 physical map and proceed with
characterization of genes important to human biology. New methodologies to be
utilized in the longer temi include DNA sequencing, functional expression in
mammalian cells in culture, creation of transgenic mouse strains as models for human
disease states, and the identification of functional genes by genetic complementation.

The immediate goals are to ( 1 ) improve engineering, robotic, and computational
systems for preparation and management of arrayed cosmids and their subsequent
processing: (2) expand the arrayed library of chromosome-1 1 cosmids to
10,000-33.000 members: (3) establish a corresponding repository and database for the
distribution of these resources and the correlation of results from other laboratories:
(4) pursue detection and characterization of expressed genes, especially those relevant
to disease, while completing the physical map: (5) continue the conceptual and practical
development of the multiplex strategy including extension of the probe-pooling strategy
from two- to higher-dimensional arrays of the cosmids, establishment of the optimal
library size, further suppression of effects of troublesome high-copy-number sequences,
and exploration of the utility of PCR techniques for probe preparation: (6) develop a
correlative approach to integration of chromosome-1 1 map data acquired through
multiplex walking, PFG and linking clone analyses, linkage data obtained through the
use of a variable number of tandem repeats ( VNTR), RFLP, and minisatellite probes
and radiation hybrids: and (7) continue the characterization of DNA/genes adjacent to
the chromosome-1 1 translocation breakpoints obtained from clinical sources and pursue
the identification of disease genes mapped to chromosome 1 1.

70

Novel Methods for Physical Mapping of the Human
Genome Applied to the Long Arm of Chromosome 5

Michael McClelland. Carol A. Westbrook.* Mike Weil. John Hanish. Mike Nelson.

Yogesh Patel. Settara C. Chandrasekharappa.* Michelle M. Le Beau.* and

Michelle Rebelsky*

California Institute of Biological Research. La Jolla. CA 92037

(619)535-5476

*Department of Medicine. University of Chicago. Chicago, IL 60637

The goal of this project is to develop and assess new approaches to megabase mapping
of a suhchromosomal region, specifically applied to chromosome 5. bands q23-31. This
region has been selected because, at 25 Mb. it is of manageable size and represents an
approach to the larger chromosomes. Our megabase map will consist of restriction sites
for enzymes that cleave infrequently and will link a series of probes that have been
mapped to 5q. The region is delineated, and the probes sublocalized. by means of
hybrids containing translocations and deletions in chromosome 5. some of which have
been prepared from leukemic cells of patients that carry chromosome-5 abnormalities.
The technology we plan to develop includes ( I ) enzymologic strategies and (2) methods
for the directed production of unique-sequence probes from the region of interest and
will include linking clones. The multimegabase strategies have been quite successful:
we have developed a reliable method for producing a partial digest of DNA in agarose.
In addition, several methylase/D/7/( 1 combinations are being evaluated, including one
that cleaves a 12-bp specificity. These approaches should generate fragments of over
500,000 bp in the human genome and facilitate the linking of probes. The map will have
interesting biological uses because the region contains the gene(s) for radiation/
mutagen-induced leukemia, as well as for a variety of growth factors and receptors.

Abstracts:
Physical Mapping

Center for Human Genome Studies, Los Alamos
National Laboratory

Robert K. Moyzis, C. E. Hildebrand. R. L. Stallings, and N. A. Doggett

Lite Sciences Division. Los Alamos National Laboratory. Los Alamos. NM 87545

(505) 667-3912. FTS 843-3912

The Los Alamos Center tor Human Genome Studies will provide coordination,
technical oversight, and direction for the following interdisciplinary elements of the
Human Genome Program at Los Alamos: physical mapping, new technology
development, and infomiatics. The center will also develop collaborative research and
development programs with the private sector and with other centers for human genome
research.

The goals of this project are ( 1 ) to develop concepts and to advance technology for
genomic physical mapping and (2) to construct a physical map of human chromosome
16 that will include an ordered set of overlapping DNA fragments encompassing the
chromosome. The physical map will integrate phage, cosmid, and yeast artificial
chromosome (YAC) contigs ordered by repetitive sequence "fingerprinting" with the
genetic linkage map. identified gene sequences, and the cytogenetic map into a tool that
will allow rapid access to any region of chromosome 16 for analysis and eventual large-
scale sequencing. The significance of this work lies in the immediate application of the
knowledge and tools (1 ) to understand human genetic disease. (2) to clarify the
molecular bases for genetic disease susceptibility, especially in regard to energy-related
chemical or radiation exposures, and (3) to reveal the molecular details underlying long-
range chromosome architecture and dynamics.

72

Genome Organization and Function

Robert K. Moyzis. Julie Meyne, and Robert Ratliff

Life Sciences Division. Los Alamos National Laboratory, Los Alamos, NM 87543

(5031 667-3912, FTS 843-3912

The ultimate objective of this program is to determine the molecular mechanisms by
which higher organisms organize and express their genetic information. Applications of
these basic investigations will include the development of novel approaches for (a)
detecting of human genetic diseases and (b) measuring the effects of low-level ionizing
radiation and/or carcinogen exposure. A combination of biochemical, biophysical, and
recombinant DNA techniques is being used to identify, isolate, and determine the roles
of DNA sequences involved in long-range genomic order. Currently, major efforts are
focused on determining the organization and function of human repetitive DNA
sequences. Major findings in the last year included (a) the use of synthetic repetitive
DNA oligomers to target in situ hybridization to specific human chromosomes and (b)
the isolation of the human telomere. Future studies will be directed toward (a) the
further definition and isolation of "functional"" repetitive DNA regions and (b) the
cloning, in yeast artificial chromosome vectors, of human telomere adjacent DNA
fragments. Defining the mechanisms responsible for organizing the mammalian
genome, as well as determining the genetic and nonmutational alterations
accompanying abnormal phenotypic change, is important to identifying the effects of
environmental contaminants from energy-related technologies. Determining the genetic
variability in these mechanisms provides a rational basis for establishing thresholds for
toxic substance exposures, for making valid cross-species extrapolations, and,
ultimately, for identifying individuals at risk.

73

Abstracts:
Physical Mapping

Developing a Physical Map of Human Chromosome 22
Using PACE Electrophoresis and Large Fragment Cloning

Melvin I. Simon. Bruce Birren. and Hiroaki Shizuya

Biology DiviMon, California Institute ot Technology. Pasadena. CA 91 106-4107

(818)356-3944

The goal of this project is to derive a set of overlapping clones covering human
chromosome 22. Much of the work involves the development of new or improved
methods for cloning large DNA fragments, and for handling, analyzing, and
overlapping these clones. To create an overlapping clone map of human chromosome
22. a set of new bacterial artificial chromosome (BAC) vectors will be developed that
should support the cloning of large DNA fragments about 200 kb in length. These BAC
vectors are based on the E. coli F-factor and will be constructed to contain promoters
for walking, a multiple cloning site, a cos site for cleavage reactions, rare-cutting sites
surrounding the insert, and two selectable markers. A library of chromosome 22 will be
constructed in these vectors, in modified YAC vectors, and in cosmids. Source DNA for
the libraries will come from hamster/human hybrid cells that contain either intact or
deleted chromosome 22. For the hybrids with deletions, the pulse alternating current
electrophoresis (PACE) system will be used to separate the deleted chromosome.
Human clones will be selected from the libraries by screening with total human DNA.
Fingerprints of the clones from the different vector systems will be achieved by partially
digesting the cloned DNA and labeling the cos sites with either radioactive or
fluorescent tags. The cos sites will be cut with terminase and labeled by a hybridization-
ligation reaction. Since each cos end has a different sequence, different oligos can be
ligated to each site and a partial-digest map can be created from each end of the clone.
The use of fluorescent tags attached by ligation allows the simultaneous use of different
fluorochromes at each cos site while separate restriction analysis would be done for
radiolabeled oligonucleotides. Detection of the restriction fragments would be
performed on the PACE pulsed-field gel electrophoresis system. Based upon the partial
digest data, computer algorithms will construct the overlap map.

74

Techniques for Determining the Physical Structure
of Entire Human Chromosomes

Cassandra L. Smith. W. Michels, J. S. Cheng. H. Fang. J. Gingrich. D. Wang,
and Y. Wu

Human Genome Center. Lawrence Berkeley Laboratory. Berkeley. CA 94720
(415) 486-6800. FTS 4.'^ 1-6800

Large-fragment DNA methods are being used to construct a macro-restriction map of
the smallest human chromosome. Isolation of a human telomere yeast artificial
chromosome clone enabled the ends of the map to be defined. About 30 single-copy
DNA probes with previously assigned genetic map locations along the length of the
chromosome are being employed as anchor points. These probes were used to identify
corresponding large Not 1 and Mlii I DNA fragments by hybridization to pulsed-field gel
fractionated DNA restriction digests. The map between the anchor points is being
reconstructed by combining several approaches: assigning other bands by using single-
copy probes with known regional locations; assigning neighboring bands using the 15
thus-far-isolated chromosome-specific Not I linking probes: and interpolating between
anchor points using partial digests phased either by Smith-Bimstiel type approaches or
by using sites that are polymorphic, when different cell lines are compared, as signa-
tures of particular regions. A series of clones of repeated DN As are being used to
identify all the chromosome-2I restriction fragments present in these hybrid rodent cell
lines. These approaches have allowed us to identify about 40 Mb that come from
chromosome 21 and to link up Not I fragments of at least 8 Mb near the (/ telomere and
additional significant regions along the cj arm. Additionally, these strategies have
allowed us to determine that D21S13, the locus most closely linked to the Alzheimer's
disease gene on chromosome 21, is in fact located on the same 1.6-Mb fragment as
locus D21S16. Although the map is not yet complete, it reveals interesting features such
as the uneven distribution of putative genes along the chromosome and a greater-than-
expected gradient of enhanced recombination near the (/ telomere.

75

Abstracts:
Physical Mapping

Correlation of Physical and Genetic Maps of Human
Chromosome 16

Grant R. Sutherland. David F. Callen. Valentine J. Hyland, John C. Mulley, and

Robert I. Richards

Department of Histopathoiogy. Adelaide Children's Hospital, North Adelaide,

South Australia 5006, Australia

011-618-267-7333

The goal of this project is to construct a detailed physical map, which will be correlated
with the linkage map, of human chromosome 16. The methods to be used for con-
struction include the following:

(DA panel of mouse/human hybrid cell lines, which contain only parts of chromosome
16, will be developed. The panel will be achieved by fusing human cells that contain
rearrangements of chromosome 16 with mouse A9 cells and selecting for the human
APRT gene on the end of the long ami of chromosome 16. The cytogenetic and
molecular characterization of the panel will allow detailed physical mapping of cloned
DNA sequences and genes that are expressed in the hybrid cells. The cell panel
produced should divide this chromosome, which contains approximately 3.3% of the
human genome, into about 50 intervals of average-size 2 Mb and thus provide a means
of mapping any cloned DNA sequence from chromosome 16 into these relatively small
regions. Sequences that map into such a region should then be useful to generate
restriction maps using pulsed-field gel electrophoresis. This project should lay the
foundation for construction of a restriction map of chromosome 16.

(2) Anonymous cloned DNA fragments of chromosome 16 will be selected from
various chromosome- 16-specific libraries and mapped using the hybrid cell panel. In
selected intervals these fragments will be used to identify restriction fragment length
polymorphisms (RFLPs) for which the CEPH (Centre d'Etude du Polymorphisme
Humain) panel of families will be typed to correlate the physical and linkage maps.
Probes to cloned genes that have been mapped to chromosome 16 will be obtained, and
these genes will be physically mapped: where these probes detect RFLPs, their linkage
relationships with other cloned segments will be determined using the CEPH families.
Probes on chromosome 16 that have been put through the CEPH families and used to
generate linkage maps will be obtained from other researchers on a collaborative basis
and physically mapped to further define the correlation of the physical and genetic maps
of this chromosome.

76

Mapping Instrumentation

Automated Methods for Large-Scale Physical Mapping

Tony J. Beugelsdigk and Robert Hollen

Mechanical and Electronic Engineering Division, Los Alamos National Laboratory.

Los Alamos. NM 87545

(505) 667-3169, FTS 843-3169

The preparation of an ordered-clone collection from human chromosome-specific DNA
libraries, necessary for both low- and high-resolution physical maps, offers the next
challenge in the task of mapping the entire human genome. This research will focus on
instrumenting the front-end processes required in constructing a low -resolution physical
map. specifically, the ability to propagate automatically and purify human DNA
fragments for subsequent analysis and use. We are assembling the necessary automated
hardware designed to manipulate, simultaneously, through numerous preparative steps,
a large number of samples containing small volumes (0. 1-0.5 mL), and, finally, to
deliver the samples to solid support filters for binding. The samples will automatically
be placed on the solid support in a precisely indexed array for multistage analysis and
automated data acquisition.

Based on extensive experience in designing robotic and automated equipment, we
anticipate that practical problems in large-scale mapping programs will become evident
only after attempts are made to apply new methods to actual map production. For this
reason, instrumenting the construction of chromosome-specific physical maps will
evolve through a multidisciplinary program. This strategy offers an advantage in that
automated devices will become both research and production tools and will be
appropriate targets for technology transfer.

77

Abstracts:

Mapping

Instrumentation

Human Genome Center, Lawrence Berkeley Laboratory

Charles R. Cantor, C. Bustamante, J. Gingrich, A. Hassenfeld, J. Jaklevic,
W. Johnston, J. Katz, W. F. Kolbe, S. Levene, S. Lewis, M. Maestre, and E. Theil
Human Genome Center, Lawrence Berkeley Laboratory. Berkeley, CA 94720
(413)486-6800, (FTS) 451-6800

Researchers at the Human Genome Center at Lawrence Berkeley Laboratory (LBL) are
developing innovative techniques in instrumentation and automation to accommodate
the size and complexity of the experimental procedures used in physical mapping
methods. In addition to improving existing laboratory methods, emphasis will be placed
on developing advanced techniques for separating large DNA fragments. Technology
for the flow separation of chromosomes will be developed. Modem nuclear radiation
detectors or optical and ultraviolet imaging systems will be used to explore methods for
direct imaging of electrophoresis gels. The use of differential-polarization imaging to
achieve enhanced sensitivity for direct viewing of DNA will be investigated.

Brief abstracts of the individual projects are listed below.

Optimization of Pulsed-Field Gel Electrophoresis (A. Hassenfeld, J. Jaklevic. J.
Katz, W. F. Kolbe, and S. Levene) — Engineers at LBL have constructed a test bed for
all of the various configurations of PEG electrophoresis. This test bed provides
precision control and recording of the conditions within the gel in real time during the
run. Optimizing the speed and precision of DNA isolation will, in turn, shorten crucial
steps in the mapping process while retaining the high resolution of the current PEG
techniques. Among the variables currently being explored are short, intense secondary
electric field pulses and combined electric and magnetic fields.

Micromanipulation of DNA (M. Maestre and C. Bustamante) — Some of the
infomiation from the PEG studies is being applied to the development of a system for
handling and manipulating single DNA molecules. The goal is to develop techniques for
isolating specific, single DNA molecules for other procedures such as PCR. cutting by
enzymatic or physical means, or direct visualization. The rationale for this technique is
to construct a network of electrodes that are about the same size as large DNA
molecules (electrodes of 10-20 mm separated by 20-50 mm). A key concept is the use
of inhomogeneous fields. If a DNA molecule is to be manipulated in a way that places
different sections of the molecule in specific positions in the electrode net, it is essential
that these parts experience different forces and different electrostatic fields. The motion
of single DNA molecules has been made visible with the use of the intensified
epifluorescence microscope after labeling with fluorophores (acridine orange). Direct
manipulation of the molecule is then performed through the computer-assisted control
of the local electric fields by the microelectrodes. The microelectrodes have been
constructed and tested, and DNA molecules can be moved in predictable directions. T4
phage DNA was used in the preliminary testing. The DNA molecules stretched to a
length of 49 mm; this length is very close to the length of 52 mm reported for the T4
phase DNA.

78

Automated Image Analysis (W. Johnston, S. Lewis, J. Jaklevic, and E. Theil) —
Most mapping protocols rely upon visualization of DNA. Therefore, development of
hardware and software for automatic capture, analysis, indexing, and storage of images
is needed to advance the generation of physical maps whether by PFG analysis,
confocal microscopy, or STM. The gel imaging system being developed here
emphasizes automatic electronic filtering of peaks for band identification. The filter
procedure is designed to remove constant or linear background while compensating for
the weak signal that may be present from shoulders on better defined peaks. Other
aspects of the imaging system include development of a simple-to-use image database
and automatic lane compensation. The overall goal is a highly integrated acquisition,
analysis, storage, and retrieval system for any image.

Robotics (J. Gingrich, S. Lewis, J. Jaklevic, and E. Theil) — Robotic techniques are
being developed to automate and accelerate the labor-intensive steps that currently limit
the rate of generating physical maps. This effort is resulting in modification of existing
hardware as well as the development of new software. Applications currently being
tested for robotic automation include screening yeast colonies for those containing
YACs; processing DNA samples for PFG analyses: and collecting and processing DNA
samples separated by PFG for further electrophoretic analysis, sequencing, or
amplification by PCR.

79

Abstracts:

Mapping

Instrumentation

Genomic Instrumentation

Jack B. Davidson

Instrumentation and Controls Division. Oak Ridge National Laboratory, Oak Ridge.

TN 37831-6010

(615) 574-5599. FTS 624-5599

We are taking a two-level approach to instrumentation needs in DNA mapping and
sequencing. On the first level, developments to improve present gel-based techniques
are extensions of our approach to filmless autoradiography. Using ultralow-light-level
digital television, we detect macromolecules labeled with 'H, "C, "S. or '-P directly
from dried gels impregnated with a liquid scintillator or by use of an intensifying
screen. Originally developed for two-dimensional protein distributions, the method has
improved the speed and accuracy of data acquisition and may be useful for imaging the
blots found in gene mapping. Upon resolution and field-coverage improvements, large
conventional sequencing gels and the blots used in G. M. Church's multiplexing system
could be imaged directly. Because light is the detected entity, the basic system can be
applied to gels tagged with fluorescent dyes as well as radioactive labels. A related
development is a "lensless" radiation microscope for imaging beta particles in in situ
hybridization studies and in neuronography. One goal— to visualize radiolabeled genes
on chromosomes— requires a 20- to 50-fold improvement in resolution over present
capability.

80

High-Resolution DNA Mapping by Scanning Transmission
Electron Microscopy (STEM)

James F. Hainfeld. Martha N. Simon, Stephen G. Will

Biology Department, Brookhaven National Laboratory, Upton, NY 1 1973

(516) 282-3372, FTS 666-3372

Mapping DNA directly with STEM may complement the fast-growing technology of
automatic sequencers. Several tests will be made to determine the feasibility, speed, and
reliability of this method. There are several advantages of a direct physical microscope
approach to sequencing: ( 1 ) very long sequences could be done (i.e., lO'^-lO^ bp in
length); (2) if successful, the method could be several orders of magnitude faster than
chemical methods: and (3) since long pieces of DNA are used, the problems
encountered with repetitive sequences would be circumvented. Preliminary results have
been obtained using the following test system. A 622-bp sequence from pBR322 was
excised with restriction enzymes and purified. Next, a 128-bp T7 piece was inserted at
position 276 (giving a total of 720 bp). The denaturing and renaturing of equal
quantities of the 622-bp and 720-bp fragments resulted in 50'7f fonnation of hetero-
duplexes (one 622 strand paired with a 720 strand) and left the extra bases as a single-
stranded loop. A 26-mer oligonucleotide that was complementary to a region of the
single-stranded insert was synthesized. A chemical modification added a sultTiydryl at
the 3 -end of this oligonucleotide, and the undecagold cluster was covalently attached to
it. The oligonucleotide and heteroduplexes were then mixed under renaturing conditions
and examined using STEM. Control heteroduplexes with no gold clusters show a kink
at the position of the 128-bp single-stranded insert, and the total length and length to the
insert are consistent with the proposed model. When the gold-oligonucleotide was
hybridized, the gold cluster was visible as a tiny bright dot at the "V vertex of the
DNA. The gold cluster was about 10 A from the base it labels (3 bp), and the accuracy
of positioning a base from the end of DNA segments with STEM is 2 bp. A total
potential positional accuracy of 3-5 bp should prove useful in the physical mapping of
genomes.

81

Abstracts:

Mapping

Instrumentation

Thermal Stability Mapping of DN A by Random
Fragmentation and Two-Dimensional Denaturing
Gradient Electrophoresis

L. S. Lerman. Nashua Gabra, Eric Schmitt. and Ezra Abrams

Department of Biology, Massachusetts Institute of Technology. Cambridge. MA 02139

(617)253-6658

The themial stability of the double helix in a standard solvent is fully determined by the
base sequence. Within a long DNA molecule, each local region (ranging in length from
a few dozen bases up to several hundred ba.se pairs) undergoes a transition from an
ordered helix to a disordered, randomized configuration (melting) within a narrow
temperature span, typically from I to 3°C for the change from 95*"/^ helical to 5%
helical. It is convenient to characterize the transition in each region, or domain, by the
Tm, the temperature at which there is a 50-50 equilibrium between the helical and
melted forms. Within human genomic DNA there is substantial variation in the local
Tm, as much as about 35 degrees, often with distinct and sharp boundaries between
adjacent domains. While the pattern and characteristics of this sequence of domains in
long DNA molecules is inferred principally by statistical-mechanical theory, the domain
content is more directly observable in short DNA molecules by means of absorption
spectroscopy as a function of temperature, or by denaturing gradient electrophoresis.
Since the Tm of each domain is changed only very slightly by the substitution, addition,
or deletion of one or a very few bases, the sequence of domains provides a robust
counterpart to the base sequence. The domain map is less sensitive to trivial individual
variation (including methylation) than a restriction map and reflects biological function
more closely.

In two-dimensional separation of randomly fragmented genomic DNA. each random
fragment is identified by X. Y coordinates representing its length and the Tm of the
domain with the lowest Tm in the fragment. All fragments in which that domain is the
lowest will find a similar gradient level, regardless of their length. This distribution and
the response to specific sequence probes provide a means, in principle, for determining
the spacing, order, and Tm among those domains that have a lower Tm than the average
and those with the highest Tm. It will provide measurements of the nucleotide distances
between each of these domains and any arbitrary set of sequence identifiers or probes.

Our current effort is concerned with refining and calibrating various aspects of the two-
dimensional denaturing gradient technique and related procedures. The.se efforts include
(I) using iron-peroxide nicking and SI cleavage for the preparation of fully random
distribution of fragments from lambda DNA and yeast artificial chromosomes
containing long human genomic inserts; (2) reducing the breadth of bands produced by
very long DNA molecules in the denaturing gradient: (3) analyzing the band broadening
observed when the domain with lowest Tm is surrounded by higher melting domains;
and (4) developing optical, mathematical, and computing procedures for calibrating gel
photographs or autoradiographs in terms of quantitative, point-to-point distributions of
DNA.

82

New Approaches to DNA Mapping: Synthetic
Endonucleases

Betsy M. Sutherland and Gary A. Epling

Biology Department, Brookhaven National Laboratory. Upton. NY 1 1973

(516)'282-3293. FTS 666-3293

Recognition and mapping of functionally important DNA regions (e.g.. regulatory and
coding regions and initiation sequences) can he greatly facilitated by specific DNA
cleavage at such sites. Synthetic endonucleases. able to cleave at regions of functional
importance, will be created by coupling DNA site-specific binding proteins via linker
arms to light-activatable cleaving moieties: specific binding function is provided by the
DNA binding protein; cleavage activity is provided by the activatable cleaving
molecules. A prototype system of Rose-Bengal (RB) coupled via a hexanoic acid linker
to a DNA lesion site-specific monoclonal antibody will be developed for other specific
DNA-binding proteins, including the T7 RNA polymerase and mammalian transcription
initiation factors.

Coupling of RB-hexanoic acid to T7 RNA polymerase via l-ethyl-3-(3-
dimethylaminopropyl) coupling (EDO was found to yield RB-tagged polymerase,
which can specifically bind to a plasmid containing a T7 promoter. However, the
conditions for EDC were sufficiently stringent to result in low final yields of active
polymerase. We designed and synthesized an RB triethylene glycol succinate activated
ester, which can be added to polymerase under buffer conditions optimal for enzyme
stability. Levels of the RB addition that yielded maximum specific binding of the
polymerase to T7 promoter sites were determined. Preliminary results indicate that the
RB-labeled T7 RNA polymerase can mediate the cleavage of T7 DNA to a level of at
least 2.4 cleavages per DNA molecule. The sites of cleavage by the tagged polymerase
are being determined.

83

Abstracts:

Mapping

Instrumentation

Quantitation in Electrophoresis Based on Lasers

E. S. Yeung

Environmental Sciences Program, Ames Laboratory, Ames, lA 5001 1
(515)294-8062

The goal of this project is to develop a novel laser-based imaging technique for
quantitation in gel electrophoresis and in capillary electrophoresis. No stains are
required: thus cost-efficiency, reliability, convenience, and speed of processing are
increased. The fact that the scanning technique uses no mechanical parts adds to the
positional accuracy (resolution) of the measurement. The research is based on indirect
fluorometry and acousto-optic imaging. In indirect fluorometry, a fluorescing ion is
used to elute the sample and thus produce a large fluorescence background signal
throughout the gel. When one of the components of the samples appears, the fluore.scing
ion is displaced; a lower fluorescence signal will then be observed. Since electro-
phoresis is based on charged species, electroneutrality requires a one-to-one
displacement of the fluorescing ion. This negative signal allows nonfluorescing species
to be detected with the high sensitivity nonnally associated with fluorescing species
only, and without staining. The response should be uniform and predictable because it is
derived from the same fluorescing ion. Preliminary results indicate that indirect
fluorometry is feasible for monitoring nucleotides and DNA fragments. Applications to
DNA mapping and sequencing will be pursued.

84

Sequencing Technologies

Scanning Tunneling Microscopy of DNA

Rodney L. Balhurn and Wigbert Siekhaus

Biomedical Sciences Division. Lawrence Livennore National Laboratory.

Livemiore, CA 94550

(415) 422-6284. FTS 532-6284

Researchers at Lawrence Livermore National Laboratory, and at other institutions, have
recently shown that scanning tunneling microscopy (STM) can be performed on the
DNA molecule with angstrom resolution. In addition, STM in the spectroscopic mode
(scanning tunneling spectroscopy or STS) has been used to characterize the electronic
structure of semiconductor substrates and their interaction with molecules adsorbed on
such substrates. The goals of this project are ( 1 ) to develop the instrumentation and
techniques required for imaging naked double- and single-stranded DNA at or near
atomic resolution using STM and (2) to devise methods for obtaining spectroscopic
information with STM that allow us to distinguish between the four bases and to
sequence DNA. To accomplish these goals, the four nucleotides and various single- and
double-stranded DNAs will be imaged, after electrophoresis, onto graphite and other
substrates. Experiments will be performed to determine how specific counterions and
the hydration state of the deposited DNA affect imaging. To identify individual bases
and specific molecular tags, measurements of synthetic and tagged sequences of known
length will be made so that various types of spectroscopy (work function, laser-
enhanced vibration, and photon emission) can be performed on the molecules. Our
application of STM and STS to the analysis and sequencing of DNA should directly
impact the progress of the human genome effort by eventually providing a new,
electronic method for sequencing DNA at three orders of magnitude faster than existing
methods. The techniques and instrumentation developed during the course of this
project will also be directly applicable to the analysis of biological samples in general.

85

Abstracts:

Sequencing

Technologies

Transposon-Facilitated DNA Sequencing

Douglas E. Berg, Clara M. Berg, and Henry Huang

Department ot Microbiology and Immunology. Washington University,

St. Louis. MO 631 10

(314)362-2772

Two types of derivatives of transposon Tn5 will be constructed to facilitate the
sequencing of cloned DN As. One type is designed for in vivo insertion at many sites in
DNAs cloned in lambda phage and in cosmids and other plasmid vectors, and for
sequencing in both directions from each site of insertion. The other type will be
embedded in cosmids and used to generate nested deletions with one variable end point
in the cloned DNA and one end point fixed at a transposon end; the set of nested
deletions will similarly permit sequencing of the entire stretch of cloned DNA without
need for subcloning of random fragments. The Tn5 element for insertion into lambda
and cosmids will contain a supF (suppressor tRNA) gene as a selectable marker. Its
transposition to lambda will be selected by plaque formation, while its transposition to
plasmids will be selected by suppression of a chromosomal amber mutation. Constructs
for making nested deletions by intramolecular transposition will contain a Tn5
transposase gene, whose expression is turned on by IPTG. so that deletions can be made
at will. The construct will also contain a conditionally lethal gene for selection of the
transposition-induced deletions. Deletion-generating Tn5 derivatives will be constructed
as complete cosmid vectors for the construction of new recombinant DNA libraries and
as cassettes for insertion into cosmids that already contain cloned DNAs. Both types of
Tn5 derivatives will be adapted for multiplex sequencing.

86

Computer-Assisted Multiplex DNA Sequencing

G. M. Church. G. Gryan. S. Kieffer-Higgins. L. Mintz, M. J. Rubenfield.

and M. Temple

Department of Genetics. Howard Hughes Medical Institute. Harvard Medical School,

Boston. MA 02138-3800

(617)732-7562

Several laboratories are sequencing genomes (ranging from 1 to 15 Mbp) from each
phylogenetic kingdom. The genome closest to completion is E. coli (20% of 4.7 Mbp).
These sequences will define consensuses for classes of protein domains, evolutionary
conservation, and change. While participating in this quest, we have developed a new
multiplex DNA sequencing method |Church et al.. Science 240, 185-188 ( 1988)], In
multiplex DNA sequencing. 480 sequencing reaction sets, each tagged with specific
oligonucleotides, are run on a single gel in 1 2 pools of 40 and transferred to a
membrane. We hybridize 75 such membranes simultaneously. The resulting sequence
film images are digitized, and sequence interpretations are superimpo.sed on the
enhanced two-dimensional images for editing. The computer program (REPLICA) uses
internal standards from multiplexing to establish lane alignment and lane-specific
reaction rules by discriminant analysis. The automatic reading phase takes one hour per
film (3 kb) on a Vaxstation. Images with overlapping data can be viewed side by side to
facilitate decision making. Hash-table-based routines for linking up shotgun sequences
in the megabase range are compatible in speed with the rest of the software.

87

Abstracts:

Sequencing

Technologies

Sequencing of Megabase Plus DNA by Hybridization:
Method Development

R. Crkvenjakov, R. Drmanac. Z. Strezoska. and I. Labat

Center for Genetic Engineering, Vojvode Stepe 283. P.O. Box 283. 1 1000 Belgrade,

Yugoslavia

38- n -49 1-391

The DNA sequence of a genome can be constructed by joining shorter overlapping
oligomer sequences of DNA. Sequencing by hybridization (SBH) is an implementation
in which the oligomer sequences are accessed through a DNA hybridization
methodology. Oligomer probes are used to ascertain the presence or absence of
complementary sequences in arrayed clones, with the library of overlapping clones
representing the genome. The sequence data gathered earlier provide for the ordering of
the library. The data on the unique regions of overlapping clone pairs have a particularly
important role and serve for the resolution of sequence branch ambiguities that would
otherwise arise in the developing .sequence of single clones. Through computer
modeling, an optimal system configuration is defined that specifies the family of
100,000 oligomer probes needed and the characteristics of the library. For practical
application, probe-template hybrids that are perfectly base-paired must be
distinguishable from those with a base-pair mismatch. Now that an understanding of the
requisite themiodynamics has been achieved and the resolving capacity demonstrated,
oligomers — at least as small as octamers — can serve as probes. These size reductions
are important because costs of oligomer synthesis are thereby greatly decreased. The
continuing development program includes optimization of probe labeling, library
cloning, and management procedures; measurement of the actual error rate in data
acquisition: and a stringent test of the theory. The latter is a computer simulation of data
processing and sequence assembly for a DNA of yeast genome size. A scheme for
extensive miniaturization of the probe-clone array system is under development. The
aim of the project is to provide experience for decision making on whether or not to
proceed to a sequencing pilot plant stage.

*This project is being supported under terms of a Scientific and Technological
Cooperation Treaty between the United States and Yugoslavia, with Yugoslavia
providing the majority of the funds.

88

Rapid Preparation of DNA for Automated Sequencing

John J. Dunn and F. William Studier

Biology Department, Brookhaven National Laboratory, Upton, NY 1 1973

(316)282-3012, FTS 666-3012

A strategy that uses a library of oligonucleotide primers of length eight, nine, or ten has
been developed for direct sequencing of cosmid DN As. The statistics of priming
indicate that a library sufficient for determining the sequence of hundreds of thousands
of different cosmids could be readily assembled. This strategy would greatly reduce the
cost and effort of human genome sequencing. Any needed primer would be instantly
available at a cost of considerably less than 0.1 cent per nucleotide of sequence
obtained. Mapping, subcloning, or preparation of multiple DNA samples would not be
necessary, and the wasteful redundancies of random sequencing strategies would be
eliminated. The success of this strategy requires only that a considerable fraction of all
octamers, nonamers, or decamers be able to prime selectively. Work is under way to
establish conditions where this will be the case. The transposon gamma-delta is being
modified to carry genetic signals that enable bacteriophage T7 to replicate and package
plasmid DNAs. The ability of such an element to insert these signals into a cosmid
DNA and thereby to facilitate preparation of the DNA for sequencing is being tested
using a cosmid that carries a complete genomic copy of the receptor gene for polio
virus.

89

Abstracts:

Sequencing

Technologies

Scanning Tunneling Microscopy

T. L. Ferrell, R. J. Warmack, and Dave Allison

Health and Safety Research Division, Oak Ridge National Laboratory,

Oak Ridge, TN 3783 1

(615) 574-6214, FTS 624-6214

This project includes the operation and continued development of scanning tunneling
microscopes for basic physics research related to health and environmental problems.
The primary focus of this research is the use of scanning tunneling microscopes in
sequencing the human genome. A scanning tunneling microscope with single-atom
resolution is used to image atomic structure on surfaces, to alter atomic positions, and to
probe the dynamical phenomena caused by collective electron motion and motion of
ions. Development includes extending capabilities to a wider range of materials, to
identification of atomic species, and to studies of biological samples. Methodologies are
developed for atom-by-atom studies of biological molecules important in understanding
the human genome and in providing other applications in surface science.

90

Multiplex DNA Sequencing

Robert Weiss and Raymond F. Gesteland

Howard Hughes Medical Institute, Department of Human Genetics, University of Utaii

Medical School, Salt Lake City, UT 841 12

(801)581-5190

We have developed a method for rapid DNA sequencing that employs multiplex
probing, as originally described by Church and Gilbert. A large number (25-50) of
DNA samples, each cloned in an equal number of special vectors, are prepared together
and as a mixture are sequenced using dideoxy sequencing from a universal primer
whose sequence is carried by each vector. After conventional separation by gel
electrophoresis, the mixed DNA pattern is transferred to a membrane. Each of the
individual sequences is then revealed by repetitive rounds of probing, washing, reading,
stripping, and reprobing with labeled oligonucleotide, each of which is unique for a
sequence in one of the vectors. By fixing a number of such membranes in a drum, the
'-P-labeled probing can be done (without handling the membranes) to obtain
10.000-20,000 bases of sequence in each cycle with a cycle time of 1-2 days, including
time for autoradiography — with minimal labor. By using a set of transposons containing
appropriate primer and identifier sequences, we are developing, with help from Diane
Dunn, a method for creating in vivo random subclones that are ready for multiplex
sequencing. An optical lab has been set up by Jeff Ives and Achim Karger with the help
of Joel Harris (Chemistry Department) to compare the feasibility of fluorescence and
chemiluminescent tags for the multiple probes. The efficiency of multiplex sequencing
has been diminished by a bottleneck at the step of reading autoradiograms. To solve this
problem and to deal with data that might come from charge-coupled display (CCD)
images of the fluorescent or chemiluminescent membranes, a research group —
including Jeff Ives, Mike Murdock, and Tom Stockham and Neil Cotter (Electrical
Engineering Department) — has been assembled. Harold Swerdlow is investigating the
feasibility of using gel electrophoresis in microbore capillaries (70 mm) to do
sequencing.

91

Abstracts:

Sequencing

Technologies

DNA Sequencing Using Stable Isotopes

K. B. Jacobson. H. F. Arlinghaus,* G. M. Brown. R. S. Foote. F. A. Larimer,

R. A. Sachleben. N. Thonnard.* and R. P. Woychik

Biology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-8077

(615)574-1204. FTS 624-1204

*Atom Sciences, Inc.

Thi.s project utilizes a new DNA-sequencing approach that could increase the rate of
sequence determination 100-t'old or more as compared to current methods that use
radioisotopes. In this procedure, stable isotopes of sulfur, tin, iron, mercury, and other
elements can be used to label DNA it.self or oligonucleotide probes that will be used to
locate DNA fragments after electrophoresis. Resonance ionization spectroscopy (RIS)
will be used for localization and quantitation of these elements and all their stable
isotopes in an optimal fashion. The sensitivity of RIS, as implemented in Sputter-
Initiated Resonance Ionization Spectroscopy (SIRIS), should be comparable to that of
radioisotopes and fluorescent labels that are in current use. Furthemiore. the multiplex
method of DNA sequencing will be adapted for use with stable isotopes so that 40
labels can be used simultaneously. This adaptation will require that the maximum
number of stable isotopes of a given element be attached to a series of oligonucleotides
in a stable configuration, so that the label does not interfere with accurate hybridization.
Several chemical approaches are presented that should lead to properly labeled probes
that meet these criteria. Current state-of-the-art SIRIS will be used to demonstrate this
new sequencing approach, and the design of the instrument will be modified to adapt it
specifically to DNA sequence determination. The use of radioisotopes is eliminated
along with the attendant problems related to radiation exposure of personnel, the
prohibitive costs of radioactive waste disposal, and the need to cope with the short half-
lives of reagents that contain radioisotopes. A patent application is pending for this
technique. To carry out the goals of this project will require close collaboration of
physicists, chemists, and molecular biologists.

92

Sequencing of Linear Molecules

Joseph M. Jaklevic and W. F. Kolbe

Instrumentation Division. Lawrence Beri<eiey Laboratory. Berkeley. CA 94720

(4LS) 486-.'^647. FTS 431-5647

Innovative methods for determining the linear sequence of nucleic acid bases along
mapped fragments of human chromosomal DNA are required for the efficient
implementation of human genome sequencing. This project involves the investigation of
physical analytical methods for determining the base sequence of DNA. The initial
approach focuses on the development of techniques for manipulating individual DNA
molecules or ordered arrays of identical molecules in a manner that will allow
sequencing of the individual bases using direct spectroscopic methods. An intermediate
step will be an attempt to combine the manipulation of ordered arrays with solid-phase
restnction enzyme chemistry to achieve an advanced method for physical mapping of
intennediate-sized fragments. Since the spatial resolution and analytical sensitivity
required for DNA sequencing are significantly beyond existing capabilities, the
feasibility of combining several innovative technologies will be explored. Initially, we
will investigate low-temperature solid matrices as media for both physical manipulation
and chemical isolation of individual molecules. Subsequent alignment of the DNA
strands using electromagnetic fields will be investigated using direct imaging methods
such as scanning tunneling microscopy and fluorescence microscopy of labeled
molecules. Methods for attaching cloned DNA fragments to appropriate substrates will
also be incorporated into the matrix studies. Spectroscopic methods applicable to the
detection of the DNA molecules at the required level of sensitivity will also be
explored.

93

Abstracts:

Sequencing

Technologies

Advanced Concepts for Base Sequencing in DNA

J. H. Jett, R. A. Keller, J. C. Martin. E. B. Shera

Life Sciences Division, Los Alamos National Laboratory, Los Alamos, NM 87545

(505) 667-3843, FTS 843-3843

We are addressing the problem of rapidly sequencing the bases in large fragments of
DNA. The ideas presented represent the combined effort of a multidisciplinary task
force composed of physicists, physical chemists, cellular and molecular biologists, and
organic chemists. To reduce mapping requirements, the emphasis is on sequencing
methods that are rapid, require little DNA. and are capable of sequencing large
fragments. After evaluation of several physical approaches to sequencing, the decision
was made to proceed with a modified-tlow-cytometer approach that employs laser-
induced fluorescence to detect individual fluorescent molecules. A large fragment of
DNA. approximately 40 kb in length, will be labeled with base-identifying tags and
suspended in the flow stream of a flow cytometer capable of single molecule detection.
The tagged bases will be sequentially cleaved from the single fragment and identified as
the liberated tag passes through the laser beam. We are projecting a sequencing rate of
100 to 1000 b/s on DNA strands approximately 40 kb in length.

94

Ultrasensitive Fluorescence Detection of DNA
in Sequencing Gels

Richard A. Mathies and Konan Peck

Chemistry Department, University of California, Berkeley, CA 94720

(415)642-4192

Our goal is to develop an ultrasensitive apparatus for the detection of fluorescently
labeled DNA fragments and probes in sequencing gels. A detection system has been
developed that is sensitive enough to observe the fluorescence emitted by individual
fluorescent molecules.' The idea is to detect the bursts of fluorescence arising from the
passage of individual molecules through a tightly focused laser beam. A theory for the
signal-to-noise ratio in laser-induced fluorescence has been developed and used to select
the optimal laser power and molecular transit time (Mathies et al., in preparation). This
theory has been satisfactorily tested by studying the fluorescence from phycoerythrin as
a function of the incident laser power and the transmit time of the molecule through the
laser beam. Using these optimized conditions, we have observed the fluorescence of
individual molecules of phycoerythrin in a flow system as they pass through the laser
beam. The calculated autocorrelation function of the detected photons clearly shows
that we are observing correlated bursts of fluorescence from individual molecules. Also,
the frequency of these bursts is linearly related to the concentration as expected for
single-molecule events. Using a hard-wired version of our single-molecule detector, we
have been able to detect phycoerythrin at concentrations as low as 1 x 10" M (Peck et
al., Proc. Natl. Acad. Sci. USA. in press). This detector is three orders of magnitude
more sensitive than conventional fluorescence detection systems. A patent on this on-
line, single-molecule counter has been filed.' We are now in the process of applying this
new technology to the detection of individual molecules on substrates and in gels.

'R. A. Mathies, K. Peck and L. Stryer, "High Sensitivity Fluorescent Single Particle and
Single Molecule Detection Apparatus and Method," patent filed.

95

Informatics

Abstracts

The Human Genome Information Resource

Christian Burks. T. Michael Cannon, Michael J. Cinkosky. James W. Fickett,

C. Edgar Hiklebrand. James H. Jetl. Rebecca J. Koskeia, Frances A. Martinez,

Debra Nelson. Robert M. Pecherer. Karen R. Schenk, Robert D. Sutherland,

David C. Torney. and Clive C. Whittaker

Theoretical Biology and Biophysics Group, Los Alamos National Laboratory,

Los Alamos. NM »1545

(305) 667-7510, FTS 843-7510

Overview. The goal of the Los Alamos National Laboratory (LAND Human Genome
Information Resource (HGIR) — to develop information management and analysis tools
for physical mapping data — reflects an interest in linking databases from two or more
biological disciplines with the long-term objective of extending similar research to other
related data sets such as nucleotide sequences and genetic maps.

The HGIR project has focused initially on developing strategies and tools for
facilitating the flow of data from electrophoretic gels into computers and, once in the
computer, into various formats for access and analysis. The reasons for this focus are
the desirability of examining and manipulating "real" experimental data and HGIR's
close ties with the LANL Life Sciences Division experimental group, who are currently
developing a cosmid-based physical map of human chromosome 16. The flow of data
begins with the digitization and processing of electrophoretic images. Data corre-
sponding to the clones analyzed on the electrophoretic gels are then passed into a
computerized database — the Laboratory Notebook (see below). The availability of these
data for subsequent analysis of clone fingerprints leads to the development of contig
maps and — eventually — comparison to other, related data sets that will allow for
higher-order assemblies and ordering of contigs.

At the end of this flow path, the need for more sophisticated data management, analysis,
and interface tools becomes evident: therefore, the thrust of the HGIR project will shift
to the development of these tools and their use. To facilitate the cross-linking among
multiple levels of physical mapping data, as well as between physical maps and other
related data sets (e.g., sequences and genetic maps), our data structure design work is
being undertaken with this emphasis (and the future extension to yet unrecognized data
structures) in mind. Finally, we plan to design an on-line system and set of interfaces to
provide these data and tools to a user community more broad than that of the current in-
house activity.

Laboratory Notebook. The Laboratory Notebook database is designed to be useful, with
little modification, for most mapping strategies. Representing an electronic (and
advantageous) alternative to the ubiquitous paper notebook traditionally used in the
laboratory to maintain experimental data, the Laboratory Notebook was designed to
manage data resulting not only from experiments, but also from information relating to

96

the corresponding materials, methods, and procedures. The database resides in Sybase, a
relational database management system, on a Sun 4 computer operating under UNIX.

We have developed tools to support the flow of data from the laboratory into the
database and, once in the database, into various fonns suitable for analysis and
presentation. For example, electrophoretic gel data (e.g., DNA fragment sizes) are
transferred directly to the database from the Biolmage Visage 1 10, an image-processing
workstation. Fingerprint annotation is developed and attached to the DNA fragment
data items through a similar mechanism. Reports from these data are created for direct
input to fingerprint analysis software.

The user interface was designed to present a conceptual view of the data without
burdening the user with the need to understand the structure of the database and details
of data storage. Developed using the Sybase's APT-forms, the Laboratory Notebook
requires very little training to use.

97

Abstracts:
Informatics

Human Genome Center, Lawrence Berkeley Laboratory

Charles R. Cantor, W. Chang. D. Gusfield, M. Hutchinson. W. Johnston. G. Lavvler.
P. Li. V. Markowitz. D. Naor. F. Olken. and C. L. Smith
Human Genome Center. Lawrence Berkeley Laboratory. Berkeley. CA 94720
(415) 486-6800. ( FTS ) 4.S 1 -6800

Researchers at the Human Genome Center at Lawrence Berkeley Laboratory (LBL) are
developing computational tools needed to analyze the map. clone, and sequence data
generated from human genome research and to provide the technical foundation for the
construction of a Human Genome Information System (HGIS). Plans call for the
development of logical data models, physical structures, access methods, and query
facilities for map and sequence data. Also to be investigated are methods for fragment
overlap detection, map assembly, and approximate sequence and pattern matching.

Brief abstracts of the individual projects are listed below.

Chromosome-21 Database (W. Johnston. F. Olken. and C. L. Smith) — The current
goal in database management for maps and sequences is a prototype, integrated
chromosome-21 database and mapping system. The system is initially targeted at
groups involved in developing the physical map of chromosome 21 but can be extended
to groups studying other aspects of chromosome 21. This system will provide access to
various types of physical, cytogenetic, and genetic maps of chromosome 21; a
bibliography of chromosome-2 1 literature; map construction algorithms; and graphic
tools for drawing, displaying, and editing maps.

Computer-Assisted Laboratory Notebook (W. Johnston. M. Hutchinson,
V. Markowitz, F. Olken. and C. L. Smith) — Image storage and indexing for gen-
erating physical maps requires interfacing a data management system with a mass
storage facility. The data management group has made progress in using an extended
entity relationship as a basis for a data management system that will combine elements
of a graphical and textual interface with data interchange between numerous
repositories of data. This system will allow data to be loaded in such a way that the data
can be updated, edited, reviewed, queried, and manipulated with ease. The computer-
assisted laboratory notebook also provides a means of tracking individual experiments,
handling inventories, and integrating data for the development of physical maps.

Theory Group (W. Chang. D. Gusfield. G. Lawler. P. Li. and D. Naor) —Theoretical
work IS being done on two problems: string matching for sequence alignment and map
assembly for probed partial digestion experiments. The map assembly effort has looked
at backtracking algorithms, including various special cases (e.g.. probes on adjacent
fragments). In the coming year these researchers expect to continue their investigations
and look especially at incorporating polymorphism data and multiprobe experiments.
The string-matching work has yielded a fast algorithm for approximate string matching
with a bounded number of errors. Also expected to be improved are the error-handling
abilities of the algorithm.

98

GnomeView: A Color Graphics Interface
to the Human Genome

Richard J. Douthart, Victor B. Lortz, and David A. Thumian

Battelle Pacific Northwest Laboratory. Life Sciences Center. Richland. WA 99332

(509)375-2653

Battelle Pacific Northwest Laboratory is developing an interactive, graphical computer
interface to information and data being generated by the human genome project. Color
graphic representations of genetic objects and maps provide the user with a sense of
topology and reveal patterns that are otherwise difficult to detect. The GnomeView
software allows a user to look at the entire human karyotype and manipulate graphical
representations of genetic and physical maps and sequences. GnomeView supports the
creation and display of maps, which are rendered in graphics on a workstation using the
X Window System. The user can zoom, scroll, and select graphical objects. Descriptive
information is available by selecting particular objects or groups of objects. Textual
references may also be followed using GnomeView's hypertext features. GnomeView
includes a network-model database that permits natural and efficient representation of
the hierarchies in genomic information. Landmarks, features, and blocks of sequence
information are stored as objects in the database. Compound objects and maps are
created by reference to other objects. Specific attention has been paid to designing
database search algorithms and user interface techniques that scale well to high data
volumes. We expect that eventually the GnomeView software will contribute
substantially to obtaining closure of the various types of maps currently being
generated. For instance, a restriction enzyme map can be updated from analysis of
sequence information that sets exactly the position of cut sites. Different maps covering
the same region can be compared visually and analytically for inconsistencies.

99

Abstracts:
Informatics

Identification of Genes in Anonymous DN A Sequences

C. A. Fields and C. A. Soderlund

Center for Advanced Computing in Molecular and Cellular Biology, Computing

Research Laboratory, New Mexico State University, Las Cruces, NM 88003-0001

(505)646-2848

The objective of this project is the development of practical software to automate the
identification of genes in anonymous DN A sequences from the human and other higher
eukaryotic genomes. A prototype automated sequence analysis system, gm2, has been
implemented in the programming language of C for Unix version 4.2. This system
accepts for input; DNA sequences; consensus matrices for locating splice sites,
translational start sites, and polyadenylation sites: match-quality cutoffs for consensus
searches; and base frequency and codon usage standards for coding regions and introns.
It produces, as output, schematic models of the possible genes contained in the sequence
that show the locations of the coding sequences, introns, and control signals. Exten-
sively tested on sequences in the 10-kb size range containing known genes of up to
10 exons, gnil is capable of generating complete and correct analyses showing all
possible alternative splicing patterns. Run times for such analyses on a Sun 3/60
workstation range from less than 1 min to about 45 min and depend on the stringency
of the search parameters used. Current effort is focused on implementing procedures
for analyzing sequences that contain only partial genes and on implementing a more
efficient algorithm for first-pass analysis using low-stringency parameters.

100

Special-Purpose VLSI-Based System for the Analysis of
Genetic Sequences

T. Hunkapiller, M. Waterman, R. Jones, M. Eggert, E. Chow, J. Peterson, and L. Hood
Department of Biology, California Institute of Technology, Pasadena, CA 91 125

(818)356-6408

The current size of genetic sequence databases means that extensive similarity analyses
based on robust mathematical models require large, expensive computer hardware not
generally available to most investigators. We are currently involved in developing a
hardware alternative — a VLSI-based system (i.e., very large scale integrated) — that will
give most laboratories access to these rigorous algorithms at the level of affordable
workstations and/or PCs. We are designing a board-level biological information signal
processor (BISP) coprocessor assembly. BISP is a systolic implementation of dynamic
programming methods for the determination of local sequence similarities and
alignments. Each BISP board will include an array of BISP processor chips responsible
for performing the dynamic programming methods and a tightly coupled coprocessor
(i860) that will provide complete alignments to the host central processing unit (CPU)
from the scoring information and locations provided by the systolic array. Difference
tables and gapping penalties are completely user-definable. Also, there are no arbitrary
limits on sequence lengths of either the query or database sequences, and searches can
be made simultaneously for multiple query sequences (depending on the size of the
sequences and the systolic array). The array will process at up to 20 megacharacters per
second. At this maximum speed, a query sequence the size of the systolic array could be
compared against all of the current GenBank® (both orientations) in under 3 s. The size
of the array is extensible and will range from about 500 processor cells to several
thousand. The BISP chip is being implemented in 1 mm cMOS technology, based on
custom standard cell methods. Each chip will have multiple identical processor cells,
local control circuits, and a results cache. The nearly full-custom processor cell design
is nearing completion, and the control circuit logic and floor plan have been generally
detailed. Our present schedule calls for tested design vectors to be completed by late fall
and handed over to the fabrication service for prototype chip delivery by winter 1990.

101

Abstracts:
Informatics

Human Genome Management Information System

Betty K. Mansfield. Judy M. Wyrick. John S. Wassom, Po-Yung Lu.

Mary A. Gillespie, and Sandy E. McNeill

Health and Safety Research Division, Oak Ridge National Laboratory,

Oak Ridge, TN 3783 1 -6050

(615) 576-6669. FTS 626-6669

The Human Genome Management Information System (HGMIS), sponsored by the
U.S. Department of Energy (DOE I at the Oak Ridge National Laboratory, has the
following roles in the Human Genome Program: ( I ) to assist the DOE Office of Health
and Environmental Research (OHER) in communicating issues relevant to human
genome research to DOE contractors and grantees and to the public and (2) to provide a
forum for exchange of information among individuals involved in genome research or
the development of instrumentation and methodologies to implement genome research.
To fulfill these communications goals. HGMIS is producing technical reports. DOE
Human Genome Program reports, a quarterly newsletter, and an electronic bulletin
board. These documents/facilities are available to all interested persons upon request.
The first technical report will assess instrumentation and methodology development
relevant to DNA mapping and sequencing. The DOE Human Genome Program reports
will contain information on human genome research and development activities
supported by the DOE program, as well as background information. The Human
Genome Quarterly newsletter features technical articles, meeting reports, a calendar of
genome events, announcements, and other information relevant to genome research.
Accessible via modem through direct dial or via the user's host mainframe computer
network, the electronic bulletin board contains information organized by categories
(e.g., menu, general information, news, and comments from OHER; summaries and
highlights of research projects; meeting announcements/calendar; literature highlights;
DOE Human Genome Program contacts; and international activities). HGMIS
welcomes comments, suggestions, and contributions from the genome research
community.

102

Applications of Logic Programming and Parallel
Computation in Genetic Sequence Analysis

Ross Overbeek, Ian Foster, and Steve Winker

Mathematics and Computer Science Division, Argonne National Laboratory,

Argonne, IL 60439

(312) 972-7856, FTS 972-7856

Several applications of logic programming and parallel computations to genome
research problems are being pursued in close collaboration with G. Church (Harvard
Medical School), C. Woese and G. Olsen (University of Illinois at Urbana), and M.
Liebman (Amoco Technology Co.). ( 1 ) For interpretation of sequencing gels, a program
has been developed that mimics human expert behavior and includes extensive
backtracking to explore sets of alternative interpretations. (2) For multiple sequence
alignment, a new algorithm has been devised. It implements several heuristic methods
to produce alignments of up to 500 sequences with lengths of 2000 nucleotides. With
the use of bilingual programming (logic programming with C routines that handle well-
defined computationally intensive subcomputations), prototype algorithms could be
rapidly tested and executed on multiprocessors. (3) For prediction of secondary
structure manifesting in RNAs, an approach based on covariance analysis has been
developed that heavily utilizes multiple sequence alignment. Current applications are
the explorations of possible interactions between 5S, I6S, and 23S RNAs. (4) For the
problem of constructing databases with very diverse data types, uses of logic
programming are being explored. To increase capacities to handle large data sets in
disk, usual logic programming environments have been extended by the addition of
predicates. The result is a prototyping environment that can manage the increasing
volume of sequence data. (5) For comparisons of relative cost and performance,
sequence similarity search algorithms are being compared on several commercially
produced multiprocessors. These include the shared memory MIMD (multiple
instruction, multiple data) machines, SIMD (single instruction, multiple data) machines
[similar to the Connection Machine™ (Thinking Machines, Inc.) and DAP™ (Active
Memory Technology)], and distributed memory machines. Logic programming
technology is also being utilized here to explore solutions that are portable over a wide
range of machine environments.

103

Abstracts;
Informatics

Computational Support for the Mapping of Complex
Genomes

Karl M. Sirotkin. Daniel S. Caugherty. David C. Tomey. and Frederic R. Fairfield
Center for Human Genome Studies. Group T-IO, Mail Stop K710. Los Alamos National
Laboratory, Los Alamos. NM 87545
(505) 667-7510. FTS 855-0479

At Los Alamos National Laboratory (LAND, we are determining how to construct
contigs (contiguous segments of DNA) from fingerprint data and are making tools to
evaluate a number of methods to optimize the rate of contig extension and completion
of a map. From the fingerprint data generated at LANL, contigs are assembled by using
calculated overlap-probability values for pairs of cosmid clones. These overlap
probabilities are determined from a statistical model that incorporates the positions of
the fingerprint data on a gel and the reproducibility of this data. Extensions of these
results will include calculations of the likely structure of a contig (including clone
positions and partial restriction maps) and analysis of the position-dependence of the
chromosome- 16 fingerprint data.

To evaluate mapping methods, we have created a set of modular programs that
detemiine the limitations on contig size when a particular mapping strategy is used.
A person using these programs can create a test genomic segment, generate clones from
this segment, extract fingerprint data from each clone, test various strategies for
determining pairwise overlap between clones, reassemble the genome from these
overlaps, display the resulting contigs, and evaluate the success of the reassembly. The
programs have been structured to evaluate many mapping strategies and to assemble
contigs from current data. To facilitate the use of these programs by others, both
program and data files are readable by the user, and the same symbolic parameter names
are used throughout the data and program files. In addition, we have designed the
program structure so that it is easy to tailor the installation for individual users.

These modeling programs have been implemented in two phases. The first phase used
only exact fingerprint data: in the second phase, there were controlled levels of error in
the fingerprint data. By using exact data, four different strategies reassembled similar
percentages of a genome segment into contigs, although the exact contigs were
different. By combining information from all of these methods, the coverage of the
genome by contigs increased. With errors in the fragment lengths, it is not yet clear
which strategies for reassembly of fragments into contigs make better use of the data.
Small changes in the experimental errors seem to have large effects on contig
generation. With this project, we hope to optimize map construction rates by making
contigs, monitoring our progress, and circumventing potential problems with map
completion.

104

Small Business Innovative Research (SBIR) —

Phase I (1989 Awards)

Instrumentation for Automated Colony Processing

Norman G. Anderson

Large Scale Biology Corporation, Roci<vilie, MD 20855
(301)424-5989

The objective of this project is ( 1 ) to process large numbers of clones through colony
picking, (2) to purify colonies by recloning as necessary, (3) to reposition the colonies
in large rectangular arrays, (4) to process the DNA from each colony to allow the
transfer of recombinant DNAs to the nitrocellulose or other filter materials, (5) to probe
the arrays en masse with DNA probes under different conditions of stringency, and (6)
to identify and order overlapping clones. We propose to develop an automatic cloning,
repositioning, and hybridization analysis system based on proprietary technology that
can ultimately handle sets as large as 2 million clones. The system is designed to do
large matrix hybridization analyses at different and closely controlled stringencies and
to recover the filter materials used for multiple reprobing. Key elements in the design
are positive clone identification, arrangements for the production of duplicate sets of
large arrays, and long-term storage of these arrays. The entire system is designed to run
automatically in a remote environment under sterile conditions.

105

Abstracts:
SBIR-Phase I

Increased Speed in DNA Sequencing by Utilizing LARIS
To Localize Multiple Stable Isotope Labeled Fragments

Heinrich F. Arlinghaus

Atom Sciences. Inc., Oak Ridge. TN 37830
(615)483-1113

The need to map and sequence the DNA in the human genome is of crucial impoilance
to future medical advances and economic competitiveness. Utilization of DNA probes
labeled with dozens of stable isotopes, instead of a single radioisotope, could increase
the rate of sequence determination 100-fold or more, compared with current methods.
The new approach described here — laser atomization resonance ionization spectroscopy
(LARIS) — will be used to localize and quantify these isotopes in DNA fragments that
had been separated by polyacrylaniide gel electrophoresis. The unique selectivity and
ultimate sensitivity made possible by the LARIS technique allow detection of isotope-
tagged DNA fragments in the 10"*-mol. or lower, range without isobaric or other
interferences. This technique can be applied not only to genome sequencing but also to
hybridization methods used in genetic engineering. For the proposed experiments in
Phase I, we will use the same experimental apparatus used by the ongoing sputter-
initiated RIS (SIRIS) DNA sequencing measurements. Doing this allows comparison
and evaluation of these two techniques. During Phase II, recommendations will be made
for the most fruitful directions in DNA sequencing development, and an RIS system
dedicated to DNA sequencing will be specified.

106

PC Program for Automated Analysis of Stained DNA Gels

Jeffrey M. Stiegman

BioPhotonics Corporation. Ann Arbor. Ml 4X106
(313)426-8299

Electronic image processing of one-dimensional ( 1-D) or two-dimensional (2-D)
electrophoretic gels requires specialized software for analysis. Although several
commercially available packages provide generic image enhancement and spatial
analysis, a high degree of user modification is necessary to produce useful data for 1 -D
or 2-D gels. Dedicated software for electrophoretic gels has focused primarily on 2-D
protein separation. However, the application of electronic image analysis to 1-D
separations of DNA fragments has received relatively little attention.

The goal of the BioPhotonics Corporation project is to develop computer software for
capturing, enhancing, and documenting electronically acquired images of 1-D
electrophoretic separations of tluorescently stained DNA fragments and for sub-
sequently converting this data to molecular weight and concentration information to be
used for statistical analysis and restriction fragment mapping. The emphasis of this
project is to automate fully the procedures for locating band positions, detennining
relative DNA concentration, calculating molecular weights, and analyzing migration
abnormalities. This approach will provide improvement in time and accuracy many
times over that of manual methods.

107

Abstracts:
SBIR-Phase I

Rapid Genome Analysis on a Workstation with an
Associative Coprocessor

Charles D. Stormon, James Brule, and Hamid Bacha
Coherent Research, Inc., Syracuse. NY 13202
(315)426-0929

The goal of this research project is to produce a high-performance, low-cost sequence
analysis workstation that will allow genetic researchers to perfomi complex analyses on
genome data. Phase I research involves feasibility studies of processing genome data at
a workstation with an associative coprocessor. Many of the DNA sequence analy.ses that
molecular biologists require involve finding similarities between different molecules or
within one molecule. Several different approaches, represented by dozens of programs,
have been taken to solve this problem. With increasing sequence length (or larger
numbers of fragments) to be analyzed, these methods can quickly become compu-
tationally impractical. Recently, parallel processors have become available, and they
provide an avenue of speedup for sequence analysis.

By implementing a simplified computational model in VLSI (very large scale
integrated), our associative Coherent Processor^*^ (the AP-DS) provides the same
degree of parallel processing as either the Connection Machine™ (Thinking Machines,
Inc.) or the DAP^" (Active Memory Technology) at a much lower cost. The research in
this project will show the feasibility of processing complex genome queries on a
workstation-class machine with attached AP-DS coprocessor.

108

New Recording Media for an Automated Genome
Mapping System

George M. Storti

Quantex Corporation, Rockville. MD 20850

(301)258-2701

New recording media for an automated genome mapping system is the goal of this
project. The media are different formulations of Quantex's electron trapping (ET^^' )
materials that have the capability of recording beta particle events and events produced
by visible light of relatively short wavelength. Consequently, the presence of
radioactively tagged and dye-tagged DNA fragments can be recorded, and event
information can be integrated. Readout will be performed by an X-Y scanning system
that allows for large signal-to-noise ratios. Compatibility of a scanning system with
large-area DNA gels will be evaluated.

109

Abstracts:
SBIR-Phase I

Improvements in DNA Sequencing Methods by
Incorporating Chemiluminescent Detection

John C. Vovta

Tropix. Inc.. Bedford. MA 01730
(617) 271-0045

The research plan for Phase I of this project will involve feasibility studies of the use of
chemiluminescence in DNA sequencing. With the advent of enzymatically induced
chemiluminescence from stable precursors and the improved instrumentation for
luminometry. ultrasensitive detection of a wide variety of agents under many different
conditions is now possible. With the implementation of the chemiluminescent substrate
technology, we believe that it will be possible to improve several aspects of DNA
sequencing methodology such as stabilizing reagents, eliminating radioisotopic labels,
shortening the incubation times, improving detection sensitivities, and greatly
simplifying the detection instrumentation. To verify the viability of such a system, we
plan to use an avidin/biotin detection system with alkaline phosphatase as the signal-
generating label. In one study, biotinylated primer strands and other necessary reagents
will be used in the dideoxy-sequencing reactions and separated via gel electrophoresis.
Subsequently, the separated DNA will be visualized with avidin-labeled alkaline
phosphatase and a chemiluminescent substrate. The use of chemiluminescent substrates
in multiplex-sequencing protocols will also be investigated. In Phase I, the luminescent
detection of nucleotides will be accomplished with X-ray and instant films. The
anticipated improvements in chemiluminescence-based DNA sequencing will be fully
explored in Phase II with the implementation of full -performance protocols coupled
with the design and prototype generation of a simple and inexpensive luminescent gel/
blot scanner capable of high throughput. Fmally. we believe that these new and
improved sequencers will offer speed, cost effectiveness, and ease of use in DNA
sequence determinations.

Small Business Innovative Research (SBIR) —

Phase II (1988, 1989 Awards)

Chemiluminescent Labels for Polynucleotides
Electrophoretically Separated in Agarose Gels*

Edward M. Davis

SymBiotech. Inc.. Cheshire. 07 06410
(203)272-4190

The objective of this project is to continue the work, started during Phase I. which
demonstrated the feasibility of employing chemiluminescence to visualize poly-
nucleotides separated by electrophoresis in agarose gels. Chemiluminescence is a very
powerful labeling method and, under optimal conditions, yields detection limits equal to
or better than radioisotopic labels. Much Phase II work will be directed toward
improving the detection limits for DNA in electrophoresis gels. This improvement will
be accomplished by developing ways to decrease background luminescence and
increase the efficiency of biotinylation of DNA. as well as to explore more powerful
luminescent reagents. Ultimately, the aim is to develop highly sensitive and safe labels
that can be detected with inexpensive photography or, alternatively, can be recorded,
processed, and stored by an electronic digital imaging system. The potential of
employing chemiluminescence for labeling DNA fragments in sequencing gels will also
be explored. It is anticipated that several products will be developed from this work
including a combination electrophoretic device for labeling and photographing
luminous DNA. In addition, new reagents for improved luminescence and biotinylation
of DNA will be developed for commercialization.

*1988 award for two years.

Abstracts:
SBIR— Phase II

The Separation of Large DNA Fragments with Oscillating
Electric and Magnetic Fields*

Gunter A. Hofniann

BTX/Biotechnologies and Experimental Research, Inc., San Diego, CA 92109
(619)270-0861

The accurate and fast separation of large DNA fragments is a crucial technology
needed for mapping the human genome. Existing methods such as pulsed-field gel
electrophoresis appear to have severe limitations. This project uses a novel approach:
Lorentz-force-mediated separation in the form of oscillating electric and magnetic
fields. DNA exhibits a large induced dipole moment at low frequencies, with a
relaxation time dependent on the length of the fragment. The planned separation method
makes use of the polarizability of DNA fragments by subjecting them to an oscillating
electric field with a superimposed perpendicular oscillating magnetic field in a liquid
without matrix. The resulting unidirectional Lorentz force moves the DNA fragments
perpendicular to both the electric and the magnetic field with a drift velocity dependent
on the DNA polarizability. which depends in turn on the DNA size and configuration.
Phase I studies demonstrated that DNA fragments can be polarized by an oscillating
electric field and that a superimposed magnetic field results in a unidirectional Lorentz
force and a unidirectional drift velocity. The drift velocity increases with the size of the
DNA fragment, in contrast to conventional electrophoresis. For large DNA fragments,
the drift velocity is two to three orders of magnitude higher than the drift velocity
observed in experiments in pulsed-field gel electrophoresis. A theoretical model with a
porous sphere simulation of the DNA molecule showed promise in yielding some
qualitative agreements with the observed experimental dependencies of the drift
velocity. Phase II studies will develop several prototype apparatuses of increasing
complexity that will allow the rapid, accurate separation of large DNA fragments. A
wider parameter range will be investigated, especially separation at higher frequencies,
and optimum operating conditions will be determined. The limits of the crossed-field
separation method will be investigated in experiments and theory.

•1989 award for two years.

112

An Image Acquisition and Processing System for the
Analysis of Fluorescence from Stained DNA Gels*

Ronald A. McKean

KMS Fusion. Inc.. Ann Arbor. MI 48106
(313)769-8500

Current methods that use fluorescence techniques to analyze stained DNA gels are
inadequate because the test results ( 1 ) cannot be easily accessed by a computer and (2)
cannot be precisely reproduced. The lack of available instrumentation to convert a
fluorescing image into a digitized record for computer entry and the lack of any
technique for standardizing analyses performed under varying conditions severely limit
the usefulness of this analysis. Overcoming these problems is essential as the need
increases for an efficient means of performing both the analysis and the statistical
review of the results. The goal of this project is to develop an instrument that will
digitize stained DNA directly from agarose gels, standardize the results, and provide
statistical analysis features. This instrument will quickly scan the gel with an optical
system capable of high photometric resolution as well as very high spatial precision.
The digitized data will then be electronically preprocessed (e.g., background sub-
traction, filtering, data compression) to ensure correct, noise-free data structured for
storage with minimal memory requirements. The stored data, then available for use in
imaging the gel, will produce statistical comparisons and generate graphical displays.
The data will be archivable using media such as magnetic tapes or disks. Phase I efforts
have been highly successful. The results demonstrated ( I ) acquisition of high-quality
image data directly from agarose gels without elaborate or high-cost components, (2)
feasibility of standardizing DNA results under varying electrophoretic conditions, and
(3) preliminary optical and electronic designs for this instrument. Phase I results thus
show the feasibility of the instrument that will solve many problems associated with
agarose gel analysis of DNA.

*1988 award for two years.

113

Abstracts:
SBIR-Phase II

A Novel, Low-Cost System for Automated DNA
Sequence Reading*

John West

BioAutomation. Inc.. Bridgeport. PA 19405
(215)275-4540

Analytical instruments tor the automated measurement of DNA are just emerging.
Automated sequence readers commercialized to date are expensive (over $90,000 per
instrument). As a result, in more than 10,000 sequencing laboratories worldwide, fewer
than 3% have acquired such instrumentation. Because of the high cost, adoption of
automated DNA-sequencing instrumentation has been slow even though interest in the
technology is high. First-generation sequence readers also lack the throughput required
for large-scale sequencing projects. Technological feasibility of an autoradiography-
based approach incorporating the use of application-specific, highly integrated circuits
and parallel processing was demonstrated in Phase 1. During Phase II. full working
prototypes will be developed using this approach. These prototypes will ( 1 ) provide a
significant increase in accessibility of automated sequence-reading equipment to smaller
laboratories, and (2) provide the quantum jump in throughput required by large-scale
sequencing projects.

Being proprietary, the specific technical approach to be followed is omitted from this
abstract.

* 1989 award for two years.

114

Index to Project Investigators

Ezra Abrams

82

Edward M. Davis

HI

David Allison

90

Larry L. Deaven

53

Norman G. Anderson

105

N. A. Doggett

72

Grai Andreason

69

Richard J. Douthart

99

S. E. Antonarai^is

62

R. Drmanac

88

H. Arenstorf

61

John J. Dunn

89

Heinrich F. Arlingiiaus

92. 106

M. Eggert

101

Raghbir S. Athwal

32

Gary A. Epling

83

Hamid Bacha

108

M. Esposito

64

Rodney L. Balhom

85

James Eubanks

69

David F. Barker

63

Glen A. Evans

69

Clara M. Berg

86

Kathryn C. Evans

69

Douglas E. Berg

86

Pamela R. Fain

63

Tony J. Beugelsdigk

77

Frederick R. Fairfield

104

Bruce Birren

74

H. Fang

75

Elbert W. Branscomb

66

T. L. Ferrell

90

G. M. Brown

92

James W. Fickett

96

James Brule

108

C. A. Fields

100

Christian Burks

96

R. S. Foote

92

C. Bustamante

64,78

Ian Foster

103

David F. Callen

76

Nashua Gabra

82

T. Michael Cannon

96

Emilio Garcia

66

Charles R. Cantor

64, 78, 98

Joe M. Gate wood

58

Anthony V. Carrano

59,66

Raymond F. Gesteland

91

C. Thomas Caskey

68

J. Calvin Giddings

54

Daniel S. Caugherty

104

Mary A. Gillespie

102

S. C. Chandrasekharappa 71

J. Gingrich

64, 75.

78

W. Chang

98

David E. Goldgar

63

Shizhong Chen

69

Joe Gray

59

J. S. Cheng

75

G. Gryan

87

E. Chow

101

D. Gusfield

98

George M. Church

87

James F. Hainfeld

81

Michael J. Cinkosky

96

John Hanish

71

Radomir Crkvenjakov

88

Reece Hart

69

Jack B. Davidson

80

A. Hassenfeld

78

Abstracts

15

Abstracts: Index to
Project Investigators

Richard P. Haugland

55

Gary Hennanson

69

Peter C. Hewitt

55

P. A. Hieter

62

C. E. Hiidebrand

33,

72,

Robert Hoilen

77

Gunter A. Hofmann

112

Leroy Hood

101

Henry Huang

86

Hans Huber

57

T. Hunkapilier

101

M. Hutchinson

98

Valentine J. Hyiand

76

K. Bruce Jacobson

92

Joseph M. Jakievic

78.

93

J. H. Jett

94,

96

W. Johnston

78,

98

R. Jones

101

Pieter J. de Jong

59,

66

R. Kandpal

61

Hee-Chol Kang

55

J. Katz

78

Daniel Kaufman

69

R. A. Keller

94

S. Kieffer-Higgens

87

W. F. Kolbe

78,

93

Rebecca J. Koskela

96

I. Labat

88

Carlisle P. Landel

69

F. A. Larimer

92

G. Lawler

98

Michelle M. Le Beau

71

David H. Ledbetter

68

Ester P. Leeflang

58

L. S. Lerman

82

96

S. Levene

64,78

Kathy A. Lewis

69

S. Lewis

78

P.Li

98

Jon Longmire

53

Victor B. Lortz

99

Po-Yung Lu

102

M. Maestre

64,78

Donna R. Maglott

56

Betty K. Mansfield

102

V. Markowitz

98

J. C. Martin

94

Frances A. Martinez

96

Richard A. Mathies

95

Michael McClelland

71

M. K. McCormick

62

David McElligott

69

Ronald A. McKean

113

Sandy E. McNeill

102

Julie Meyne

73

W. Michels

75

L. Mintz

87

H. W. Mohrenweiser

66

R. Mortimer

64

Robert K. Moyzis

53, 72, 73

John C. Mulley

76

D. Naor

98

David L. Nelson

68

Debra Nelson

96

Mike Nelson

71

William C. Nierman

56

Arnold R. Oliphant

63

F. OIken

98

Ross Overbeek

103

S. Pari moo

61

116

Yogesh Patel

71

Grant R. Sutherland

76

Robert M. Pecherer

96

Robert D. Sutherland

96

Konan Peck

95

A. Swaroop

61

J. Peterson

101

Stanley Tabor

57

Ken O. Pischel

69

M. Temple

87

Kimball 0. Pomeroy

69

E. Theil

78

Robert Ratliff

73

N. Thonnard

92

Michelle Rebelsky

71

David A. Thurman

99

Robert 1. Richards

76

David C. Tomey

96,

Charles C. Richardson

57

Barbara Trask

59

M. J. Rubent'ield

87

Joseph Trotter

69

John Rush

57

Ger van den Engh

59

R. A. Sachleben

92

Marvin A. Van Dilla

59

Mary Saleh

69

John C. Voyta

110

M. Salmeron

64

D. Wang

75

Karen R. Schenk

96

Gui-Lin Wang

58

Carl W. Schmid

58

D. Ward

61

Eric Schmitt

82

R. J. Wamiack

90

E. B. Shera

94

John S. Wassom

102

Hiroaki Shizuya

74

M. Waterman

101

Wigbert Siekhaus

85

Mike Weil

71

Martha N. Simon

81

Robert Weiss

91

Melvin I. Simon

74

Shenman M. Weissman

61

Karl Sirotkin

104

John West

114

Thomos Slezak

66

Carol A. Westbrook

71

Cassandra L. Smith

64,75,98

James E. Whitaker

55

C. A. Soderlund

100

Clive C. Whittaker

96

R. L. Stallings

72

Stephen G. Will

81

Jeffrey M. Stiegman

107

Huntington F. Willard

63

M. Stoneking

64

Steve Winker

103

Charles D. Stormon

108

R. P. Woychik

92

George M. Storti

109

Judy M. Wyrick

102

Z. Strezoska

88

Y. Wu

75

F. William Studier

89

E. S. Yeung

84

Betsy M. Sutherland

83

Kathy Yokobata

59

Jun Zhao

69

104

117

Resource Development

DNA sequence data output. Genetically modified T7 DNA polymerase lacking
3' to 5' exonuclease activity was used in the presence of manganese ions
(Mn-'*) to catalyze the four separate polymerization reactions prior to sequencing
the vector, M13 mp18 DNA, on an Applied Biosystems Model 370A Automated
Sequencing System. Each polymenzation reaction was earned out with a primer
labeled with a different fluorophore and a different dideoxynucleoside
triphosphate chain terminator [Smith et al., Nature32^, 674 (1986)]. After
completing the polymerization reactions, the mixtures were combined and run in
a single lane on a denaturing polyacrylamide gel. The fluorescently labeled
bands separated during electrophoresis were analyzed using the Applied
Biosystems 370A. The red. blue, green, and black peaks correspond to ddTMP-.
ddCMP-, ddAtVlP-, and ddGMP-terminated fragments, respectively. Shown is a
part of three consecutive panels of a 450-nucleotide region of Ml 3 mpl 8 DNA;
the sequence of the first 20 nucleotides is TCGTACTCTAGAGGATCCCC. The
height of the signal decreases with the length of the fragment, because
statistically there are fewer terminations at each position with increasing length
of the DNA strand being sequenced.

Because native T 7 DNA polymerase has an inherent 3' to 5' exonuclease
activity that causes premature terminations of polymerizations at pause sites,
T7 DNA polymerase has been modified genetically so that it no longer has 3' to
5' exonuclease activity. However, when the genetically modified T7 DNA
polymerase discriminated against the dideoxynucleoside triphosphates, uneven
band intensities resulted when the usual polymerization reaction component —
magnesium (Mg-*)— was present. Substituting Mn-* in the polymerization
reaction alleviated the problem of nonuniform DNA band intensity [Tabor and
Richardson, Proc. Natl. Acad. Sci. USASe. 4076 (1989)]. (Photograph provided
by S. Tabor and C. C. Richardson, Harvard Medical School.)

Appendices

37-15

iTwmrw

.DAT Clvl34 Cpmments: ^ „

119

Fig. 1. The human genome at four levels of detail. Apart from reproductive
cells (gametes) and mature red blood cells, each cell of tfie human body
contains 23 pairs of chromosomes, each a packet of compressed and entwined
DNA. Each strand of the DNA is a huge natural polymer consisting of repeating
nucleotide units — each of which is comprised of a phosphate group, a sugar
(deoxyribose), and a base (either guanine, cytosine, thymine, or adenine). In its
"normal" state, DNA takes the form of a highly regular double-stranded helix, the
strands of which are linked by hydrogen bonds between guanine and cytosine
and between thymine and adenine. Each such linkage is said to constitute a
"base pair": some 3 billion bp constitute the human genome. The specificity of
these base-pair linkages underlies the mechanism of DNA replication illustrated
here. Each strand of the double helix serves as a template for the synthesis of a
new strand; the nucleotide sequence (i.e., linear order along the DNA strand) of
each strand is strictly determined. Each daughter double helix is thus not only a
twin, but also an exact replica of its sole parent. (Figure and caption text
provided by the Human Genome Center, Lawrence Berkeley Laboratory.)

120

Primer on Molecular Genetics

Appendix A

121

Appendix A:
Primer on Molecular
Genetics

The text for this primer was adapted
from material and figures that were
supplied by Charles Cantor and
Sylvia Spengler of the Human
Genome Center, Lawrence
Berkeley Laboratory.

Fig. 2. A diagram of a very short
section of the human genome,
which contains about 200 million
times the amount of DNA shown
in this section. [U.S. Congress,
Office of Technology Assessment.
Mapping Our Genes — The Genome
Projects: How Big, How Fast? OTA-
BA-373 (Washington, D.C.: U.S.
Government Printing Office, April
1988).]

Base

Major
groove

Introduction to the Genome

The genome is the total complement of genetic material present in a single cell.
This genetic material is deo.wribonucleic acid — DNA — the blueprint tor all
cellular activities for the lifetime of the cell or organism. Each of a person's 10
trillion cells contains essentially the same complement of DNA (except for mature red
blood cells, which contain no chromosomal DNA).

Chromosomes

The human genome, containing ?i billion base pairs of DNA. is divided into 23 pairs of
physically separate units called chromosomes. Chromosomes, which are located in
cellular nuclei (Fig. 1. p. 121). contain roughly equal parts of protein and DNA. Each
chromosome has a single DNA molecule (packaged in a complex hierarchy and whose
number of bases average \5i) million) that would be 2 in. long if released from the cell
and stretched out. DNA molecules are the largest molecules now known.

Chromosomes can be seen under the light microscope. Cytological stains reveal a
pattern of light and dark bands that reflect local variations in the percentage of Adenine
and Thymine versus that of Guanine and Cytosine in particular regions. Differences in
size and banding pattern allow each of the 24 chromosomes to be distinguished from the
other. With microscopic examination, one can detect occasional major chromosomal
abnormalities that indicate differences in the genomes of some individuals. However,
the majority of DNA differences are more subtle and can be detected only by molecular

analysis. These abnomialities in DNA
may be responsible for inherited
diseases or cancer.

DNA

The DNA structure consists of a long,
ribbon-like chain made of two strands
and divided into subunits. Each
position along a DNA strand can be
occupied by one of four nitrogenous
bases — adenine (A), thymine (T).
cytosine (C), or guanine (G) — strung
together along a backbone of sugars
and phosphates (Figs. I and 2). The
sequence of bases along the sugar-
Hvdroeen phosphate backbone encodes the

bond

Sugar-phosphate
backbone

groove

122

genetic information. The two DNA strands are related by the following rules for
matching — pairing — the four bases:

• A on one strand is always matched by T on the other to form an A-T base pair.

• G on one strand is always matched by C on the other to form a G-C base pair.

Adherence to these rules ensures that a duplicate information set is available to correct
errors and thus minimize the mutation rate.

When a cell divides, its DNA divides too. Using the rules for forming base pairs, each
strand directs the synthesis of a complementary new strand. Each daughter cell receives
one old and one new DNA strand (Figs. 1 and 3).

Fig. 3. Replication of DNA.

Complementary DNA strands
separate; each strand then
becomes the template for
synthesis of a new strand.

123

Appendix A:
Primer on Molecular
Genetics

Genes

Genes — the functional unit of genetic infonnation — encode the DNA sequences
required to make a protein. Genes also contain information that is used to regulate the
kind and amount of protein made in a particular type of cell.

The human genome has 50.000 to 100.000 genes. A typical gene is 90% stuffer —
introns — whose function is unknown. A human gene may contain up to 30,000 base
pairs, but only about 10% of the base pairs are known, with certainty, to contain useful
information. This infonnation is read out in discontinuous blocks, called exons (Fig. 4).
Since three bases code for one amino acid (the building blocks of protein.s), the average
size of the protein coded by a gene will be 1000 amino acids. In addition to introns and
sequences that regulate the amount of protein made, special regions within the DNA
help organize it into chromosomes and control its replication.

Fig. 4. Structure of human genes.

When a gene is turned on by
signals at the regulatory region, the
entire gene — introns and exons — is
copied. Then the introns are spliced
out to make the shorter mRNA that
is translated into protein by the cell.

3x10^ bp In the human genome
50.000 - 100.000 genes
30.000 base pairs per gene

A gene is 10% information (exons) and 90% stuffer (introns)

Portion of the

chromosome Exon Exon Exon Exon

containing a gene 4' ^ ^ ^

A noncoding region
including
regulatory region

M

leader

Messenger RNA
(mRNA)

Protein

t

Introns

t t

Introns Introns

\

Noncoding
region

Copying DNA and splicing out Introns

processed gene

taU

Translation

124

Mapping and Sequencing the Genome

To determine the order of billions of pairs of raw DNA sequence, first the human
genome must he broken down into genes or other fragments small enough to be
propagated by cloning techniques and characterized for identification. Next, these
chromosomal fragments will be ordered — mapped — to correspond to their respective
locations on the chromosomes. Finally, automated techniques will be utilized to
determine the base sequence of the ordered fragments — the ultimate goal of the Human
Genome Project.

Molecular Techniques Used in the Human Genome Project

Improvements in mapping and sequencing techniques are a major focus of the genome
project. The efforts will include developing automated probe and mapping methods, as
well as optimizing techniques to extract the maximum useful information from maps
and sequences. Some of the techniques currently being employed are described below.

Using Restriction Enzymes

Isolated from various bacteria, restriction enzymes serve as microscopic scalpels that
recognize short sequences of DNA and cut the DNA molecules at those specific
recognition sites (Fig. 5).

Examples
GClGC

cg:cg

GAA; TIC
CTT:AAG

Y
GCGG ; CCGC

CGCC : GGCG

No. of Bases
in Recognition

Site

Frequency of
Occurrence

1
256

1

4096

1

4 bases
6 bases

8 bases

64,000

Fig. 5. Typical restriction enzyme
cutting sites. The recognition site
has a twofold symmetry around a
point (the dashed line). The cutting
sites (arrows) are on both strands of
the double helix (above and below
the solid line) and. when cut,
generate smaller, linear fragments
of DNA.

125

Appendix A:
Primer on Molecular
Genetics

On the average:

• 4-ba.se recognition sites will yield pieces 256 bases long,

• 6-base recognition sites will yield pieces 4,000 bases long, and

• 8-base recognition sites will yield pieces 64.000 bases long.

Since hundreds of different restriction enzymes have been characterized. DNA can be
cut into many different types of smaller fragments.

Electrophoretic techniques (Fig. 6) can be used to .separate DNA fragments according
to length with a 1 -base-pair resolution up to 1.000 base pairs, with 1% resolution up to
I million base pairs, and with 5% resolution up to 10 million base pairs. This great
separation power forms the basis of most methods for handling and analyzing
DNA molecules.

Fig. 6. Schematic results of
electrophoresis of DNA. The

fragments resulting from restriction
enzyme cutting can be separated
by gel electrophoresis. The smallest
fragments move most rapidly
toward the positive electrode. DNA
fragments can be detected by uv
absorbance, fluorescence, radio-
activity, or other labels. The
electrophoresis pattern itself can be
of interest, since variations in the
pattern from a given chromosomal
region can sometimes be associ-
ated with variations in genetic traits,
including susceptibility to certain
diseases. This kind of "genetic
linkage" has been used to establish
the locations of important genes.

Intact DNA molecule (sizes in base pairs)

7,000

1,000

600

7.000-

1,500

-► 3.900-

3,900

Cut with
restriction
enzyme 1.500-

1.000-
600 ■

electrophoresis

(-^)

fragment sizes in number of bases

126

Cloning DNA Fragments

In vivo cloning. By propagating the fragments inside living cells. DNA pieces can be
immortalized, and the amount of specific DNA can be amplified (Fig. 7). Bacteria are
most often the hosts, but yeast and mammalian cells are also used. These cloning
procedures provide unlimited amounts of material for experimental study.

Cloning

1

X 10^ bp

cut

T

1

V

aurlfy

Grow in bacteria or in yeast
to immortalize

Libraries

Bacteriophage

■ cloned in bacteria

size:

Up to 5 X 10'* bp

coverage:

Need 10^ clones to cover a human genome

YACs

- cloned In yeast

size:

1 - 10 X 10^ bp

coverage:

Need 10 '^clones to cover a human genome

Fig. 7. Cloning and libraries. A

DNA fragment can be cloned in
bacteriophages or yeast artiticial
chromosomes (YACs). The
minimum number of separate
fragment clones needed in a library
to cover the human genome of
3x10^ bp is shown.

127

Appendix A:
Primer on Molecular
Genetics

In vitro cloning. A new approach to cloning, the polymerase chain reaction (PCR)
amplifies DNA molecules in the test tube without using a host cell (Fig. 8). PCR is
especially valuable because the reaction is easily automated and can handle samples too
small or too to.xic to be managed inside cells.

If a set of DNA fragments contains a complete sample of an entire genome or
chromosome, it is called a library (Figs. 7 and 9).

Fig. 8. DNA amplification by the
polymerase chain reaction (PCR).

PCR is a recently developed
method for cloning human DNA — a
necessary prelude to any intensive
mapping or sequencing efforl. PCR
begins with the annealing of primers
to the single-stranded fragments
that are to be replicated. With the
primers attached, a DNA replication
enzyme can complete the
complementary strands thus
started. The succeeding PCR cycle
then operates on the newly
produced duplicates, as well as the
original fragments. Each cycle thus
doubles the amount of the selected
DNA fragments in the reaction
mixture. Twenty-five cycles, which
can be completed in less than three
hours, can theoretically produce
amplification by 30 million-fold: in
practice, amplification by a factor of
over one hundred thousand can be
readily achieved.

,i 7 \t'<i>,:l, ilr<imli.

\r^7 Extend complementary strands
^-^ by emymatK rephcatton

^^4/ ffepedl steps I and 2

128

Mapping the Genome

The human genome may be "understood" at several levels (Fig. 10). Geneticists have
already charted the approximate positions of over 1000 genes, and a patchy start has
been made at establishing high-resolution maps of the genome. The coarsest maps are
chromosome and genetic maps, which define the chromosomal locations of genes. The
physical map is an ordered set of DNA fragments made from restriction enzyme
fragments. The ultimate map is the base pair sequence for the human genome.

i-^' Libnin
III (linud
tragments^ ,5,

Fig. 9. Linking a library of cloned
DNA fragments. Several strategies
are available for ordering an
unordered library of cloned restric-
tion fragments. The one schemati-
cally illustrated here involves
sequencing the ends of the
fragments, together with a corre-
sponding family of much shorter
linking clones. The linking clones,
themselves generated by restriction
enzymes, are fragments that
contain the cutting site (in this
simplified example. GGCC) for the
enzyme used to produce the clone
library. In practice, linking clones
are typically 50 to 100 base pairs
long, and the fragments being
linked are likely to be many
thousands of base pairs long. If the
library contains contiguous frag-
ments and if the family of linkers is
complete, assigning an order to the
fragments is a simple matter of
sequence matching — a puzzle-
solving task best done by computer.

129

Appendix A:
Primer on Molecular
Genetics

Fig. 10. Multiple levels of mapping of human chromosomes. At the lowest level of
resolution, the human genome is characterized by 22 pairs of autosomal chromosomes
and two sex chromosomes — X and Y. Individual chromosomes can be distinguished by
size, position of centromeric constriction relative to telomeres (centromeric index), and
patterns of banding induced by staining with DNA-specific dyes after treatments that
partially remove chromosomal proteins (G- and Q-banding). Cloned DNA probes can be
used to determine by in situ hybridization the locations of specific genes or markers (A
and B in this diagram) along a chromosome. To establish the order of — and relative
distances between — genes or markers along specific human chromosomes, polymorphic
DNA markers or fragments of gene regions are used to follow the inheritance of specific
alleles in multigeneration families with large numbers of siblings. Distances on genetic
linkage maps are given as percent recombination between markers: a distance of 1%
recombination [1 centiMorgan (cM)] is generally estimated to represent approximately
1 Mbp of DNA. Radiation hybrid techniques can be used to obtain a higher-resolution
(hundreds to thousands of kilobase pairs) physical map [S. Goss and H. Harris, Nature
255, 1445-1458 (1975) and D, Cox et al.. Genomics^, 397-407 (1988)]. The next level
of resolution can be represented by the analytical physical — or macrorestriction — map
that is developed using rare-cutting restriction enzymes and pulsed-field gel
electrophoresis to separate megabase-sized DNA fragments. Markers are located on
macrorestriction maps relative to restriction sites, and distances can be measured in
kilobase pairs.

CHROMOSOME ( Siim)
CYTOGENETIC MAP

CHROMOSOMAL DNA
{4-8 X 10''^nl)
(-1.3 X lo'bp)

LINKAGE OR
GENETIC MAP

RADIATION HYBRID
MAP

PHYSICAL MAPS

MACRORESTRICTION

MAP

OVERLAPPING SET
OF CLONES

SEQUENCE TAGGED SITES
(STS)

DNA SEQUENCE

c

-mK

*

f

^XSOOO EXPANSION MARKER* (GENES)\

Disease Locus

RECOMBINATION
FREQUENCY

11 It

DIseas i Gene

PFG MAP (20 kt)-10 Mb)

YACs (100 - 1000 kb)

COSMIDS (40 kb)
PHAGE (17 kb)

■ m iMi m m

^^p<Z><!^^^ ,1 BASE PAIR,

DISEASE GENE SEQUENCE

As an example of the value of long-
range restriction maps, distances
can be determined between markers
flanking a disease locus. Over-
lapping sets of cloned DNA from
phage or cosmid vectors and from
yeast chromosomes — all engineered
to accept foreign DNA — can be used
for constructing these macro-
restriction maps. A clone subset
provides the substrate for obtaining
the DNA sequence — the ultimate
physical map. Maps from the
cytological level to the DNA
sequence are retained in various
databases. The proposal regarding
sequence-tagged sites (STSs) has
offered that evenly spaced STSs
would be used as identifiers to
interrelate different levels of maps
and to diminish requirements for
storage of large numbers of
biological reference materials
[M. Olson et al., Science 245,
1434-1435 (1989)]. (Figure orovided
by C. E. Hildebrand, Los Alamos
National Laboratory.)

130

Chromosome maps

In a chromosome map, genes or DNA fragments are assigned to their respective
chromosomes (Fig. 10). With in situ hybridization, the DNA (tagged with either a
fluorescent or radioactive label) is used to seek out and bind to its complementary
strand in an intact metaphase chromosome.

Until recently, the best chromosome maps could be used to locate a DNA fragment only
to within a region of about 10 million base pairs — the size of a typical chromosome
band. However, new methods promise to improve the resolution of the maps and allow
narrowing of the range to around 1 million base pairs.

Genetic maps

Any molecular or physical characteristic that differs between individuals and is
inherited is a potential genetic marker. The human genetic map is constructed by
observing the pattern of inheritance of pairs of genetic markers. DNA sequence
differences are especially useful markers because they are plentiful and easy to
characterize precisely. Two markers located near each other on the same chromosome
will tend to be passed together from parent to child (Fig. 1 1 ).

B

father
blue eyes
HD H

mother

H

—

B —

-

E —
H _

—

children

blue eyes
HD

blue eyes blue eyes
only HD

ombinant: f

requen

cw of this event measures

distance between genes for blue eyes and HD

Fig. 11. How genetic maps are
made. Each parent contributes one
of each type of chromosome
through the haploid egg and sperm
so that the offspring will have a
diploid chromosome set. In this
illustration, the vertical lines show
just one pair of chromosomes for
each individual in a family. The
father has, in this example, two
traits that will be visible in any child
that inherits them: blue eyes (B)
and Huntington's disease (HD). The
fact that one child received only a
single trait, B, indicates that one of
the father's chromosomes
rearranged in the process of
producing the child. (Note: In this
example the mother's eyes were
also blue.)

131

Appendix A:
Primer on Molecular
Genetics

Meiotic recombination, a relatively rare event, produces DNA strand breakage and
rejoining; the result is separation of two markers originally on the same chromosome.
The frequency of occurrence of these rare events provides an estimate of the distance
between two markers. The closer they are. the less likely that a recombination event
will fall between them and separate them.

The power of the genetic map is that an inherited disease can be located on the map,
even though the molecular basis of the disease is not yet understood and the gene for
the disea.se is not yet identified. Thus, genetic maps can be used to locate the
neighborhood of important disease genes. The current resolution of the human genetic
map is 10 million base pairs, about the same as the chromosome map.

Physical maps

Physical maps are ordered sets of DNA fragments made from restriction enzyme
fragments. This mapping is done by either the top-down or the bottom-up approach
(Fig. 12). Both of these approaches require purifying and analyzing large numbers of
DNA fragments.

Top-down mapping. In top-down mapping, a single chromosome is cut into large
pieces, which are then ordered, subdivided, and mapped further. DNA pieces can be
located in regions about 100.000 to I million base pairs in size.

Bottom-up mapping. In bottom-up mapping, the chromosome is cut into small pieces.
These pieces — the resulting clones — are then placed in order according to structural

Fig. 12. Two physical mapping
strategies. The top-down strategy
gives low resolution but high
connectivity. The bottom-up
strategy has low connectivity
because not enough clones can be
mapped to give neighboring
contigs.

Top-Down Strategy

chromosome

fragments

single fragment
digest

Results: Macro-restriction map: complete but low resolution
Bottom-Up Strategy
fragments ^^^^^^^^^^>» ^^^^^^— ^^^—

contigs

z"/"z /-\ -r\

YAC or cosmld clones
Results: Map based linked library: detailed but incomplete

132

features, such as restriction map data, that have two. and then multiple, clones in
common. TTiey then form contiguous blocks of DNA (contigs). The resulting ordered
library of DNA pieces is 10.000 to 100.000 base pairs in size.

By using chromosomes purified by flow-sorters (a technique pioneered by the national
laboratories) or in hybrid cell lines, one can concentrate on mapping a single
chromosome at a time.

Sequencing

The DNA sequence is the ultimate physical map. Sequencing is also done by two basic
approaches. Both of these methods work because very high resolution separations of
DNA molecules are achievable with gel electrophoresis. In Maxam-Gilbert sequencing
(Fig. 13A). DNA is cleaved at individual specific bases, and the lengths of the resulting
fragments are determined. In Sanger sequencing (Fig. 13B), DNA replication is stopped
at one of the four types of bases, and the lengths of the resulting DNA fragments are
determined. Virtually all the steps in these sequencing methods are now automated.

Double Stranded
DNA ot Unknown

Chemical Trealmml

loCul Chain al One or
Two Specific
NocleoUdes

-

Reaction Mixtures

Get Electrophoresis

Detection of

Radioactive Bands

Only (Autoradiography!

G + A T - C

»1-

•-

-rn-,- ^-r

"a

C T T

Prod

ucts from
Reaction

DNA Polymerase I ^-^ .,, .~.

dCTP ; . >■ V

dCTP

dTTP ; KH Hy

H H
Dldroxynucleoilde (ddNTP any one ol Ihe 4 bases)

dJTTI- .JdCTT

■■'"'^•te.,.

RcacUon Mixtures

Ge] EtectrophoTCsIs

Auloradiof^phy lo

Dclecl Radioartlvr

Etands

I I ' I ' i I ' Ij

Fig. 13A. DNA sequencing by the Maxam-Gilbert method

{Office of Technology Assessment, 1988).

Fig. 13B. DNA sequencing by the Sanger method. (Office of
Technology Assessment, 1988).

133

Appendix A:
Primer on Molecular
Genetics

End Games: Completing the Maps and Determining
the Sequences

Starting maps and sequences is relatively easy; finishing them is very difficuU. New
methods are needed to expedite this end game.

Chromosome Walking with Primers

An approach being used to finish sequences is chromosome walking with primers
(Fig. 14). In this technique, the walk begins with a primer at a known site and proceeds
in a linear fashion — one step at a time — while adjacent regions that were previously
unknown are being identified and sequenced. The larger region then serves as the new
primer. All primers are synthesized chemically. The difficulty with this technique is the
number of syntheses required for a large project.

Fig. 14. Chromosome walking.

chromosome
- chromosome

rhromosome

1 1 known primer

synthesis of DNA

T stop 1

sequence synthesize

d DNA
new known sequence

new known primer

1 1

V

V//////A

+

1

synthesis

chromosome

\/77/7//y\—

r//////XA-

One "step"

134

Single Chromosome Dissection

Another approach to finishing maps is to utilize single-chromosome dissection by
physical methods (Fig. 15). This method isolates DNA pieces from the regions that are
not yet mapped or sequenced and then applies the previously described mapping
methods to those pieces.

Locating Specific Genes

The current human genetic map has about 1000 markers, or 1 marker spaced every 3
million base pairs. About 100 genes will lie between each pair of markers. In some
regions of particular interest, genetic maps have been made that are five to ten times
more detailed. By combining genetic and physical map infomiation for a region, the
stage is set for locating new genes (Fig. 16). The genetic map of these markers basically

unibld, cut

extend

DNA chain

tether

tether

chromosome ifr molecule

Fig. 15. Chromosome or single-
molecule strategies.

135

Appendix A:
Primer on Molecular
Genetics

gives gene order. Rough information about gene position is sometimes available also,
but these data have to be used with caution, because recombination is not equally likely
at all places on the chromosome: thus, the genetic map, compared to the physical map,
stretches in some places and compresses in others. It is as though the genetic map were
drawn on a rubber band.

How difficult it is to find an actual disease gene of interest depends largely on what is
already known about that gene and, especially, on what sort of alterations in the gene
have resulted in disease (Fig. 17). If disease results from a single altered DNA base,
spotting the disease gene is very difficult; sickle cell anemia is an example of such a
case, as are probably the majority of major human inherited diseases. When disease

Fig. 16. Finding genes. A genetic
map will reveal that a given gene of
interest (C) lies in a region, perhaps
encompassing 10 million base pairs
(10 Mbp). The physical map allows
one to dissect that region and to
locate likely pieces on which the
particular gene may reside.

Genetic
map

cloned
DNA

A

1 - 10 Mb

— D

c

■O

B

restriction
fragments

1 Mb

Goal; Find the fragment that contains gene C.

136

results from a large DNA rearrangement, this anomaly can usually be detected by
alterations in the physical map of the region or even by examination of the
chromosome. The location of these alterations pinpoints the site of the gene.

To identify — without a map — the gene responsible for a specific disease, is analogous
to finding a needle in a haystack. Finding the gene is even more difficult, because no
matter how close one gets, the gene still looks like just another piece of hay. However,
maps make finding genes much easier. They tell where to look in the haystack. The
finer the map, the fewer the pieces of hay one has to test to see which is the gene of
interest.

normal gene
30,000 bp

single base
change

deletion

insertion

translocation

Fig. 17. Possible DNA
abnormalities that can
produce an inherited defect.

137

"The Alta Summit, December 1984"

Appendix B

139

Appendix C:
"The Alta Summit,
December 1984"*

Robert Miillan
Cook-Deegan

Alta is a ski area nestled among the Saguache Mountains in Utah, a winding 40-
niinute drive southeast from Salt Lake City. From December 9 to 13, 1984,
visitors were isolated by repeated blizzards. The slopes were covered most
mornings with Utah's renowned fine light powder, which beckoned skiers to cut its
virgin surface.

For those 3 days. Alta was also a capital of human genetics. Many historical threads in
the fabric that later became the Human Genome Project wind through that meeting,
although it was not a meeting on mapping or sequencing the human genome. Through
happenstance and historical accident, Alta links human genome projects to research on
the effects of the atomic bombs dropped on Hiroshima and Nagasaki 40 years earlier. If
genome projects prove important to biology, then historians will note the Alta meeting.

The Alta meeting was sponsored by the Department of Energy (DOE) and the
International Commission for Protection Against Environmental Mutagens and
Carcinogens. It was initiated by David Smith of DOE and Mortimer Mendelsohn of the
Lawrence Livermore National Laboratory, who turned over final organization to
Raymond White of the Howard Hughes Medical Institute at the University of Utah.

The purpose was to ask those working on the front lines of DNA analytical methods to
address a specific technical question: could new methods permit direct detection of
mutations, and more specifically could any increase in the mutation rate among
survivors of the Hiroshima and Nagasaki bombings be detected (in them or in their
children)? The idea behind the Alta meeting came from another meeting on March 4

*Reprinted with permission from
Academic Press. Inc.. Genomics 5.
661-663 (October 1989),

TABLE 1

Participants at ttie Alta l\/leeting, December 1984

David Botstein

Mortimer Mendelsohn

Elbert Branscomb

John Mulvihill

Charles R. Cantor

Richard Myers

C. Thomas Caskey

James V. Neel

George Church

Maynard Olson

John D. Detahanty

David A, Smith

Charles Edington

Edwin Southern

Raymond Gesteland

Sherman Weissman

Michael Gough

Raymond L, White

Leonard Lerman

140

and 5, 1984. in Hiroshima, at which new DNA analytical tools were deemed second
highest priority for human mutations research, just behind establishing ceil lines from
atomic bomb survivors, their progeny, and controls. Those attending the Alta meeting in
December (see Table 1 ) were drawn from a variety of backgrounds, and many had
never met each other. Most said in interviews later that they came to the meeting quite
skeptical, but left thinking it had been one of the best scientific meetings they ever
attended (Interviews, 1987. 1988).

The principal conclusion of the meeting was. ironically, that methods were incapable of
measuring mutations with sufficient sensitivity, unless an enormously large, complex,
and expensive program were undertaken. Technical obstacles thus thwarted attainment
of the main goal of the meeting, yet the meeting left a profusion of new ideas in its
wake, some of which later washed ashore to be incorporated into various genome
projects. Five years later, there is still no sensitive assay for human heritable mutations,
but there are genome programs at NIH. at DOE, and in several foreign nations.

Excitement about the new methods blossomed at Alta despite, or perhaps because of,
the wintry isolation. As Mortimer Mendelsohn noted in his internal report to DOE:

It was clear from the outset that the ingredients for a successful meeting [were
present]. . . and the result far exceeded expectation. Once the point of the
exerci.se was clear to everyone, a remarkable atmosphere of cooperation and
mutual creativity pervaded the meeting. Excitement was infectious and ideas
flowed rapidly from every direction, with many ideas surviving to the end.
(Mendelsohn, 1985).

John Mulvihill began the meeting by reviewing epidemiological studies of human
mutations. Studies that could theoretically have detected a threefold increase in
mutations had not found any. James Neel spoke about measurement of mutations among
Hiroshima-Nagasaki survivors, estimating that the likely mutation rate was 10 '^ per
base pair per generation (or roughly 30 new mutations per genome per generation),
indistinguishable from that of Japanese controls and in the same general range as that
estimated by epidemiological methods and detection of protein variants among other
"normal" populations. Several of the technical consultants commented on the passionate
devotion Neel brought to the study of Hiroshima and Nagasaki victims, and how his
demeanor set the tone for lively and cooperative exchanges throughout the meeting.

Existing methods had failed to detect an anticipated increase in mutations among the
more than 12.000 children of Hirsohima-Nagasaki survivors (whose parents received an
average 43 rad). Calculations showed that to measure a 30% increase in the mutation
rate, roughly what would be expected from the average dose, one would have to
examine 4.5 x 10'" bp in the children, and 4 to 5 times more in the parents (Delahanty,
1986). In fact, the DNA methods were at least an order of magnitude short of being able
to detect the expected impact from atomic bomb exposure among survivors; they could
only detect differences expected from radiation exposure well above the lethal dose
(and hence not measurable). The question was whether there were new technical means

141

Appendix B:
"The Alta Surnmir

that would get around the problems. The answer was no. but the process of thinking
about it forced many novel ideas to the surface.

George Church began to ruminate on the ideas that culminated in multiplex sequencing.
He said later that discussions with Maynard Olson. Richard Myers, and others helped
him crystallize his inchoate ideas. (David Smith recalled watching George Church
disappear in a cloud of new-fallen powder one afternoon, and worrying about the future
of DNA sequencing technology.)

Richard Myers showed work using RNase I to cut (and thus make detectable) single
base pair mismatches; he and Leonard Lerman showed early data using gradients of
denaturing agents embedded in electrophoresis gels as a way to detect heteroduplexes
and mismatches. Myers credits his roommate for the conference. Maynard Olson, with
clarifying his ideas and permitting him to expand the RNase I method to mismatches
other than C-A mutations. In a trip report to the Office of Technology Assessment.
Michael Gough characterized the Church and Myers presentations as technological
wonders and called the two young scientists, then largely unknown, the "two biggest
surprises" of the meeting (Gough. 1984).

Charles Cantor showed how his and David Schwartz's first pulsed-field gel electro-
phoresis method could separate megabase-sized DNA fragments, resolving individual
yeast chromosomes and thus introducing an enormously powerful method to assess
DNA structure on this scale. He also showed his and Cassandra Smith's first macro-
restriction digest of the Escherichia coli genome, which suggested the tantalizing
possibility of physically mapping entire genomes by combining restriction cleavage and
pulsed-field gel electrophoresis.

Maynard Olson showed early results of attempting to construct a physical map of
Saccharofuyccs ccrcvisiac using overlapping clones, and also showed good separation
of megabase-sized DNA using a modification of the Schwartz-Cantor electrophoresis
technique. Mendelsohn's DOE report noted that "while Olson's method would not
presently be chosen for analyzing human mutation rates, his philosophy of paying
careful attention to and investing in the quantitative, methodological details of DNA
technology had a recurrent and important impact on the meeting" (Mendelsohn, 1985).
Olson later brought the same core ideas to the National Research Council Committee
on Mapping and Sequencing the Human Genome, where those ideas, combined with
an expansion of goals to include genetic mapping, helped to forge a consensus that
dedicated genome projects were scientifically worthwhile (National Research
Council, 1988).

At Alta, Elbert Branscomb described the state of the art in using flow cytometry and
immunofluorescence to detect altered protein products on the surface of red cells.
Branscomb later became the computer modeler and one of the architects for the

142

Livemiore cosmid map of chromosome 19. now under construction. Tom Caskey
reviewed progress on understanding mutations in the HPRT locus, and Sherman
Weissman reviewed data on the HLA locus. David Botstein. as always exuding volcanic
enthusiasm peppered with sharp humor, speculated about pushing the restriction
fragment length polymorphism (RFLP) techniques to their limits — perhaps enough to
detect mutations in the range of 10^ per base pair per generation. Unfortunately, this
was still shy of what would be needed to detect mutations among the Hiroshima-
Nagasaki survivors, unless an unrealistically massive effort were mounted. Ray White
talked about applying RFLP methods to the Y chromosomes originating from a single
Mormon progenitor of 1850 (who by now has thousands of male descendants) to
examine changes in the part of the Y chromosome outside the pseudoautosomal
region — a part of the genome where changes should accumulate.

Edwin Southern wound up the .scientific session by addressing the gap between
cytogenetic detection and molecular methods, and his presence was noted by more than
one participant as a moderating influence on the intellectual pyrotechnics. Southern's
discussion of measuring uv-induced mutations might be seen to presage the radiation
hybrid mapping methods brought to fruition in 1988 by David Cox and Richard Myers,
although the two approaches are quite independent in origin.

Michael Gough returned from Alta to Washington to work on the OTA report on
detecting heritable mutations. The report had been requested by Congress in anti-
cipation that controversies over Agent Orange, radiation exposure during atmospheric
testing in the 1950s, and exposure to mutagenic chemicals might find their way to court,
where a neutral as.sessment of the technical feasibility of detecting mutations would be
essential. Gough directed preparation of Technologies for Detecting Heritable
Mutations in Human Beings until he left OTA in 1985 (U.S. Congress, 1986). Several
Alta participants served either as contractors or as advisory panel members for that
study. Charles DeLisi, then newly appointed director of the Office of Health and
Environmental Research at DOE, read a draft of this report in October 1985, and while
reading it first had the idea for a dedicated human genome project (DeLisi, 1988). The
Alta meeting is thus the bridge from DOE"s traditional interest in detection of mutations
to DeLisi 's push for a Human Genome Initiative, and provides one of several historical
links between genome projects and another massive technical undertaking of the 20th
century — the Manhattan project.

Acknowledgements

Thanks go to the many Alta participants and others who reviewed drafts of this
historical sketch — Elbert Branscomb, Charies Cantor, Charles DeLisi, Michael Gough,
Mortimer Mendelsohn, Richard Myers, Maynard Olson, David Smith, and Ray White —
and to those who helped provide background in interviews (see Ref. (4)).

143

Appendix B:

"The Alta Summit"*

References

1. DELAHANTY. J.. WHITE. R. L., AND MENDELSOHN, M. L. ( 1986).
Approaches to determining mutation rates in human DNA. h4iitai. Res 167:
215-232.

2. Di LISI, C. (1988). The Human Genome Project. Anicr. Sci. 76: 488-493.

3. GOUGH. M. (1984). Notes from the DOE's Utah Meeting on DNA Methods for
Measuring Human Mutation Rates. Trip Report to the Office of Technology
Assessment. 21 December 1984.

4. Interviews with David Bolstein. 22 August 1988; Charles Cantor. 19 August 1988;
George Church, 14 November 1988; Charles DeLisi, 6 January 1987 and 7 October
1988; Maynard Olson. 28-30 April 1988; David Schwartz, 6 January 1987; and
David Smith. 22 December 1988.

5. MENDELSOHN, M. L. (1985). Infomial Report of a Meeting on DNA Methods for
Measuring the Human Heritable Mutation Rate, Lawrence Livermore National
Laboratory Report UCID-20315. January 1985.

6. National Research Council, Committee on Mapping and Sequencing the Human
Genome ( 1988). "Mapping and Sequencing the Human Genome," National
Academy Press, Washington, DC.

7. U.S. Congress, Office of Technology Assessment (1986). "Technologies for
Detecting Heritable Mutations in Human Beings," OTA-H-298, U.S. Govt. Printing
Office, Washington, DC.

*Reprinted with permission from
Academic Press, Inc.. Genomics 5,
661-663 (October 1989).

144

Glossary

Appendix C

145

Appendix C:
Glossary

Portions of the glossary text were
taken directly or modified from
definitions in the U.S. Congress
Office of Technology Assessment
document; Mapping Our Genes —
The Genome Projects: How Big.
How Fast? OTA-BA-373.
Washington, D.C.: U.S.
Government Printing Office.
April 1988.

A word printed in a typeface different from that of the definition text is defined
within the glossary.

Adenine (A): A nitrogenous base, one member of the base pair. A-T (adenine-
thymine).

Alleles: Alternative forms of a genetic locus; a single allele for each locus is inherited
separately from each parent (e.g., at a locus for eye color, a certain allele might
result in brown eyes).

Amino acid: Any of a class of 20 molecules that are combined to form proteins in
living things. The sequence of amino acids in a protein and hence protein function
are determined by the genetic code.

Arrayed library: Arrayed libraries represent individual primary recombinant clones
(hosted in phage, COSmid, YAC, or other vector) that are placed in two-
dimensional arrays m microtiter dishes. Each primary clone can be identified based
on the identity of the plate and the clone location (row and column) on that plate.
Arrayed libraries of clones can be used for many applications including screening
for a specific gene or genomic region of interest as well as for physical mapping.
Information gathered for individual clones from various genetic linkage and
physical map analyses is entered into a relational database and used to construct
physical and genetic linkage maps simultaneously; clone identifiers serve to
interrelate the multilevel maps. Compare libraiy. genomic library.

Autoradiography: A technique that uses X-ray film to visualize radioactively labeled
molecules or fragments of molecules; used in analyzing length and number of DNA
fragments after they are separated by gel electrophoresis.

Autosome: A chromosome not involved in sex determination. The diploid human
genome consists of 46 chromosomes, 22 pairs of au'osomes. and 1 pair of sex
chromosomes (the x and y chromosomes).

Bacteriophage: See phage.

Base pair (bp): Two nitrogenous bases (adenine and thymine or guanine and

cytOSine) held together by weak bonds. Two strands of DNA are held together in
the shape of a double helix by the bonds between base pairs.

Base sequence analysis: A method, sometimes automated, for determining the
sequence of nucleotide bases in DNA.

Blotting: See in situ colony hybridization. Southern blotting.

Blunt ends: On linear duplex DNA molecules, ends that are fully double-stranded and
base-paired, without single-stranded tails.

146

Centimorgan (cM): A unit of measure of recombination frequency. One centimorgan
is equal to a 1 -percent chance that a marker at one genetic locus will be separated
from a marker at a second locus due to crossing over in a single generation. In
human beings. 1 centimorgan is equivalent, on average, to 1 million base pairs.

Centromere: A specialized chromosome region to which spindle fibers attach during
cell division.

Chromosomes: The autoreplicating genetic structures of ceils, containing the cellular
DNA that bears in its nucleotide sequence the linear array of genes. In
prokaryotes. chromosomal DNA is circular, and the entire genome is carried on
one chromosome. Eukaryotic genomes are divided between a number of
chromosomes in which the DNA is associated with many different kinds of
proteins.

Clone bank: See Genomic library.

Cloning: The process of asexually producing a group of cells (clones), all genetically
identical, from a single ancestor. In recombinant DNA technology, the u.se of
DNA manipulation procedures to produce multiple copies of a single gene or
segment of DNA is referred to as cloning DNA.

Cloning vector: DNA molecule originating from a virus, a plasmid, or the cell of a
higher organism into which another DNA fragment of appropriate size can be
integrated without loss of the vector's capacity for self-replication; vectors
introduce foreign DNA into host cells, where it can be reproduced in large
quantities. Examples are plasmids, cosmids, and yeast artifical chromosomes;
vectors are often recombinant molecules containing DNA sequences from several
sources.

Code: See genetic code.

Cohesive (sticky) ends: On linear duplex DNA molecules, single-stranded ends that are
complementary and can base-pair either with each other to form a circular
molecule or with other linear DNAs having the same termini to form recombinant
DNA molecules.

Colony hybridization: See in situ colony hybridization.

Complementary DNA (cDNA): DNA that is synthesized from a messenger RNA
template; the single-strand form is often used as a probe in physical mapping.

Complementary sequences: Nucleic acid sequences that can form a double-stranded
structure by formation of base pairs; the sequence complementary to G-T-A-C is
C-A-T-G.

147

Appendix C:
Glossary

Contigs: Groups ot clones representing overlapping regions of a genome.

Cosmid: Artificially constructed cloning vector containing the ens gene of phage
lambda. Cosmids can be packaged in lambda phage particles for infection into
E. (v>//; this permits cloning of larger DNA fragments (up to 45 kb) than can be
introduced mio bacterial hosts in plasmid vectors.

Crossing over: The breaking during meiosis of one maternal and one paternal

chromosome, the exchange of corresponding sections of DNA. and the rejoining
of the chromosomes. This process can result in an e.xchange of alleles between
chromosomes.

Cytosine (C): A nitrogenous base, one member of the base pair, G-C (guanine and
cytosine).

C-value paradox: The lack of correlation between the amount of DNA in a haploid
genome and the biological complexity of the organism. (C-value refers to haploid
genome size.)

Determinism: The theory that for every action taken there are causal mechanisms such
that no other action was possible.

Diploid: A full set of genetic material, consisting of paired chromosomes — one

chromosome from each parental set. Most animal cells except the gametes have a
diploid set of chromosomes. The diploid human genome has 46 chromosomes.
Compare haploid.

DNA, deoxyribonucleic acid: The molecule that encodes genetic information. DNA is
a double-stranded molecule held together by weak bonds between base pairs of
nucleotides. The four nucleotides in DNA contain the bases: adenine (A), guanine
(G), cytosine (C). and thymine (T). In nature, base pairs form only between A and
T and between G and C: thus the sequence of each single strand can be deduced
from that of its partner.

DNA probes: See probes.

DNA replication: The use of existing DNA as a template for the synthesis of new DNA
strands. In humans and other eukaryotes, replication occurs in the cell nucleus.

DNA sequence: The relative order of base pairs, whether in a fragment of DNA. a
gene, a chromosome, or an entire genome. See base sei/uc?ice analysis.

Domain: A discrete portion of a protein with its own function. The combination of
domains in a single protein determines its overall function.

Double helix: The shape that two linear strands of DNA assume when bonded together.

148

Electrophoresis: A method of separating large molecules (such as DNA fragments or
proteins) from a mixture of similar molecules. An electric current is passed through
a medium containing the mixture, and each kind of molecule travels through the
medium at a different rate, depending on its electrical charge and size. Separation is
based on these differences. Agarose and acryiamide gels are the media commonly
used for electrophoresis of proteins and nucleic acids.

Endonuclease: An enzyme that cleaves its nucleic acid substrate at internal sites in the
nucleotide sequence.

Enzyme: A protein that acts as a catalyst, speeding the rate at which a biochemical
reaction proceeds but not altering the direction or nature of the reaction.

Eukaryote: Cell or organism with membrane-bound, structurally discrete nucleus and
other well-developed subcellular compartments. Eukaryotes include all organisms
except viruses, bacteria, and blue-green algae. Compare prokciryote. See
chromosome.

Exons: The protein-coding DNA sequences of a gene. Compare introns.

Exonuclease: An enzyme that cleaves nucleotides sequentially from free ends of a
linear nucleic acid substrate.

Flow cytometry: Analysis of biological material by detection of the light-absorbing or
fluorescing properties of ceils or subcellular fractions (i.e.. chromosomes) passing
in a narrow stream through a laser beam. An absorbance or fluorescence profile of
the sample is produced. Automated sorting devices, used to fractionate samples,
sort successive droplets of the analyzed stream into different fractions depending
on the fluorescence emitted by each droplet.

Flow karyotyping: Use of flow cytometry to analyze and/or separate chromosomes
on the basis of their DNA content.

Gamete: Mature male or female reproductive cell with a haploid set of chromosomes
(23 for humans); that is, a sperm or ovum.

Gene: The fundamental physical and functional unit of heredity. A gene is an ordered
sequence of nucleotides located in a particular position on a particular
chromosome that encodes a specific functional product (i.e.. a protein or RNA
molecule). See gene expression.

Gene expression: The process by which a gene"s coded infomiation is converted into
the structures present and operating in the cell. Expressed genes include those that
are transcribed into mRNA and then translated into protein and those that are
transcribed into RNA but not translated into protein (e.g., transfer and ribosomal
RNAs).

149

Appendix C:
Glossary

Gene families: Groups of closely related genes that make similar products.

Gene library: See genomic lihrury.

Gene mapping: Detemiination of the relative positions of genes on a DNA molecule
(chromosome or plasmid) and of the distance, in linkage units or physical units,
between them.

Gene product: The biochemical material, either RNA or protein, resulting from

expression of a gene. The amount of gene product is used to measure how active a
gene is; abnomial amounts can be correlated with disease-causing alleles.

Genetic code: The sequence of nucleotides, coded in triplets along the mRNA, that
determines the sequence of amino acids in protein synthesis. The DNA sequence of
a gene can be used to predict the mRNA sequence, and the genetic code can in turn
be used to predict the amino acid sequence.

(ienetic engineering technologies: See reconihinant DNA technologies.

Genetics: The study of the patterns of inheritance of specific traits.

Genome: All the genetic material in the chromosomes of a particular organism: its
size is generally given as its total number of base pairs.

Genome projects: Research and technology development efforts aimed at mapping and
sequencing some or all of the genome of human beings and other organisms.

Genomic library: A collection of clones made from a set of randomly generated
overlapping DNA fragments representing ti'e entire genome of an organism.
Compare lihrary. uridyecl library.

Guanine (G): A nitrogenous base, one member of the base pair, G-C (guanine and
cytosine).

Haploid: A single set of chromosomes (half the full set of genetic material), present
in the egg and spemT cells of animals and in the egg and pollen cells of plants.
Human beings have 23 chromosomes in their reproductive cells. Compare diploid.

Heteroduplex: A double-stranded DNA molecule in which the two strands are not
completely complementary in base sequence and hence are not completely
base-paired,

Homeo box: A short stretch of nucleotides whose sequence is virtually identical in all
the genes that contain it. It has been found in many organisms, from fruit flies to
human beings. In the fruitfly, a homeo box appears to determine when particular
groups of genes are expressed during development.

150

Human gene therapy: Insertion of nomial DNA directly into cells to correct a genetic
detect.

Human Genome Initiative: Collective name for several projects begun in 1986 by
DOE to ( 1 ) create an ordered set of DNA segments from known chromosomal
locations. (2) develop new computational methods for analyzing genetic map and
DNA sequence data, and (3) develop new techniques and instruments for detecting
and analyzing DNA. This initiative is now known as the Human Genome Program.

Hybridization: The process of joining two complementary strands of DNA. or of
DNA and RNA together, to form a double-stranded molecule.

In situ colony hybridization: Use of a DNA or RNA probe to detect by in situ

hybridization the presence of the complementary DNA sequence in cloned bacterial
or cultured eukaryotic cells.

Informatics: The study of the application of computer and statistical techniques to the
management of information. In genome projects, informatics includes the devel-
opment of methods to search databases quickly, to analyze DNA sequence
information, and to predict protein sequence and structure from DNA sequence
data.

International technology transfer: Movement of inventions and technical know-how
across national borders.

Introns: The DNA sequences interrupting the protein-coding sequences of a gene;
these sequences are transcribed into RNA but are cut out of the message before it
is translated into protein. Compare e.xons.

Karyotype: A photomicrograph of an individual's chromosomes arranged in a
standard format showing the number, size, and shape of each chromosome type;
used in low-resolution physical mapping to correlate gross chromosomal
abnormalities with the characteristics of specific diseases.

Kilobase (kb): Unit of length for DNA fragments on physical maps (equal to the
distance spanned by 1000 base pairs).

Library: An unordered collection of clones (i.e., cloned DNA from a particular
organism), whose relationship can be established by physical mapping. Compare
genomic library, arrayed library.

Linkage: The proximity of two or more markers (e.g., genes, RFLP markers) on a
chromosome; the closer together the markers are, the lower the probability that
they will be separated during DNA repair or replication processes (binary fission in
prokaryotes, mitosis or meiosis in eukaryotes), and hence the greater the probability
that they will be inherited together.

151

Appendix C:
Glossary

Linkage map: A map of the relative positions of genetic loci on a chromosome,
determined on the basis of how often the loci are inherited together. Distance is
measured in centimorgans.

Locus (plural: loci): The position on a chromosome of a gene or other chromosome
marker; also, the DNA at that position. Some restrict use of locus to regions of
DNA that are expressed. See iicne expiTssion.

Mapping: See liene nuippiiii;. Ilnkas^c map. physical map.

Marker: An identifiable physical location on a chromosome (e.g.. restriction

enzyme cutting site, gene) whose inheritance can be monitored. Markers can be
expressed regions of DNA (genes) or some segment of DNA with no known coding
function but whose pattern of inheritance can be determined. See RFLP, restriction
frai>mi'ni length polymorphism.

Meiosis: The process of two consecutive cell divisions in the diploid progenitors of sex
cells. Meiosis results in four rather than two daughter cells, each with a haplold set
of chromosomes.

Messenger RNA, mRNA: RNA that serves as a template for protein synthesis. See
genetic code.

Multifactorial or multigenic disorders: See polygenic disorders.

Mutation: Any heritable change in DNA sequence. Compare polymorphism.

Nucleotide: A subunit of DNA or RNA consisting of a nitrogenous base (adenine,
guanine, thymine, or cytosine in DNA: adenine, guanine, uracil, or cytosine in
RNA), a phosphate molecule, and a sugar molecule (deoxyribose in DNA and
ribose in RNA). Thousands of nucleotides are linked to form a DNA or RNA
molecule. See DNA. base pair, RNA.

Nucleus: The cellular organelle in eukaryotes that contains the genetic material.

Oncogene: A gene, one or more forms of which is associated with cancer. Many

oncogenes are involved, directly or indirectly, in controlling the rate of cell growth.

Phage: A virus for which the natural host is a bacterial cell.

Physical map: A map of the locations of identifiable landmarks on DNA (e.g.,

restriction enzyme cutting sites, genes), regardless of inheritance. Distance is
measured in base pairs. For the human genome, the lowest-resolution physical
map is the banding pattems on the 24 different chromosomes: the highest-
resolution map would be the complete nucleotide sequence of the chromosomes.

152

Plasmid: Autonomously replicating, extrachromosomal circular DNA molecules.

distinct from the normal bacterial genome and nonessential for cell survival under
nonselective conditions. Some plasmids are capable of integrating into the host
genome. A number of artificially constructed plasmids are used as cloning vectors.

Polygenic disorders: Genetic disorders resulting from the combined action of alleles
of more than one gene (e.g., heart disease, diabetes, and some cancers). Although
such disorders are inherited, they depend on the simultaneous presence of several
alleles, thus the hereditary patterns are usually more complex than those of
single-gene disorders. Compare sini;le-f>eiu' disorders.

Polymerase, DNA or RNA: Enzymes that catalyze the synthesis of nucleic acids on
preexisting nucleic acid templates, assembling RNA from ribonucleotides or DNA
from deoxyribonucleotides.

Poly.nerase chain reaction (PCR): A method for amplifying a DNA sequence using
the Klenovv fragment of £. coli DNA polymerase I and two 20-base primers, one
complementary to the (+)-strand at one end of the sequence to be amplified and
the other complementary to the (-)-strand at the other end. Because the newly
synthesized DNA strands can subsequently serve as additional templates for the
same primer sequences, successive rounds of primer annealing, strand elongation
and dissociation produce rapid and highly specific amplification of the desired
sequence. PCR also can be used to detect the existence of the defined sequence in a
DNA sample.

Polymorphism: Difference in DNA sequence among individuals. Genetic variations
occurring in more than I percent of a population would be considered useful
polymorphisms for genetic linkage analysis. Compare nuitation.

Primer: Short preexisting polynucleotide chain to which new deoxyribonucleotides can
be added by DNA polymerase.

Probe: Single-stranded DNA or RNA molecules of specific sequence, labeled either
radioactively or immunologically, that are used to detect the complementary base
sequence by hybridization.

Prokaryote: Cell or organism lacking membrane-bound, structurally discrete nucleus
and subcellular compartments. Bacteria are prokaryotes. Compare eiikaryote. See
chromosome.

Promoter: A site on DNA to which RNA polymerase will bind and initiate
transcription.

Protein: A large molecule composed of one or more chains of amino acids in a
specific sequence; the sequence is determined by the sequence of nucleotides in

153

Appendix C:
Glossary

the gene coding for the protein. Proteins are required for the structure, function,
and regulation of the body's cells, tissues, and organs, and each protein has unique
functions. Examples are hormones, enzymes, and antibodies.

Recombinant DNA technologies: Procedures used to join together DNA segments in a
cell-free system (an environment outside of a cell or organism). Under appropriate
conditions, a recombinant DNA molecule can enter a cell and replicate there, either
autonomously or after it has become integrated into a cellular chromosome.

Resolution: Degree of molecular detail on a physical map of DNA, ranging from low
to high.

Restriction enzyme, endonuclease: A protein that recognizes specific, short
nucleotide sequences and cuts DNA at those sites. There are over 400 such
enzymes in bacteria that recognize over 100 different DNA sequences. See

restriction enzyme cutting site.

Restriction enzyme cutting site: A specific nucleotide sequence of DNA at which a
particular restriction enzyme cuts the DNA. Some sites occur frequently in DNA
(e.g.. every several hundred base pairs), others much less frequently (e.g.. every
10,000 base pairs).

RFLP, restriction fragment length polymorphism: Variation, between individuals, in
DNA fragment sizes cut by specific restriction enzymes; polymorphic sequences
that result in RFLPs are used as markers on both physical maps and genetic
linkage maps. RFLPs are usually caused by mutation at a cutting site. See
marker.

Ribosomal RNA, rRNA: A class of RNA found in the ribosomes of cells.

RNA, ribonucleic acid: A chemical found in the nucleus and cytoplasm of cells; it

plays an important role in protein synthesis and other chemical activities of the cell.
The structure of RNA is similar to that of DNA. There are several classes of RNA
molecules, including messenger RNA, transfer RNA, ribosomal RNA, and other
small RNAs, each serving a different purpose.

Sequence: The order of the nucleotides in a nucleic acid or order of amino acids in a
protein.

Sequence-tagged sites (STSs): Short (200-500 base pairs) DNA sequences that have a
single occurrence in the human genome and whose location and ba.se sequence are
known. Detectable by polymerase chain reaction, STSs are useful for localizing
and orienting the mapping and sequence data reported from many different
laboratories and could serve as landmarks on the developing physical map of the
human genome.

154

Sex chromosomes: The X and Y chromosomes in human beings that detemiine the
sex of an individual. Females have two X chromosomes in diploid cells; males have
an X and a Y chromosome. The sex chromosomes comprise the 23rd chromosome
pair in a karyotype. Compare autosomes.

Shotgun method: Cloning of DNA fragments randomly generated from a genome.
See lihrary. genomic library.

Shuttle vectors: Cloning vectors that are capable of replicating in both prokaryotic
and eukaryotic hosts.

Single-gene disorder: Hereditary disorder caused by a mutant allele of a single gene
(e.g.. Duchenne muscular dystrophy, retinoblastoma, sickle cell disease). Compare
polygenic disorders.

Somatic cells: Any cell in the body except gametes and their precursors.

Southern blotting: Transfer by absorption of DNA fragments separated in

electrophoretic gels to membrane filters for detection of specific sequences by
radiolabeled complementary probes.

Spheroplast: Yeast or bacterial cell from which most of the cell wall has been removed
by enzymatic or chemical treatment.

Sticky ends: See cohesive ends.

Technology transfer: The process of moving scientific findings into the commercial
sector for conversion to useful products.

Telomere: The ends of chromosomes. These specialized structures are involved in the
replication and stability of linear DNA molecules. See DNA replication.

Thymine (T): A nitrogenous base, one member of the base pair, A-T (Adenine-
Thymine).

Transcription: The synthesis of an RNA copy from a sequence of DNA (a gene); the
first step in gene expression. Compare translation.

Transfer RNA. (tRNA): A class of RNA having structures with triplet nucleotide
sequences that are complementary to the triplet nucleotide coding sequences of
mRN A. The role of tRN As in protein synthesis is to bond with amino acids and
transfer them to the ribosomes, where proteins are assembled according to the
genetic code carried by mRNA.

Transformation: A process by which the genetic information carried by an individual
cell is altered by incorporation of exogenous DNA into its genome.

155

Appendix C:
Glossary

Translation: The process in which the genetic code carried by niRNA directs the
synthesis of proteins from amino acids. Compare iiciiiscrlpHdii.

Uracil: A nitrogenous base nomiaily found in RNA but not DNA: uracil is capable of
forming a base pair with adenine.

Vector: See cloniiii; vector.

Virus: A nonceliular biological entity that can reproduce only within a host cell.
Viruses consist of nucleic acid covered by protein: some animal viruses are also
surrounded by membrane. Inside the infected cell, the virus uses the synthetic
capability of the host to produce progeny virus.

VLSI: Computer jargon: literally, "very large system integrated" (i.e.. 10.000 to
100.000 transistors on a chip).

Yeast artificial chromosomes (YACs): Cloning vectors [containing centromere

(CEN) and autonomous-replication sequences (ARS)| that are derived from yeast
chromosomes, eukaryotic telomere sequences, and a number of biochemical
marker genes.

156

Acronym List

AEC

ANL*

ATCC

BNL*

CEPH

DOE

ERDA

FCCSET

HERAC*

HGCC*

HGMIS*

HUGO

JITF*^

LANL*

LBL*

LLNL*

NAS

NIH^

NLGLP*

NRC

OHER*

ORNL

OSTP

OTA

PAC^

PNL*

SBIR

Atomic Energy Commission
Argonne National Laboratory, Argonne, IL
American Type Culture Collection, Rockville, MD
Brookhaven National Laboratory, Upton, NY
Centre d'Etude du Polymorphisme Humain
Department of Energy

Energy Research and Development Administration
Federal Coordinating Council on Science, Engineering and
Technology

Health and Environmental Research Advisory Committee
Human Genome Coordinating Committee
Human Genome Management Information System (ORNL)
Human Genome Organisation (international)
Joint Informatics Task Force
Los Alamos National Laboratory, Los Alamos, NM
Lawrence Berkeley Laboratory, Berkeley, CA
Lawrence Livermore National Laboratory, Livermore, CA
National Academy of Sciences (U.S.)
National Institutes of Health, Bethesda, MD
National Laboratory Gene Library Project (LANL, LLNL)
National Research Council (NAS)
Office of Health and Environmental Research
Oak Ridge National Laboratory, Oak Ridge, TN
Office of Scientific and Technology Policy (White House)
Office of Technology Assessment (U.S. Congress)
Program Advisory Committee on the Human Genome (NIH)
Pacific Northwest Laboratory, Richland, WA
Small Business Innovative Research

* Denotes U.S. Department of Energy organizations.

^ Denotes U.S. Department of Health and Human Services organizations.

A

m

DO

■D O
m -n

-Z. -n
^o

H >

< I-

■n ro

O c:

n w

-D z

n m

-p. w

>^

H

m
CO

jn

CO

o
o

g o c

> m z

C/5 2 H

I > m

;^ fTi r

o z >

o
o

ro
o
en
-P^
en

m
en