conference funding level for your Center was under your request last year. Are we still on schedule for completing the project in fifteen years?

Answer. As you noted, the 15-year clock began ticking on October 1, 1990, at the start of FY 1991. As we are only 6 months into the first year, it is really very early for me to give a firm answer to your question. The goals that we have set for ourselves are very challenging ones. To meet them, we will have to improve our capabilities for producing genomic maps and generating DNA sequence many-fold compared to what we can do today. I believe that we can do this, and I look at the improvements that have occurred just within the past few years, from before the genome program was even organized and during its very earliest phase, to justify my confidence. I think that with the level of funding that we have currently, and with the projected increases that we hope will be appropriated over the next 2 to 3 years, the likelihood is high that the necessary technological improvements will be developed and utilized within the predicted 15-year timeframe.

ETHICAL, LEGAL AND SOCIAL IMPLICATIONS PROGRAM

Question. Dr. Watson, you have told us that you plan to devote 3 percent of the funding of the National Center for Human Genome Research to study the ethical, legal and social issues arising out of the uses of genetic information. What are the major issues in this area? What progress have you made?

Answer. Three major sets of questions have been identified by the Program on the Ethical, Legal and Social Implications of Human Genome Research as particularly important to pursue as the genome initiative proceeds.

First, issues involved in the integration of new genetic tests into health care. Human genome research is expected to greatly increase the number of gene-based diagnostic and prognostic tests available to health professionals. The social policy problems involved in integrating those tests effectively into medical practice include developing standards for a number of components of genetic health care. These include standards for: (1) insuring the accuracy and quality control of genetic tests; (2) defining the indications for testing and the design of testing protocols; (3) establishing the professional responsibilities of clinicians who perform tests; (4) protecting the confidentiality of information obtained from testing; (5) controlling access to and use of test results by third parties like health insurers; and (6) providing reimbursement for testing and test-related counseling. To help develop these standards, the NCHGR has commissioned a 2-year study by the National Academy of Sciences/Institute of Medicine, designed to assess and propose professional guidelines for the clinical introduction of new genetic

tests.

The current focus of much of the discussion of these issues is a test for the DNA mutations involved in cystic fibrosis (CF), which now makes widespread testing for CF carrier status conceivable. The professional practices and policies developed with respect to CF carrier testing are expected, to establish important precedents for the development of subsequent tests, like the genetic predictors of breast and colon cancer risk now on the horizon. As a result, the Center is developing an initiative to evaluate and establish sound policies before the test for CF carrier status becomes diffused into practice.

Second, issues involved in educating and counseling individuals about genetic test results. The primary risk that the diffusion of genetic tests poses is the misinterpretation of their findings and the resulting potential for psychological trauma, stigmatization and discrimination against those availing themselves of the tests. This risk broadens as the genetic elements of more health problems are uncovered and gene-based tests for susceptibilities and carrier states are developed.

To protect against these risks, NCHGR is soliciting and supporting projects aimed at improving professional and public understanding of these tests and their implications. For example, the NCHGR is already supporting a 10-part television series, "The Future of Medicine," which will explore the

clinical impact of genome research for a public audience. In addition, the NCHGR will be involved in over 15 open meetings, in 11 states and the District, devoted to the public discussion of the social implications of genome research during the 1990-91 academic season. Other projects involve studies of genetic risk perceptions, stigmatization and discrimination. The conclusions of these studies, and the continuing involvement of those at risk for genetic stigmatization in the program's projects, will help inform the professional deliberations of the medical community, and help give substance to educational projects seeking to improve public understanding.

Third, issues of access to and use of genetic test results by third parties, including insurance providers, researchers and employers. One way to combat the unfair use of genetic information is to protect its privacy. Because genetic information pertains to the most personal aspects of individual's lives--their health problems and reproductive plans--most wish to ensure its confidentiality. But genetic information almost always has implications for other people's welfare as well: spouses, children and extended family members. The interests of insurers, employers and biomedical researchers can also be affected. As a result, the ethical and legal bases (and limits) of such protections are still unclear. Thus, the program is supporting research projects addressing the legal status of genetic information, the information-sharing practices of biomedical researchers, and the clinical ethical issues involved in maintaining confidentiality within families.

At a January 10-11, 1991 program workshop on these issues, grantees and consultants identified two profitable avenues for the development of public protections against the unfair use of genetic information. First, the recently enacted Americans with Disabilities Act has the potential to prevent the abuse of genetic information in the context of employment. The language and intent of the Act apparently extends its protections to persons at risk for genetic disorders. ELSI researchers are currently developing specific

recommendations to the E.E.O.C. about how best to make these protections against genetic discrimination explicit in the implementation of the Act. Second, the Health Insurance Association of America, which is in the process of establishing its own policies with respect to genetic testing, has invited the NIH to work with them in developing their policy, and an interdisciplinary Insurance Task Force has been established by the program to develop recommendations for industry policy. Meanwhile, ELSI researchers are also working with state governments towards the development of model legislation in this area.

In summary, these activities have allowed the NCHGR to accomplish four important preliminary goals: (1) inaugurating open discussions of ELSI issues by the public, the scholarly community, and policy makers; (2) initiating a program of research on priority areas of concern; (3) facilitating the development of professional and social policy key issues; and (4) fostering public education about genome research and its implications.

COST OF HUMAN GENOME RESEARCH CENTERS

Question. Originally you projected that the average cost of the human genome research centers would be $4 million per year. Your current estimate is $2.9 million. Why has this estimate changed?

Answer. As you know, in the course of peer review, the cost of the research along with its scientific merit is critically evaluated. The initial human genome research center grant applications we received requested budgets averaging $4 million. As a result of very stringent peer review, the recommended budgets were reduced and we now estimate about $2.9 million per human genome research center in FY 1991 dollars. The resulting human genome research centers are of outstanding quality and operate on very lean budgets. It should be understood, however, that even the awarded levels are only current best estimates of the cost of large-scale mapping, since mapping on this scale has not been attempted before now.

I would also like to anticipate another issue related to the cost of an average human genome research center. We expect to initiate a program of

developmental grants (P20) to provide support for institutions where investigators would like to establish a human genome research center or program project, but must establish a track record in genome research, develop interdisciplinary collaborations, or generate additional preliminary data and/or resources before submitting an application. These grants will be limited in size to $750,000 total cost and will be funded according to NIH budget protocol through the center budget line. Each developmental grant will be counted as an entity within the center budget, which will greatly reduce the average cost of each entity. In the future it will be necessary for us to differentiate between human genome research centers and human genome developmental grants for purposes of discussing average cost as well as overall progress of the center program.

MAPPING CHROMOSOMES

How many

Question. I know there are 24 chromosomes in the human genome. of these chromosomes have you started to map? What plans do you have for those chromosomes that you have not started mapping?

Answer. In terms of genetic or linkage mapping, the NCHGR has started to develop "framework linkage maps" for all of the human chromosomes. Framework linkage maps, which are composed of only the most informative markers, are designed to be an important first step towards the final goal of "high resolution" maps and, at the same time, to be an early product of the Human Genome Project for use by the larger scientific community. With respect to high resolution linkage maps, those for chromosomes 1, 4, 5, 7, 10, 16, 17, 21, 22, and X, are currently being developed, representing about half the chromosomes.

-

With respect to physical maps, major projects designed to produce maps of the entire chromosome have been initiated in the NCHGR Human Genome Program Centers for five chromosomes Smaller chromosomes 4, 7, 11, 17, and X. scale physical mapping projects have begun for seven others chromosomes 3, 5, 9, 18, 21, 22, and Y. If successful, many of these latter can be expected to grow and/or stimulate larger projects that address the entire chromosome. Together with the three chromosomes (16, 19, and 21) being tackled at the DOE Genome Centers, we have already begun to map 14 of the 23 human chromosomes.

We do not anticipate problems in finding research groups interested in mapping the remaining chromosomes. We have grants pending award or have had inquiries and/or applications about mapping most of the remaining chromosomes. However, to further stimulate interest and to attract outstanding investigators into the genome field, we are preparing to initiate a program of developmental grants that will support planning and feasibility studies toward the development of the interdisciplinary research programs required to improve mapping technology and produce the high quality maps, which are two of the 5year goals of the Human Genome Project.

MAJOR ACCOMPLISHMENTS OF THE NCHGR

Question. What do you consider to be your major accomplishments in the first year of the National Center for Human Genome Research?

Answer. As you can imagine, there has been an enormous amount of activity during our first year and I believe we have accomplished a great deal. Administratively, we have successfully organized and staffed the National Center for Human Genome Research. I am particularly pleased that we have rapidly and successfully initiated two NCHGR programs that I believe are crucial to the success of our mission, the Human Genome Research Centers program and the Ethical, Legal, and Social Implications (ELSI) program. We have also recruited outstanding individuals to serve on the newly chartered Genome Research Review Committee study section and the National Advisory Council for Human Genome Research that will be responsible for the review of grant applications assigned to the NCHGR.

Scientifically, the most important developments have been in the area of physical mapping. The 5-year physical mapping goal is to "generate sets of overlapping cloned DNA that are continuous over lengths of 2 million base

pairs for large parts of the human genome."

The magnitude of this task can be illustrated by noting that, just a few years ago, the largest DNA fragment that could be cloned was about 40,000 base pairs long and theoretical calculations showed that the biggest overlapping sets of cloned DNA's of that size that could be attained was under 200,000 base pairs. In the last year, scientists at Washington University in St. Louis have unequivocally demonstrated that, using newer cloning techniques, DNA fragments several hundred thousand base pairs in size can be cloned and that overlapping sets that span more than 2 million base pairs can be achieved. Thus, we now know that the physical mapping goals that we set for ourselves are attainable.

Moreover, this work demonstrates some of the benefits that the improved technology and maps will offer. One of the genomic regions that was isolated in an overlapping clone set was that containing the cystic fibrosis gene. This work was done by a single investigator in 6 months for less than $40,000. In contrast, the efforts of at least ten laboratories over the course of 8 years were required in the original isolation of the gene. Furthermore, using the newer technique, the gene was isolated in intact form and it was surrounded by 20 times its length of normal genetic material.

Other notable achievements that have been made in the past year include the development of an alternative system, based on the Pl bacteriophage, for cloning large DNA fragments and the development of new, useful classes of markers that will speed the construction and usefulness of the human genetic linkage map.

HUMAN GENOME PROJECT BENEFITS SCIENCE

Question. Tell me,

how is the Human Genome Project benefitting science right now? Do we have to wait for the project to be completed before the information is available?

Answer. No, we certainly do not have to wait until the Human Genome Project is completed to reap its benefits. Significant advances are already apparent. There has already been a marked improvement in the underlying technologies, such as cloning and mapping techniques. For example, genome scientists have constructed markedly better collections of cloned DNA, better in the sense that the individual members of the collections are much larger than were previously available and the collections themselves are more complete. From the improved libraries, a number of clones containing interesting genes, including those associated with diseases such as cystic fibrosis, hemophilia, and a form of X-linked mental retardation have been isolated. Technological improvements have already been noted in the area of DNA sequencing as well. The first generation of automated sequencing machines has been brought on line to routine use. The continued development of better sequencing machines, as well as the adaptation of robots to take over much of the repetitive preparative work, is now an area receiving much attention.

In terms of new data, a key characteristic of the genome project, which distinguishes it from many other identifiable scientific projects,. is its incremental nature. Every time a new marker is mapped, the maps improve and immediately become more useful to the general scientific community. Improved genetic maps of several human chromosomes have become available in the past year.

Thus, from the moment the first designated genome funds were spent, the project began returning the kinds of results it promised. I am confident that benefits such as these will continue to appear throughout the course of the project, and I expect they will do so at an increasing pace.

COST OF SEQUENCING

Question. I understand sequencing of DNA is still very expensive. How much does it currently cost and what are you doing to bring the cost down?

Answer. The rate of accumulation of DNA sequences in GenBank, the DNA sequence databank, continues to increase exponentially. The increase is largely due to the improved efficiency and reliability of the commercial

automated sequencers and, as a result, the greater number of scientists who are producing sequence data. There is promising technology supported by NCHGR that I expect will improve the rate and decrease the cost of sequencing by commercially available machines in the next 5 years. Using capillary gel electrophoresis, NCHGR-supported scientists have been able to increase the number of base pairs that can be determined in a given time by a factor of about 10 compared to current methods. The use of mass spectroscopy to detect DNA fragments during sequencing is also being developed. The advantage to such a scheme is that smaller amounts of DNA can be sequenced.

Recently, the sequences of several very large (45 to 100 kilobases in length) segments of DNA were submitted to the databanks by NCHGR-supported investigators. While these are not the first sequences of this size, the numbers of sequences of this size are clearly increasing, indicating that large scale sequencing is improving.

The current cost of sequencing ranges between $2 and $5 per base pair, depending on the laboratory. It is difficult at present to predict whether the program will attain the 5-year goal of reducing the cost of sequencing to $0.50 per base pair. We are using a two-faceted strategy to approach this problem. We are supporting completely new technologies, which will dramatically improve efficiency and reduce costs, but may take 5 to 10 years before becoming available commercially. In addition, we support several pilot sequencing projects that have the goal to sequence one million base pairs of finished sequence in the next 3 years by scaling up and making incremental improvements in current methods. These projects are sequencing the genomes of important model organisms, the roundworm (C. elegans) and two bacteria (E. coli and Mycoplasma capricolum) or biologically important regions of the human genome (the T-cell receptor region). We expect that using these approaches we will be able to significantly reduce the costs over the next 5 years.

CONTRIBUTIONS OF THE HUMAN GENOME RESEARCH CENTERS

Question. The FY 1992 budget requests funds for 11 human genome research centers. How will these centers contribute to the Human Genome Project?

Answer. The human genome research centers are essential to the goals of the Human Genome Project, because they represent the interdisciplinary groups that are needed to achieve major objectives such as the mapping of a complete human chromosome. These human genome research centers will each have an outreach component through which collaboration will be established with other projects and information and materials made available to the community as a whole.

The two overall goals of the human genome research centers funded so far are mapping and technology development. Three human genome research centers (Washington University, University of California at San Francisco, Baylor College of Medicine) have chosen the development of a physical and genetic map of one or more human chromosomes as their goal. The center at the Massachusetts Institute of Technology will develop the physical and genetic map of the mouse. The other two human genome research centers (University of Michigan and University of Utah) have chosen mapping, sequencing, and gene identification technology development as their goals. All of the human genome research centers, however, include significant efforts in various aspects of technology development. Additional human genome research centers that will be funded will have similar goals of mapping and technology development.

DATA SHARING

Question. What are your plans for sharing genome data with the rest of the scientific community?

Answer. This is a very important question because the raison d'etre of the Human Genome Project is the generation of an extraordinarily valuable set of tools for use by the entire biomedical research community. Furthermore, because mapping and sequence data will be generated by the genome project incrementally and each new data item will be useful as soon as it is produced, it is vital that we develop means for distributing the data rapidly and