On Wednesday, September 13, 2000, the National Institute of General Medical Sciences (NIGMS) held a special open session to discuss a series of initiatives proposed for FY 2001 and FY 2002. The agenda, list of discussants, and initiatives proposed are available from NIGMS.
Dr. Cassman began with an overview of recent programs initiated by NIGMS. As a result of 1998 discussions with groups of investigators in research fields related to the Institute's mission, several areas of special interest were developed:
NIGMS defines complex systems as systems in which the behavior or expressed characteristic of a biological system is determined by the contributions of multiple interacting components with quantitative expressions that may vary in time and space. Quantitative analysis and modeling of these systems should result in an increased understanding of the global control and integration of biological processes. This will require the involvement of cross-disciplinary groups of scientists, including mathematicians, engineers, and computer scientists. A variety of initiatives were established in this area, all of which are summarized on the NIGMS Web site. Dr. Cassman noted that the Institute has had only moderate success in stimulating applications through these announcements.
The 1998 discussion groups also encouraged NIGMS to think about new ways to stimulate collaborations. This resulted in the "glue grants" initiative. These are programs to support collaborative efforts between groups of investigators. "Glue grants" do not support the underlying research--only the resources required for the collaboration to effectively function. The intent is to permit investigators to pool their efforts in order to reach goals that are not accessible through their own laboratories. Only one such program has been funded thus far--the Alliance for Cellular Signaling, headed by Dr. Alfred Gilman of the University of Texas Southwestern Medical Center in Dallas. Dr. Cassman noted that many of the "glue grant" applications have had a significant contribution from modeling and computational biology.
There are a number of diverse NIGMS initiatives in the area of research tools. These include technology development and instrument support for high-resolution electron microscopy, synchrotron facilities, and supplements for microarray analysis, to name a few.
The most prominent new NIGMS training program is a predoctoral program in bioinformatics and computational biology. Additionally, NIGMS has begun a new program in summer research experiences for undergraduates, aimed at placing individuals with quantitative backgrounds in biology labs.
Finally, a number of initiatives have emerged over the past few years from the NIGMS Division of Minority Opportunities in Research. Most recently, these include the Native American Research Centers for Health program and a post-baccalaureate training program.
Dr. Cassman noted that not all of the programs that the Institute has initiated fall neatly into the above categories. (It should also be noted that although the initiatives presented at this meeting largely could be seen to fall in the categories already mentioned, this was not uniformly true. As an example, NIGMS recently initiated a program announcement involving single-molecule studies.) Dr. Cassman pointed out that the described categories and the presented initiatives do not reflect the totality of the science that the Institute finds important or interesting; however, they do reflect areas of science that need stimulation. As important as is most of the science that NIGMS supports, most of it does not require additional stimulation. For most of the fields of science supported by NIGMS, there is a large body of investigators engaged in research, and the value of the research is well understood. Most of the initiatives presented are targeted to emerging areas where this is often not the case.
Many of the recent initiatives involve quantitative approaches to biology, and they can also be very large scale, involving well high-throughput data collection and analysis. Dr. Cassman indicated that this reflected a new research paradigm, which is best expressed by an anthropological metaphor--the change from hunter-gatherer to harvester societies. (This metaphor was first stated by Dr. William Gelbart of Harvard University.) The distinction is between approaches that involve small groups that are interactive but not highly collaborative, operating in a data-poor environment (hunter-gatherer) and those that involve highly collaborative, large groupings in a data-rich environment, often dealing with high-throughput collection and analysis of large bodies of data (harvester). Further, the hunter-gatherer researchers distribute information primarily through publications, while the harvester does this primarily through databases. The harvester approach is visible in many of the initiatives described by NIGMS staff, and the initiatives are intended to facilitate the requirements of science as these changes continue.
Dr. Shirley Tilghman of Princeton University pointed out that it is difficult for universities to adapt to the harvester environment, in part because it is not clear how people participating in large-scale collaborations will be rewarded. This is especially true for young people. Dr. John Kozarich of Merck Research Laboratories noted that industry has had to deal with this issue for some time, and it might be of value to discuss the problem with companies that have had such experiences. Dr. Cassman suggested that this might be an issue for future discussion.
The first initiative discussed was in the broad area of biophysical tools, involving the development of collaborations to test the capabilities of 900 MHz Nuclear Magnetic Resonances (NMR). Dr. Janna Wehrle of the NIGMS Division of Cell Biology and Biophysics described some of the potential contributions arising from the availability of these instruments. They included structure determination of proteins in the 100,000 to 150,000 Da range; better assessment of macromolecular motions; the study of partially and fully unfolded proteins; and the structures of carbohydrates. Additionally, there is the ability to partially orient molecules, improve chemical shift dispersion, and improve signal-to-noise.
The proposal is to create scientifically based user groups, focused on NIGMS grantees, that will be largely applications-oriented. NIGMS anticipates perhaps two awards in the first year, for about $10 million total. The awards will include funds for staff support.
During the discussion, it was pointed out that the use of NMR in determining structural and dynamic characteristics of proteins is about 15 years old, and the ability to examine larger proteins has been increasing rapidly, especially over the past 3 years. These advances are partly due to advances in technology. It was felt that at this juncture, the new opportunities provided by the 900 MHz instrument should be integrated with other technologies through extended cross-disciplinary collaborations.
Another NMR technology that will have a major impact on the sensitivity of existing instruments was identified--cryoprobes. These are considerably cheaper than the 900 MHz instrument--about $200,000--and might be a good target for competitive supplements. However, it was also noted that the demand for these might be overwhelming. Dr. Cassman said that the possibility of issuing an announcement for a competitive supplement for cryoprobes would be considered.
The next set of topics addressed the chemistry/biology interface, with the first initiative "Metals in Medicine," presented by Dr. Peter Preusch of the NIGMS Division of Pharmacology, Physiology, and Biological Chemistry. This initiative emerged from an analysis that showed that, although NIGMS supports a great deal of research on metalloenzymes, it supports rather little work on metal metabolism and regulation or on the use of metals or metal complexes as probes in examining cell behavior. A June 28-29, 2000, meeting, titled " Metals in Medicine," was organized by NIGMS to consider this area. The intent was to look at obstacles and opportunities in developing pharmaceuticals using the tools and concepts of metallobiochemistry and to look at emerging areas of metal metabolism and the role of metals in regulating cell processes.
The Council discussion suggested a strong interest in developing the understanding of metals in cell biology and in bringing inorganic chemists into collaborations with biologists; however, there was considerably less interest in developing metallopharmaceuticals. A broad discussion ensued on the best way to encourage applications from the research community in areas of particular interest to the Institute. Additionally, there was encouragement for bringing in other institutes and agencies on initiatives with common interests.
A second initiative in the area chemistry/biology interface, "Centers of Excellence in Chemical Methodology and Library Development," was presented by Dr. John Schwab of the NIGMS Division of Pharmacology, Physiology, and Biological Chemistry.
The proposal is to support the development of chemical methodologies for creating, analyzing, and ensuring the quality of chemical diversity (i.e., combinatorial chemistry) libraries. This would extend the ability to do high-throughput functional screening, which has become a common approach for drug discovery and for the discovery of biological effector molecules. The development of these libraries involves a process whereby multiple compounds can be generated simultaneously by techniques that involve parallel chemical transformations. Effective library development requires the interaction of synthetic organic chemistry, polymer science, analytical chemistry, and perhaps even engineering, through approaches such as robotics. The intent is to support method development in diversity-oriented synthesis. The process would involve the establishment of centers, to provide focal points for the development of the collaborations noted above, as well as interactions with biologists. The estimated costs in the first year would be $4 million.
There was considerable disagreement and debate about the value of this proposal. Some of the participants felt that this approach would be better suited to industry, and that, in fact, industry was already heavily involved. Others felt that the effort would be entirely appropriate--that industry is not making any effort to develop the chemical methodologies required for the synthesis of complex compounds by high-throughput or combinatorial approaches. However, it was also believed that more funds are needed in this area and that closer linkages with biologists are required. Yet, others felt that the role of biologists in the centers was not particularly clear, while some members were not enthusiastic about the use of centers at all. There was also a sense that the end product of this effort was not clearly delineated.
The conclusion was that the proposal should be reformulated to focus on the basic research component, to clarify the role of biologists in a program targeted to developing chemical methodologies, and to reassess the mechanism by which this program should be supported.
The next initiative discussed was the development of a generic model organism database. Dr. Judith Greenberg of the NIGMS Division of Genetics and Developmental Biology began by reminding the Council members that NIGMS is the primary supporter of research with non-mammalian model organisms at NIH, and that the Institute has a significant responsibility to consider the genomic requirements of these communities. The existing databases appear to be meeting most of the needs of their respective communities. Nevertheless, some improvements are still needed, especially to make it easier for investigators to go back and forth between the databases. In addition, other model organism communities have been approaching NIH about wanting to develop their own databases, and it would be very inefficient for each of these groups to create a database de novo.
Recently, NIGMS and the National Human Genome Research Institute (NHGRI) convened a small meeting of the leaders of the existing databases to suggest ways for improving model-organism databases. The group described the existing databases as being constructed of several layers (a data storage layer, an exchange or middleware layer, a query and display layer, and an annotation layer), each comprised of modules. They felt that the most important things that NIH could do would be to support the improvement of existing modules and to create new and better modules. As better modules would be developed, the existing databases would adopt them, which would result in the eventual convergence of the different databases on an optimum structure. More importantly, groups wanting to create new model-organism databases would not have to start from scratch, but instead, could create one by picking and choosing among modules that would be available off the shelf. They would be able to concentrate on the biology and not on the software engineering of their databases. NIGMS is therefore proposing efforts to improve the modules used in the existing databases and to develop new modules. The idea is to make the modules more robust, with written documentation and consistent formats to facilitate searches and open source. The modules could be used to create new, generic model-organism databases.
Achieving these goals would involve two separate mechanisms. First, in cooperation with NHGRI, NIGMS proposes to fund competitive supplements to the existing databases. This would enable them to make the existing modules more robust and to write the documentation for them so they could be shared. Second, NIGMS proposes to issue an RFA (request for applications) to solicit competitive research project grants. The applicants for these grants would have to demonstrate a collaboration with one or more of the existing databases so that the project would really lead to the development of a useful module that could interact with and enhance an existing database, or that would lead to the creation of a generic database.
There was a great deal of enthusiasm for this initiative, particularly for its attempt to ensure that new databases would be structured on a well-tested and consistent model. There was also discussion on the need to develop and make accessible ontologies and on the need to generate databases that can capture data on interacting systems.
The next discussion was on a set of initiatives related to BISTI, the NIH Biomedical Information Science and Technology Initiative. Dr. James Cassatt of the NIGMS Division of Cell Biology and Biophysics began by presenting background on the development of BISTI, which emerged out of a report promoted by Dr. Harold Varmus, former NIH director. The recommendations of the report were considered by an internal NIH implementation committee. NIGMS is a participant in two NIH-wide announcements that emerged from this committee's work. The first is for planning grants to establish centers, and the other is for phased innovation grants in computational biology, which allow a small starter award to phase into a larger award without additional peer review, if milestones are met. Additionally, NIGMS has established individual postdoctoral training grants, as well as institutional predoctoral training programs in computational biology and bioinformatics. NIGMS now proposes to add to these an institutional postdoctoral training program. This would bring in Ph.D.s with either a biology or a computational science background and provide some didactic cross training together with lab work before committing to the research effort in computational biology. There was general agreement that this approach would be useful. There was some concern that higher salaries than the norm for NIH postdoctoral fellows would be needed to attract the right applicants
A second initiative discussed by Dr. Cassatt was a proposal for a joint announcement with the National Science Foundation Division of Mathematical Sciences to increase the involvement of mathematicians in research to address biological problems. Dr. Cassatt noted that NIGMS' previous efforts to encourage applications from mathematicians have not been very successful. Some of the discussants were skeptical that academic mathematicians would have a significant interest in biology, and that computer scientists, engineers, and applied mathematicians might be more likely targets. Nevertheless, there was general agreement about the value of this proposal and its potential to reach investigators with backgrounds in mathematics.
Dr. James Anderson of the NIGMS Division of Genetics and Developmental Biology presented the next two initiatives. The first would encourage the development of centers in computational biology and bioinformatics. Dr. Anderson gave a brief background on existing efforts to develop the areas of computational biology. He discussed the R01/P01 program, "Quantitative Approaches to the Analysis of Complex Biological Systems." Over the past 2 years, NIGMS received 27 applications and made 14 awards, for a current year total of about $3.5 million. A supplements program has generated 18 applications with seven awards, and a program to develop short courses yielded 10 applications and six awards.
The general objectives of the centers would be to encourage institutions to lower the barriers to interdisciplinary research in computational biology; to foster recruitment and training in biology of investigators with quantitative backgrounds; to act as a focus for disseminating knowledge in this area of science; and to stimulate undergraduate training in quantitative biology. The research goals are to discover the fundamental principles that govern the behavior of complex dynamical biological systems; to insure that theory is linked with experiment; and to develop tools for analysis of complex systems. Targets for research would include developmental programming, metabolic flux, signal transduction circuitry, and organ system networks, to name a few.
There was some concern that very few institutions would be prepared to respond in a mature way to such a proposal, and that money for recruitment and for construction of appropriate facilities would first be required. Dr. Cassman noted that planning grants were available, although probably not for construction and recruitment, and that a number of institutions had already expressed an interest in applying for such centers. It was also pointed out that NHGRI already issued an announcement requesting applications for centers in genomics and bioinformatics.
The final proposal in this group of BISTI-related initiative was an RFA to promote the development and analysis of model biological systems that would be amenable to quantitative modeling. An example is bacterial chemotaxis, where a great deal of information is available to begin in-depth modeling. Dr. Cassman pointed out that this initiative is a variant on existing NIGMS announcements and could be accomplished without a new program. However, it was being presented as yet another way to encourage proposals. There was general enthusiasm for all the BISTI-related initiatives.
The final group of initiatives related to health disparities. The first, "Differences in Response to Injury," was presented by Dr. Scott Somers of the NIGMS Division of Pharmacology, Physiology, and Biological Chemistry. Injury, in this context, is injury to he organism, and it is linked to the NIGMS program in trauma and burn injury. Dr. Somers gave a summary of the phases following injury, i.e., seconds to minutes, minutes to hours, and days following the event, and the problems attendant to each of these. A major consequence of trauma is the induction of inflammation. Although this is necessary for healing, it can also emerge as a systemic response, systemic inflammatory response syndrome (SIRS), which may be fatal. There are multiple regulatory points for SIRS and a number of known inflammatory mediators, some of which are known to exist as genetic polymorphisms. For example, an allele for tumor necrosis factor, a pro-inflammatory mediator, has been correlated with a relatively negative outcome following sepsis. Are there comparable genetic factors involved in response to injury, and can they be found to vary among populations or by gender? This initiative is focused on encouraging studies on the genetic regulation of the inflammatory response or other parts of the injury response. Although model systems are the most likely starting point, human studies are not ruled out.
Council discussion included concerns about the validity of animal studies for application to humans and some questions about the difficulty of dealing with human populations. These concerns will be considered before proceeding with this announcement.
The final initiative discussed involved supplements to the pharmacogenetics research network, to identify the genetic polymorphisms that can lead to variations in the response to medication, and to learn how these polymorphisms are distributed in different racial, ethnic, and geographic groups. This will be done in close collaboration with other efforts at NIGMS to establish appropriate interactions with populations that may be studied. The discussion raised the possibility of linking this effort to programs other than the pharmacogenetics program.
Dr. Cassman then asked for an en bloc concurrence with a motion to approve the initiatives presented. He noted that in the case of the "Metals in Medicine" and "Centers in Chemical Methodology" initiatives, there was significant disagreement among the discussants, and that these initiatives would be reformulated to address the concerns expressed. The motion for approval was moved, seconded, and concurred on by vote.
A general discussion followed. Topics included issues related to peer review, the need for stable support of technical personnel, the problems of obtaining data for in vivo analysis and modeling, and training of physicians in basic research. The latter topic raised many concerns, particularly about the career paths of clinician investigators. Dr. Cassman promised to raise the issue again at a later date.
The National Advisory General Medical Sciences (NAGMS) Council was convened in closed session for its one-hundred and fourteenth meeting at 8:30 a.m. on Thursday, September 14, 2000, in Conference Rooms E1/E2, Natcher Conference Center, Building 45. Dr. Marvin Cassman, director of the National Institute of General Medical Sciences, presided as chairman. The meeting was open to the public on September 14 from 11:05 a.m. to 4:30 p.m. and was followed by the closed session for consideration of grant applications.
John N. Abelson, Ph.D.Jay C. Dunlap, Ph.D.Slayton A. Evans, Jr., Ph.D.Lila M. Gierasch, Ph.D.Wayne A. Hendrickson, Ph.D.Leslie A. Leinwand, Ph.D.Robert S. Pozos, Ph.D.D. Amy Trainor, Ph.D.Isiah M. Warner, Ph.D.Richard M. Weinshilboum, M.D.
Daniel J. Kevles, Ph.D.--not present September 13-15Angeline A. Lazarus, M.D.--not present September 13-15Leslie A. Leinwand, Ph.D.--not present on September 13Neil S. Mandel, Ph.D.--not present September 13-15
Adam Arkin, Ph.D.ProfessorLawrence Berkeley National LaboratoryBerkeley, CA
William Gelbart, Ph.D.ProfessorDepartment of Molecular Cellular BiologyHarvard UniversityCambridge, MA
Hobart Harris, Ph.D.ProfessorDepartment of SurgerySan Francisco General HospitalSan Francisco, CA
Neville Kallenbach, Ph.D.ProfessorDepartment of ChemistryNew York UniversityNew York, NY
John Kozarich, Ph.D.ProfessorMerck Research LaboratoriesRahway, NJ
Eaton Lattman, Ph.D.ProfessorDepartment of BiophysicsThe Johns Hopkins UniversityBaltimore, MD
Douglas Lauffenburger, Ph.D.ProfessorDivision of Bioengineering and Environmental HealthMassachusetts Institute of TechnologyBoston, MA
James H. Prestegard, Ph.D.ProfessorDepartment of Chemistry and Biochemistry and Molecular BiologyUniversity of GeorgiaAthens, GA
Shirley Tilghman, Ph.D.ProfessorDepartment of Molecular BiologyPrinceton UniversityPrinceton, NJ
Thomas Tullius, Ph.D.ProfessorDepartment of ChemistryBoston UniversityBoston, MA
Debra Schwinn, M.D.ProfessorDepartments of Anesthesiology, Pharmacology/Cancer Biology, and SurgeryDuke University Medical CenterDurham, NC
Daniel Hartl, Ph.D.ProfessorDepartment of Organismic and Evolutionary BiologyHarvard UniversityCambridge, MA
John W. Kozarich, Ph.D.ProfessorMerck & CompanyMerck Research LabsRahway, NJ
Terry Orr-Weaver, Ph.D.ProfessorMassachusetts Institute of TechnologyDepartment of BiologyWhitehead InstituteCambridge, MA
Joel L. Sussman, Ph.D.ProfessorDepartment of Structural BiologyWeizmann Institute of ScienceRehovot, ISRAEL
Keith D. Watenpaugh, Ph.D.ScientistRetiredCoupeville, WA
For the record, it is noted that to avoid a conflict of interest, Council members absent themselves from the meeting when the Council discusses applications from their respective institutions or in which a conflict of interest may occur. Members are asked to sign a statement to this effect. This does not apply to "en bloc" actions.
Council roster (available from NIGMS).
Dr. Brigid Hogan, Vanderbilt Medical SchoolDr. Howard Hiatt, Harvard Medical SchoolDr. Charlotte Kuh, National Research CouncilMs. Luciana Lopez, FDC ReportsMs. Pamela Moore, Capitol PublicationsDr. Georgia Persinos, Washington InsightMr. Brad Smith, American Chemical SocietyMs. Meredith Wadman, Nature
Dr. Dan Drell, Department of EnergyDr. Janice Hicks, National Science FoundationDr. Debbie Stine, National Academy of Sciences
National Institute of General Medical Sciences employees and other NIH employees:
Please see the sign-in sheet (available from NIGMS).
Dr. Cassman called the meeting to order and introduced and welcomed the guests and the five ad hoc members: Dr. Daniel Hartl, professor, Department of Organismic and Evolutionary Biology, Harvard University; Dr. John Kozarich, professor, Merck and Company; Dr. Terry Orr-Weaver, professor, Department of Biology, Massachusetts Institute of Technology; Dr. Joel Sussman, professor, Department of Structural Biology, Weizmann Institute of Science; and Dr. Keith Watenpaugh, recently retired from his position as a distinguished scientist, Pharmacia and Upjohn Incorporated.
Dr. Cassman thanked two NAGMS Council members who completed their terms of service with the September meeting: Dr. Slayton Evans and Dr. Wayne Hendrickson. Dr. Cassman announced that the new Council slate was approved, and that in January, the Council would be welcoming five new members. They are: Dr. Eaton Lattman, professor and chair of the Department of Biophysics, Johns Hopkins University; Dr. Susan Taylor, professor, University of California, San Diego, and Howard Hughes investigator; Dr. Douglas Lauffenburger, professor, Massachusetts Institute of Technology, and co-Director of the Division of Bioengineering; Dr. Debra Schwinn, professor of Anesthesiology and Pharmacology, Duke Medical School; and Dr. George Hill, professor of microbiology, Meharry Medical College. Dr. Cassman noted that NIGMS would be adding two new positions to Council, one to cover bioengineering and bioinformatics, and one in clinical research related to the Institute's primary programs in anesthesiology and trauma and burn research.
Dr. Larry Tabak, from the University of Rochester, is the new director of the National Institute of Dental and Craniofacial Research, replacing Dr. Harold Slavkin.
Dr. Carl Kupfer, who was director of the National Eye Institute for the past 30 years, has retired.
Ms. Naomi Churchill, who was the director of the NIH Office of Equal Opportunity, has left NIH. Dr. Cassman encouraged attendees to consider colleagues and others who might assume this role at NIH.
Dr. Cassman then mentioned some awards NIGMS Council members have recently received:
The minutes of the May 18-19, 2000 meeting were approved as submitted.
The minutes of the January 27-28, 2000 meeting were approved as submitted.
The following dates for future Council meetings were confirmed:
Dr. Cassman reminded the members of their responsibility and commitment and asked that they not schedule any other meetings, etc., for the dates that they had just confirmed, and that they inform their secretaries of these dates so that other commitments would not be made for them.
Dr. Cassman made a few brief remarks on the budget, noting that no action had yet been taken, and that a continuing resolution seemed likely for some indeterminate period. He also commented on the recent retreat attended by the NIH institute directors and senior staff in the NIH Office of the Director. The bulk of the discussion was on NIH internal management issues. Dr. Cassman noted that two major points of concern were the need to recruit additional personnel to handle new, labor-intensive programs, such as structural genomics, pharmacogenetics, and glue grants, as well as the need to support NIH-wide efforts such as Electronic Grants Administration.
Dr. Brigid Hogan, Vanderbilt University, reported to the Council on the National Academy of Sciences (NAS) Committee's study on the experiences of postdoctoral fellows in science and engineering. She reported that the number of postdoctoral fellows in the United States has doubled since 1980, and that over half are non-U.S. citizens. The Committee's report focused on the quality of the postdoctoral experience and the conditions of employment. Through a survey, workshop, and several meetings, the Committee examined many aspects of postdoctoral training. Salaries of biomedical postdoctoral fellows were found to be too low and benefits inadequate. Postdoctoral programs and employment conditions were found to be poorly designed. The Committee found that postdoctoral experiences were marked by poor communication, evaluation, and feedback. The Committee recommended that institutions employing postdoctoral fellows do much more to support these young scientists and that institutions form official offices to oversee postdoctoral programs. The NIH was asked to increase postdoctoral stipends/salaries and to create an office for assisting postdoctoral fellows around the United States and use its influence to improve postdoctoral training experiences. Dr. Hogan also discussed the recruitment of underrepresented minority and foreign postdoctoral fellows with the Council.
Dr. Howard Hiatt of Harvard University was the chair of the National Academy of Sciences (NAS) committee that was charged to examine national needs for research personnel in the biomedical and behavioral sciences. The report of this committee is the eleventh in a series and is responsive to the NRSA legislation that authorizes research training programs at NIH. Dr. Hiatt summarized the Committee's studies of current and future demand and supply for scientists in these fields. Young scientists are spending more time in both predoctoral and postdoctoral training positions and many reported being unhappy with their job prospects. He reported that their study found a substantial movement of employment of new Ph.D. graduates from the academic sector to the industrial sector, and that Ph.D. production in the biomedical sciences is considerably higher than available research positions in all employment sectors. The NAS report also discussed changes in the way graduate students are supported with NIH funds. Currently, twice as many graduate students are supported as research assistants on R01 grants as through NRSA training programs. This ratio is a reversal of the 1975 situation. Dr. Hiatt also pointed out the sharp increase in the number of U.S. graduate students and postdoctoral fellows who are not U.S. citizens or permanent residents. He reviewed the Committee's recommendations, especially the one urging that NIH increase the size of the predoctoral training grant programs and reduce the support of R01 research assistants. The Committee also suggested that NIH examine its policies for supporting non-citizens and its programs for increasing the number of underrepresented minority graduate students. Dr. Hiatt discussed these issues with the Council.
Intellectual property (IP) is a challenging issue with respect to several of NIGMS' new initiatives, especially large-scale, high-throughput efforts such as those in pharmacogenetics and structural genomics. A panel of six speakers discussed different aspects of the issue.
Dr. Cassman introduced Mr. Norman Latker, a patent attorney from Browdy and Neimark in Washington, DC, as one of the individuals involved in the initiation of the Bayh-Dole legislation. Mr. Latker summarized the history of Bayh-Dole, beginning with invention ownership disputes over inventions funded by the Department of Health, Education, and Welfare (DHEW) during the 1960s. The most prominent of these involved the commercialization of Gatorade, where the industrial organization involved in development sued the DHEW for clear title. This was settled on terms that served as the basis for a uniform DHEW Institutional Patent Agreement (IPA) policy in 1969 that left ownership to grantee institutions, with the understanding that they would manage and license these rights.
However, problems with implementing this agreement led in 1980 to the legislation that became known as the Bayh-Dole Act. This established a uniform government-wide patent policy guaranteeing ownership of Federally funded inventions by nonprofit organizations and small business. This Act repealed 33 conflicting agency statutes. Other subsequent legislation and policies further strengthened the central intent of Bayh-Dole to implement the commercialization of Federally funded discoveries.
The royalty returns to academic institutions are a measure of the success of Bayh-Dole. Even more important are the unseen successes - notably, the increased collaboration between industry and government research organizations.
Dr. Lisa Brooks, a program officer in NHGRI's Genetic Variation Program and Genome Informatics Program, gave the perspective of NHGRI. The basic philosophy is that data produced through large-scale efforts should be quickly placed in the public domain, without patents on the bulk of the data. This was agreed to by participants at an international conference dealing with high-throughput DNA sequencing. The agreement is that all sequences 1-2 kB or larger should be deposited in the public database every 24 hours, and that sequencers may not patent bulk sequences. This is fairly straightforward, since there is no easy or quick way of defining function given the rate of data production. The sequence itself is very basic information, without any utility in the absence of further studies.
This is equally the case with Single Nucleotide Polymorphisms (SNPs). The government funds SNP discovery through awards to researchers and in collaboration with the SNP Consortium (TSC), an industry consortium (which also includes the Wellcome Trust, a British non-profit foundation). Awardees agreed that SNPs could not be patented in bulk. Researchers have 90 days from initial discovery to validate and map their SNPs before placing them in the public database. TSC is filing defensive patents to establish patent priority and then placing the SNPs in the public domain, ensuring that no one can own exclusive patents on the SNPs sequences they discover. However, if a given SNP can be identified experimentally as being associated with a specific function, it can then be pursued by the researcher and a patent based on specific utility related to involvement in disease can then be issued.
In two other circumstances, NHGRI, together with several other NIH institutes, has been involved in issuing a Declaration of Exceptional Circumstances (DEC). In the case of two programs that involve the production of mouse knockouts and full-length cDNAs, the DEC states that Bayh-Dole will not be applicable, and the producers will not be able to patent the material. In both cases material resources are produced. The argument is they are tools that should be made generally available without prior constraints. This does not prohibit any subsequent patenting of these materials, by the producers or others, when functional relationships are established.
During the question period, there was a concern that for a particularly valuable SNP someone might wait longer than the 90 days required to develop it further before depositing the data. The answer was that deposition did not preclude them from patenting, since a full year was available to patent after deposition.
Dr. Jack Spiegel of the NIH Office of Technology Transfer discussed some of the major components of the NIH patent policy as applied to intramural research conducted by NIH, as opposed to those which arise from Federal funding of external entities. The essence of the policy is the need to advance public health, and so patents are sought only when further investment is required to develop the invention. The patent becomes the incentive for private industry to invest. Intellectual property protection is not sought when the products are research tools or research materials. However, it is not always easy with early-stage technologies to be able to distinguish that which is purely a research tool from those outcomes that may have diagnostic or therapeutic applications at a later stage in development. In cases where this is not clear, patents are obtained, and the licensing policy is such that in most cases a non-exclusive license is granted. In those few (10 percent) cases where exclusive licensing is granted, it is to ensure that a product is developed. Benchmarks are then put in place to guarantee that the developmental plan is followed to get the product to market.
Although NIH believes that this is appropriate, our grantees are not bound by this patenting and licensing policy. Any effort to enforce this is usually through DECs, which are extreme measures that should not often be employed. Dr. Spiegel said he felt that such measures were not appropriate for the programs that NIGMS is undertaking (e.g., structural genomics). However, NIH has issued a series of guidelines in the last year for the extramural community regarding patent rights and the access to research tools. These do not have to be followed, but they provide grantees with a view of NIH's expectations.
Dr. Spiegel then addressed the question of the kind of inventions that may arise from the programs being initiated by NIGMS, and whether there are any particular patent issues that need to be considered. One such issue is the concept of an invention needing to have specific utility before it can be patented. The U.S. Patent Office has identified three types of applications arising from DNA sequencing that have appeared in the past few years. The first are those sequences that have no specific utility associated with them, e.g., anonymous ESTs. While a few such EST patents have been issued, the latest guidelines from the U.S. Patent Office promise no further issuance of this kind of patent. It is unlikely that NIGMS' programs will generate anything comparable.
The second generation of applications is sequences whose utility is postulated based on homology. The U.S. Patent Office stance regarding these is somewhat ambiguous. If, for example, the homology was only good enough to identify a protein structure as being an enzyme, this is probably too general to meet their standard for utility. However, if some more precise identification of function could be made, it might be patentable. The final guidelines are still not out. This class of patents is clearly relevant to NIGMS' structural genomics program, and perhaps to other NIGMS programs as well.
The third generation of applications involves homologies that lead to a much more precise identification of function, including, perhaps, cellular as well as biochemical function. This would be clearly patentable, and often a consequence of structural genomics studies.
Given these options, a possible "nasty" issue in structural genomics could be the identification of a functional domain of a protein, and the determination of its structure. Could any patent of such a domain also cover all other proteins with a similar domain? This could lead to stacking, in which a protein would have several functional domains each of which is the intellectual property of a different group. This could be combined with all the other related patents on aspects such as algorithms needed to solve the structure, methods to purify the protein, etc. An interested company could then be faced with royalty stacking that might be a barrier to commercialization. Similarly, researchers might also find this a disincentive to further studies, since any outcome could only be commercialized after dealing with these previous stacked patents.
The analogy is to a toll-road, and the need is to avoid tollbooth gridlock. Further, this could also breed public distrust of the research enterprise, with consequent congressional intervention.
The discussion began with a question on the comparisons between the current situation in biotechnology and previous experience in other industries. Dr. Spiegel responded that in the semiconductor industry something similar seems to have happened, but that it apparently had no chilling effect on the progress of technological development in that field. One questioner asked whether there is something unique or at least different about biotechnology. One answer was that the biotechnology patents are often on research tools rather than products. It was noted that this was not true in chemistry, for example. An industry representative pointed out that there is an effort being made by some (usually large) corporations to put as much information in the public domain as possible, on the assumption that they can exploit it faster than others, without having to pay the tolls. Another participant from industry stated that his organization does patent research tools, to accumulate chits that can then be traded with other corporations that have patents on important steps in a process. Essentially, these are defensive patents.
A concern was expressed that the absence of patent protection might lead to unforeseen consequences, with the subsequent inability to have an entry into the marketplace. It was suggested that non-exclusive licenses might be a better option. Then cross licensing, even for research tools, would allow the tollbooths to be removed or diminished. An analogy was provided by one discussant to the jukebox industry, where a pooling agreement was established. In this case all the rights were pooled and revenues were allocated on the basis of use. It was suggested that some kind of public organization could be established that would pool patent rights, with revenues allocated based on the degree to which the particular patented discoveries are used.
A final point was that the frequent use of the public domain might reduce the incentives of the inventor to report an invention.
Dr. Alfred Gilman, the principal investigator of the Alliance for Cell Signaling in the Department of Pharmacology at the University of Texas, Southwestern Medical Center, Dallas, discussed the approaches taken by the Alliance. This is a large consortium, primarily supported by NIGMS, which focuses on the understanding of cellular signaling networks. (The National Institute of Allergies and Infectious Diseases, the National Cancer Institute, six large pharmaceutical companies, and several private donors and foundations are also contributing funds.) The consortium currently includes over 50 participating investigators from 20 academic institutions and several industrial entities. The majority of the experimental work will be done in seven centers located at five different academic institutions.
The information arising from the centers' activities will be immediately released and put into the public domain. Intellectual property rights will not be retained by the participating investigator or the centers or universities in which they are located. The argument is that the material and information produced will be primarily research tools, not products, and that these need be made broadly available with minimal constraints in order that the maximum number of interested scientists can take advantage of them. Another consideration in establishing this policy was that the input for anything the Alliance releases would likely come from many different sites, and that assigning rights would, consequently, be extremely complicated.
During the discussion, sentiments were expressed that giving up all intellectual property rights could be counter-productive, and that if the issue was ensuring immediate release of data, that could be accomplished with provisional patents. This can be done by immediate filing with no claims attached and would protect patent rights for 1 year. Since many investigators may be involved, some collective mechanism, such as in the jukebox analogy, could be used to assign rights. There was considerable disagreement about the difficulties of assigning credit in a large consortium, and the value of doing so. However, it was felt to be extremely important that the ultimate agreement among all the investigators and institutions on a common approach to intellectual property rights was arrived at without external (i.e., government) direction.
Dr. Richard Weinshilboum of the Department of Molecular Pharmacology and Experimental Therapeutics at the Mayo Foundation, Rochester, MN, discussed the activities of the Pharmacogenetics Research Network from his perspective as Chair of the Network's Intellectual Property Subcommittee. The Network is a series of collaborative research groups supported by NIGMS and other NIH institutes, aimed at increased understanding about the way inherited variation contributes to individual differences in drug responses. The information collected will go into a public database. The group is quite different from the Alliance for Cellular Signaling in that its members are an assemblage of peer-reviewed applicants, and the only formal point of common involvement is through the database, PharmGKB. Further, it is more likely to discover products of commercial value than will the Alliance, and thus will handle intellectual property differently.
There are two general approaches in the Network--genotype to phenotype and phenotype to genotype. The focus is on humans, and the end product in both examples is likely to be a diagnostic. Possible products may thus be common sequence variants of functional importance. To address issues of patent rights, the Network has considered allowing provisional patents, which would allow immediate release of information and yet retain patent rights when the functional genomics is done. Other issues that have arisen are conflict of interest in validating a diagnostic or therapeutic agent, confidentiality, and, a major concern, namely, who owns the data in the PharmGKB database.
Dr. Michael Davis, a patent attorney from Klauber and Jackson in Hackensack, NJ, discussed a primary goal of health research, which is to generate biological information and turn it over to the pharmaceutical industry to develop into products for the public good. The question is whether this can best be done by developing intellectual property rights or by giving them away. There are three means of protection--patents, trade secrets, and database protection. Database protection is still not well defined; trade secrets do not permit release of information to the public; this leaves patents as perhaps the optimal means of protection. Can meaningful patent protection be arrived at for the kind of technology discussed today? The atomic coordinates of a protein, for example, are not patentable, but the use of the atomic coordinates in a drug discovery method is patentable.
The incentives for patenting are clear--the reward for spending the time and effort to invent without fear of later competition from others that will appropriate the products of this effort. The public also benefits since patents require full disclosure of the invention. One approach to optimally utilize the products of NIH-funded research is to protect the intellectual property by filing patent applications and then provide non-exclusive licenses to the pharmaceutical companies at a reasonable fee. New provisional rights have been created that speed up the time that intellectual property can be protected after filing a patent application. After November 29, 2000, patent applications can be published within 4 to 6 months of their filing, and provided that a patent eventually issues for the claimed invention, a reasonable royalty can accrue from a third party who practices the claimed invention in the United States, beginning on the date that the third party receives notice of the publication of the application.
The point was made that the approaches of both the Alliance and the Pharmacogenetics Network provide different but reasonable approaches to moving along the highway that leads to commercialization. It was noted that the differences result from the Network being a group of individual laboratories, while the Alliance laboratories are the centralized efforts of many contributors.
It was clear that many different approaches are available to address intellectual property rights, and that for structural genomics in particular, the optimum approach has yet to be established.
Dr. Cassman brought to the attention of Council members the procedures for the conduct of the meeting. Council members were reminded that all of the review materials furnished are privileged information. Although most conflicts of interest involving institutional affiliation already had been identified, members were asked to absent themselves during discussion of any application in which there was a personal conflict that was not readily apparent.
A summary of applications reviewed by Council is attached (available from NIGMS).
The meeting adjourned at 12:00 p.m. on Friday, September 15, 2000.
I hereby certify that the foregoing minutes are accurate and complete to my knowledge.
