Enabling Research at the Intersection: Promoting Training, Support, and Communication Across Disciplines
A great deal of substantive science is being done now where the life and physical sciences intersect, and even more transformative research is on the horizon. While the potential benefits for society are profound, realizing that full potential will require significant changes in how we educate, train, and support those undertaking this research.
The historically rigid department structure at universities, the programmatically isolated stove-piped nature of much federal funding, the different ways in which life science and physical science research are organized, and the largely separate spheres of communication that isolate life scientists from physical scientists serve as barriers to the multidisciplinary connections highlighted in this report. Intentionally or not, our system pits one scientific area against another in competition for a limited pool of resources. To obtain the benefits from research at the intersection of the physical and life sciences, it will be necessary to overcome these obstacles and to create a scientific structure that truly reflects the scientific needs and opportunities of twenty-first-century science.
Breakthroughs occur when scientists from a variety of disciplines either individually or collaboratively work on important interdisciplinary and multidisciplinary problems. Therefore, we need a new generation of scientists with both rigorous disciplinary training and the ability to communicate and work easily across disciplines. This chapter addresses the third task in the committee’s charge—namely that it explore ways to enable and enhance effective interactions between the life and physical sciences, discussing some of the important areas where change is needed so that the full potential from research in the area can be realized.
CONNECTIONS BETWEEN DISCIPLINES
The degree of connection that can take place in work between disciplines varies, and the demands on those participating in and supporting that work will likewise differ. This section discusses several categories of connections; even though any such categorization is approximate at best. Actual crosscutting research efforts are more a part of a continuum rather than belonging in clearly separate categories.
The first degree of connection is simply a collaboration of experts from different disciplines, with each contributing expertise from his or her own field without crossing over into the other field (“multidisciplinary”). A physicist might build an imaging device for her neurobiologist collaborator, or a computer scientist might analyze complex sequence data generated by his geneticist collaborator. Researchers participating in this type of collaborative effort do not require extensive in-depth knowledge or skill sets from their collaborator’s discipline. Although always desirable, truly interdisciplinary training in this type of minimally cross-disciplinary work is not strictly necessary. However, the collaborators must at least speak each other’s language—that is, communicate across disciplinary borders.
A second type of research is conducted by individuals who were originally trained in one of the classical disciplines but since have acquired skills and knowledge in another discipline. This might include, for example, additional coursework at the graduate level or postdoctoral training in the other field. The research performed by these individuals, either by themselves or with others similarly trained, is currently often referred to as interdisciplinary research. Such cross-training mostly happens owing to the interests and initiative of individuals. Work of this type could be facilitated by encouraging classically trained physical and life scientists to transcend their disciplines and acquire education and training in other fields. One mechanism that fosters the cross-over of individuals trained in the physical sciences into the life sciences is discussed in the next subsection.
A third category is research in new fields that have emerged from previous connections between disciplines (“interdisciplinary integration”). Such fields as biomedical engineering and biostatistics combine features of several traditional disciplines into a new discipline. The emergence of such fields is not new, as several of the distinct fields now firmly established in today’s universities, including molecular biology and biochemistry, have their origin in the intersection of past disciplines.
Culture of Separation between the Life and Physical Sciences
Conducting research at the intersection of the physical and life sciences requires bringing together not only separate disciplines but also, in many senses, separate cultures. While biologists, physicists, and chemists may not be that different in
some respects, the way they conduct, communicate, and organize their science can be very different in other respects. Building the interdisciplinary and multidisciplinary connections discussed throughout this report will require overcoming these distinctions so that scientists from a variety of disciplines can work together on problems of common interest.
The heart of biological research has been single-investigator-initiated projects of relatively short duration. Although some of the recent projects, such as the Human Genome Project, have been large ones, they are not the main source of support for biology. In contrast, much federal support for large segments of the physical sciences, such as astrophysics and high-energy physics, goes to large facilities and programs, and those projects, unlike a typical life sciences project, almost certainly have permanent staff and involve large amounts of instrumentation and construction support.
Physical scientists who participate in these large-scale research efforts also are accustomed to awarding credit to large numbers of investigators and are comfortable with scientific papers that have hundreds of authors. In contrast, publications in the life sciences tend to include no more than one, two, or three principal investigators as authors, along with a handful of graduate students, postdoctoral researchers, and others who conduct the actual research. As the number of authors per manuscript increases, it becomes progressively more difficult for most of the authors to receive adequate credit for their contributions. Physics seems to have found mechanisms to circumvent this problem, but the issue of how investigators are evaluated remains one of the major cultural divides between the two disciplines.
Life and physical scientists have typically been members of largely separate scientific communities, attending different meetings and reading different journals. The committee encourages universities, professional societies, and funding agencies to seek ways to connect researchers across disciplines. The Keck Futures Initiative of the National Academies provides one model for bringing together researchers from across disciplines (see Box 6-1). Recommendation 3 in the 2008 National Research Council report Inspired by Biology (National Research Council, 2008b) may also be helpful in this regard, as it suggests summer courses that bring together life scientists and physical scientists and allow researchers to be introduced to other disciplines.
Another successful model of interdisciplinary community building has been developed by the Kavli Institute for Theoretical Physics and the Aspen Center for Physics. These institutions bring physicists and biologists together for extended workshops in a format that allows new collaborations to germinate. This format has been adapted from long-standing practices of the theoretical physics community but has proven very effective in the interdisciplinary setting: Most of the life scientists introduced to the highly interactive experience of these workshops choose
National Academies’ Keck Initiative
National Academies has launched a 15-year effort to realize the untapped potential for research that crosses disciplinary boundaries, thanks to a $40 million grant from the W.M. Keck Foundation. The Keck Futures Initiative brings together approximately 100 top scientists from a variety of disciplines to consider a series of questions and challenges. Following several days of conversation and engagement in groups, participants have the opportunity to apply for seed grants that will enable launching ideas generated at the conference.
Several of the Keck Futures conferences have focused on research at the intersection of the physical and life sciences, including the 2003 meeting, “Signals, Decisions and Meaning in Biology, Chemistry, Physics and Engineering”; the 2004 conference “Designing Nanostructures at the Interface Between Biomedical and Physical Systems”; and the 2008 focus, “Complex Systems.” Additional information about the Keck Futures Initiative is available at http://www.keckfutures.org/
to participate over and over, resulting in steady growth of the interdisciplinary communities associated with these institutions. Workshops like these influence the research agendas of the participating scientists and play an important role in developing a common language uniting different scientific communities. They lay the foundation for breakthroughs not possible when the disciplines work in isolation.
Culture and Organization of Academia
It is not only the culture of disciplines that complicates research at the intersection of the physical and life sciences but also the culture and traditions of academia. In particular, most universities are divided into colleges and then into academic departments, usually along traditional disciplinary lines such as biology, chemistry, and physics, which then form the basic administrative units for the university. The hiring, promotion, and granting of tenure for faculty, graduate programs, and “credit” given for instruction generally are all determined with respect to these departments. Although many faculty members have joint appointments in more than one department, they often must meet tenure criteria and take on teaching responsibilities in a “home” department.
Because their work often does not squarely fall within the purview of a single department, faculty members working at the intersection of the physical and life
sciences may find it difficult to get support in any one department. Although a full discussion of university tenure policies is well beyond the scope of this report, the disciplinary-based nature of promotion and tenure is an impediment to multidisciplinary education and research. These statements are not intended to declare that the traditional disciplinary structure is misguided or shortsighted, but rather that alternative support structures may be called for in light of these particular challenges.
The organization of university research has, however, shown promise of reform. In particular, multidisciplinary centers organized around shared research topics or common research goals are becoming more common. These centers can provide an alternative model for organizing research activities in a way that complements—but does not replace—existing departments. Centers enable universities to move rapidly into emerging fields, provide a home for faculty working at the intersection of disciplines, and develop courses and train students free of traditional departmental and disciplinary constraints. They provide opportunities increasingly viewed as important to members of the next generation of scientists, many of whom are attracted to work on certain problems rather than in particular subdisciplines. Because these centers can achieve the multidisciplinary goals discussed in this report without replacing the existing university structure, they are an attractive mechanism for promoting research at the intersection of the physical and life sciences in the medium term. Recommendation 1 calls for further support for these efforts.
RECOMMENDATION 1. Federal and private funding agencies should expand support for interdisciplinary and multidisciplinary research and education centers. In particular, extramural funding should be provided to establish and maintain center infrastructure and research expenses. Initial (e.g., 5-year) salary support for investigators performing research that spans disciplines should also be included, with continuing salary support for faculty associated with the center provided by the host institution(s) or department(s). To support these centers, universities will need to implement multidepartment hiring practices and tenure policies that support faculty working collaboratively within and across multiple disciplines, establish shared resources, and provide incentives for departments to promote multi-departmental research and cross-disciplinary teaching opportunities.
ORGANIZATION OF SUPPORT FOR RESEARCH
Research at the intersection of the physical and life sciences necessarily falls between the boundaries of disciplines, which also means that it often falls between the boundaries of how research support is organized. Several federal agencies—including the Department of Defense (DOD), the Department of Energy (DOE),
the National Institutes of Health (NIH), and the National Science Foundation (NSF)—support research at this intersection through a variety of mechanisms, from individual, investigator-initiated research grants to large centers and consortia. Which agency is most appropriate for a given proposal can be difficult to determine, because much of the work at the intersection of the physical and life sciences overlaps the interests of a variety of programs and agencies but does not fit squarely within any single funding program. The challenges exist not only between agencies, but also between divisions within an agency and between funding programs within the same division.
Such silo effects are certainly not limited to research at the intersection of the physical and life sciences but are common in any type of crossdisciplinary investigation. Hence if solutions can be identified that improve support for research at the interface between biology and physical sciences, these solutions could serve as models to improve the climate at other interfaces such as that between basic biology and medicine.
The differences in level of support among federal funding agencies, and the tendency to fund canonical research rather than research at the disciplinary interface, has led to the perception that one area is more critical than another. While federal support is indeed limited, research funding is not always a zero-sum game. The committee hopes for renewed focus on supporting and conducting the best science, including that which crosses traditional boundaries. One manifestation of this hope would be enhanced collaboration between agencies with a greater number of joint programs than at present. Another recent report makes a similar point for realizing opportunities within the life sciences: cross-agency collaboration is essential for supporting the needs of science and society (National Research Council, 2009).
Federal funding agencies and private sponsors of research have supported research at the intersection of the physical and life sciences for several decades. Such support usually has been for highly specific programs and almost always for programs contained within a single agency.
The NSF has played an important role in pioneering such research. Its molecular biophysics program, administered by the Division of Molecular and Cellular Biosciences, has funded research at this interface since the late 1960s and 1970s and has had a seminal impact. This program fostered the development of techniques such as X-ray crystallography and nuclear magnetic resonance before they became common, and later expanded to support research on theory and simulations. The Physics of Living Systems program, administered by the NSF Division of Physics, evolved from the earlier biological physics programs and supports scientists using the tools of physics to study biological problems at the molecular level. Of the 11 Physics Frontiers Centers, 2 have biological physics as their focus.
NIH’s National Cancer Institute recently initiated a program under its Center for Strategic Scientific Initiatives. This program, which seeks to promote the types of collaborations proposed in this report, addresses outstanding issues in research on cancer. Begun in 2008, it has conducted three workshops and is proposing to fund four, five, or six centers for 5 years, at an annual budget outlay of $15 million to $20 million. The program has the potential to provide valuable guidance on the encouragement and support of cross-disciplinary work in the life and physical sciences.
Perhaps the most interesting federal support for such research has come from the Defense Advanced Research Projects Agency (DARPA). DARPA has funded innovative projects in such areas as the detection of infectious disease agents and methods for rapid development and deployment of novel therapeutics against infectious disease, a number of which involved work at this intersection. DARPA support has some features that differentiate it from other federal support. For example, it provided extramural funding directed to targeted objectives, often involving large grants over a relatively short time. Reviews were carried out at a single location with rigorous security measures. As a result, investigators seemed more willing to expose novel ideas. Reviewer workloads were lighter than for comparable study sections at other agencies, meaning that there was more time for the review panel to discuss the pros and cons of particular projects in detail. Because DARPA program managers also had considerable discretionary authority in making funding choices, agency funding priorities could be shaped by the professional staff.
High-risk research at the intersection of the physical sciences and life sciences seems to have received a more encouraging welcome from the private sector than from the federal sector. Perhaps this is not surprising as small companies supported by venture capital are accustomed to supporting potentially transformative research whose failure rates approach 90 percent. Examples of successful work at this intersection include the development of technology platforms for diagnostics and for therapeutics. Ongoing efforts include the development of new tools for the analysis or manipulation of biological systems such as high-resolution optical microscopy, and cryo-electron microscopy and tomography. Such platforms inevitably require interdisciplinary teams of biological and physical scientists, engineers, and software developers, as well as a willingness to try unproven technologies. It is impossible to conceive of most federal support mechanisms taking such risks, but the committee hopes that the federal government will devote at least a small portion of its funds to potentially transformative research at the intersection of the physical and life sciences.
In addition to the need for more funding for research at the intersection of the physical and life sciences, there is a need to assess what changes should be made in the mechanisms for administering programs that support such research
so that those funds will be utilized more efficiently. How can interdisciplinary and multidisciplinary proposals be appropriately ranked in competition with single-discipline proposals? What new models are needed to support research that spans funding programs and agency boundaries?
Current support mechanisms for interdisciplinary work require that research proposals be submitted to specific funding agencies and often require acknowledging when simultaneous submissions are being made to other funding agencies. In times of tight funding it is hard for one agency’s reviewers to avoid downgrading an application submitted to more than one agency in favor of a proposal that has no other possible sources of support. A simple solution would be to allow submissions to multiple agencies without prejudice during the review process—for instance, information about pending support might not be revealed in the prospectus being considered by review panels.
A more meaningful solution would be to establish crossdisciplinary funding that spans agency divisions or even entire agencies. These joint programs would need to undergo a single, cross-agency review process rather than independent, parallel-review procedures at each participating agency. One example of such a joint program is in the area of mathematical biology and is being supported by both NSF and NIH (see Box 6-2).
It is, admittedly, more difficult to evaluate interdisciplinary research proposals. The community of established reviewers whose skills are well-founded in a pair of disciplines is small. People in one discipline will usually favor that field since it is what excites them, it is what they most easily understand, and it is where they can most easily recognize rigor and innovation. And having separate reviews by people with expertise in each of the respective areas and then merging the scores will disadvantage the investigator, as each of the disciplinary experts is likely to undervalue the proposal. In the private sector, where multidisciplinary research is more common, evaluations are prolonged, individuals with many perspectives are involved, the number of proposals is relatively small, and a project is revisited and reevaluated continually. These characteristics work well but are not a solution for federally funded work, because the sheer volume of proposals would make them impractical.
The committee recognizes that successful programs for funding innovative interdisciplinary science exist within funding agencies, several of which were described above. However, opportunities are lost due to lack of cooperation between agencies to make programs that are more than the sum of the parts. Therefore, the committee calls on the White House and its Office of Science and Technology Policy (OSTP) and Office of Management and Budget (OMB) to develop standing mechanisms that will facilitate, rather than impede, interagency collaborations. In particular, in Recommendation 2 the committee calls upon OSTP and OMB to work through the National Science and Technology Council (NSTC) to establish
NSF/NIH Joint Program in Mathematical Biology
The Directorate for Mathematical and Physical Sciences at the NSF and the National Institute of General Medical Sciences at the NIH have established a joint program in mathematical biology.1 The goal of the program is to engage practicing mathematicians in the core of biomedical research.
Proposals are reviewed by a single review panel that incorporates expertise from both the life sciences and mathematics, ensuring that candidate proposals incorporate rigorous mathematics and engage substantive biological questions. Most successful proposals demonstrate a clear commitment to substantive collaboration between one or more biologists and one or more mathematicians. The joint review panel considers review criteria for both NIH and NSF so that it is not necessary to conduct separate reviews at each agency.
Both agencies must sign off on each award, but the grants are ultimately awarded by either NSF or NIH and subject to the award requirements at that agency. The decision on which agency makes the award is at the option of the agencies, so investigators apply to a single program.
The success of the program is helped by the commitment of both NIH and NSF and the involvement of program and review staff from both agencies. The agencies have also helped to build a community of scholars in the program; they organized a meeting of principal investigators in 2003, and another is being considered in 2009.
a standing interagency working group on multidisciplinary research under the NSTC’s Committee on Science.
RECOMMENDATION 2. The Office of Science and Technology Policy (OSTP) and the Office of Management and Budget (OMB) should develop mechanisms to ensure effective collaboration and cooperation among federal agencies that support research at the nexus of the physical and life sciences. In particular, OSTP and OMB should work with federal science agencies to establish standing mechanisms that facilitate the funding of interagency programs and coordinate the application and review procedures for such joint programs. Moreover, the National Science and Technology Council should establish a standing interagency working group on multidisciplinary research within its Committee on Science, with focus on the intersection of the physical and life sciences.
Interagency Working Group on Plant Genomes
The Interagency Working Group on Plant Genomes was established in May 1997 under the direction of the NSTC’s Committee on Science to pursue crop genomics in the public sector. The group’s National Plant Genome Initiative (NPGI) focused on developing genomics tools that would transition discoveries in model plants such as Arabidopsis to crop species.
The group was charged to “(1) identify science-based priorities for a plant genome initiative; and (2) determine the best strategy for a coordinated Federal approach to supporting such an initiative, based on respective agency missions and capabilities” (National Research Council, 2008a, p. 16). Since its founding, the initial group of participating agencies has expanded and the coordinated NPGI now incorporates most of the federal investments in plant genomics. Although small in overall investment, the U.S. continues to lead the world in the productivity of plant science research.
There are several successful models of NSTC working groups and subcommittees charting a coordinated research agenda, including those focused on plant genomes and global change research. As described in Box 6-3, the former provides a coordination function in the area of plant genomics. The latter subcommittee serves the coordinating body for the U.S. Global Change Research Program, which is made up of 13 federal department and agencies and includes an integration and coordination office to implement the program’s strategic plan.
Following the model of the NPGI (see Box 6-3), the interdisciplinary working group on multidisciplinary research should begin by identifying all of the agencies with an interest in research at the physical and life sciences and overcoming the barriers to multidisciplinary connections pointed out in this report. The group would then be charged with developing multiagency solicitations and review procedures that would support research that falls between existing government funding programs.
SUPPORTING TRANSFORMATIVE RESEARCH
Although transformative research has tended to occur at the boundaries of existing disciplines, cross-disciplinary proposals often have difficulty surviving the review process intact. Most U.S. research funding is based on the prospective analysis of a research proposal and significant preliminary results are almost required as evidence that the proposed plan can succeed. In fact, there is often a sense that
investigators must cite their past research in proposals that support their future research. In an era of limited resources, it is perhaps natural for review committees to favor proposals that are likely to succeed—that is, relatively conservative proposals that extend the boundaries of past research incrementally—and to avoid taking chances on proposals that might be transformative but that also have a high risk of failing.
The challenge of supporting high-risk, high-reward research has drawn significant attention in recent years. For example, the recent ARISE report from the American Academy of Arts and Sciences recommended that all federal agencies should have programs that focus on supporting innovation with relatively simple and rapid application processes and called for an evaluation of such programs to ensure that they are, indeed, supporting at least some transformative research (AAAS, 2008, p. 36). ARISE also recommended that funding mechanisms and review processes should “nurture, rather than inhibit, potentially transformative research” by tweaking review criteria, providing more flexibility and resources for agency program managers to support exploratory projects, and by establishing interdisciplinary review panels to consider high-risk research proposals across fields.
Not only the research community but also the funding agencies themselves have called for change to support transformative research. For example, the National Science Board, which establishes the policies of the NSF, called for an NSF-wide Transformative Research Initiative that would be distinct from other programs and allow the NSF to establish new structures and procedures for transformative research (National Science Board, 2008). Similarly, an NIH initiative on enhancing peer review has recommended that at least 1 percent of investigator-initiated research awards be directed to transformative research programs. The self-study also recommended an analysis of interdisciplinary research applications to determine how they were assigned to review and how successful they were in obtaining funding, and recommended an editorial board model for review of interdisciplinary research (National Institutes of Health, 2008).
Several ongoing programs, most of them operating outside the normal peer review system, have the goal of supporting transformative research. Each of these has clear relevance to research at the intersection of the physical and life sciences. For example, since 2004, NIH has supported “individual scientists of exceptional creativity who propose pioneering—and possibly transforming approaches—to major challenges in biomedical and behavioral research”1 with the NIH director’s Pioneer Award. Based on a special review process, the program places greater emphasis on the investigator than most and even includes in-person interviews with the finalists. While the program is too new to have been formally evaluated, it
NIH Director’s Pioneer Award overview, found at http://nihroadmap.nih.gov/pioneer/.
is striking that a large fraction of Pioneer awardees work at the interface of biology and physics and that several are physical scientists, given that NIH supports few physical scientists overall. Several NIH institutes, including the National Institute of General Medical Science, have made one round of awards for Exceptional, Unconventional Research Enabling Knowledge Acceleration (EUREKA) grants. The proposed research is expected to have a substantial impact on a significant fraction of the scientific community. As with the Pioneer award, the EUREKA application and review process emphasizes significance and innovation in addition to experimental approach and other considerations.
The Howard Hughes Medical Institute (HHMI) has provided leadership in recognizing the importance of and funding work at the physical/life sciences intersection. Since 1990, HHMI has held national competitions for investigators, who then become HHMI employees while retaining their faculty positions and their laboratory location at their university or research institute. By supporting people, not projects, HHMI rewards retrospective evidence of innovative contributions more than prospective analysis of research plans both in its initial appointment of investigators and in the 5-year reviews of their appointments for renewal. Interdisciplinary work is overtly sought and valued, and a number of HHMI investigators are chemists, physicists, computer scientists, and engineers who are tackling biological or biomedical problems. Some of this multidisciplinary and interdisciplinary work now is supported in a dedicated facility HHMI recently established known as the Janelia Farm Research Campus (see Box 6-4).
To enhance the opportunity for potentially transformative research, the committee makes recommendations designed to make proposals from more than one principal investigator (PI) more common. The National Science and Technology Council and its Research Business Models Subcommittee have advocated enhancing opportunities and providing uniformity in evaluating proposals with more than one PI. The committee feels that there is room for further enhancement and hopes that multi-PI proposals and awards will become easier to achieve because they directly promote projects with equal participation of researchers from more than one discipline.
RECOMMENDATION 3. Federal and private funding agencies should enhance the ability of more than one researcher to serve as principal investigator (PI) on research projects. Each PI should receive full credit for participation on the grant, with the lead PI serving as the administrative contact.
By this recommendation, the committee does not intend to require that most or even many grant programs only provide multi-PI awards, as there is a need for the continued support of single-investigator multidisciplinary research. Rather, the optimal models for funding would include a mix of single-PI, two-PI, and multiple-PI research activities—with the particular organization dependent on the
HHMI’S Janelia Farm Research Campus
Starting in 2000, HHMI tried to discern which sorts of basic biomedical research, if any, were challenging to support through its university-based investigators program—that is, which types of research needed a new model in order to be fully realized. This analysis led to the conclusion that multidisciplinary research at universities was often frustrated by departmental decisions and prompted HHMI to build a free-standing, wholly supported multidisciplinary research institute.
The HHMI Janelia Farm Research Campus opened in 2006 in the northern Virginia suburbs of Washington, D.C. By 2010, it will house 44 research groups, each capped at a small size (six researchers in the laboratory of a “group leader,” two in the laboratory of a fellow). The small group size encourages frequent interaction and collaboration and is inspired by the small teams that contributed to the success of Bell Labs.
One of Janelia Farm’s initial research emphases is mapping and understanding the neural circuits responsible for all complex behavior. This area of research provides an interactive environment, bringing together a variety of investigators, including neuroscientists, physicists, computer scientists, and chemists. Additional information about Janelia Farm is available at http://www.hhmi.org/janelia/.
specific research. The critical need, at this time, is to break down the administrative barriers that prevent scientists from assembling in the most effective way to secure extramural funding and in conducting research.
The committee also recommends that federal and private funding agencies explicitly support potentially transformative research. By their nature, such programs should incorporate application and review procedures that are consistent with multidisciplinary research and incorporate the viewpoints from a variety of scientific disciplines. The committee also feels that these programs and any others for cross-disciplinary research should be continually assessed to be sure that they are meeting those goals.
RECOMMENDATION 4. Federal and private finding agencies should devote a portion of their resources to support potentially transformative research, including opportunities at the intersection of the physical and life sciences. These sponsors should have peer review procedures that incorporate the viewpoints of scientists from a variety of disciplines. Moreover, they should continually assess the effectiveness of these grant programs and the review procedures to ensure that they are meeting the desired aims.
EDUCATING SCIENTISTS AT THE INTERSECTION OF THE PHYSICAL AND LIFE SCIENCES
Pursuing research at the intersection of traditional disciplines will require a workforce prepared to work at the boundaries between disciplines. This will mean changing the way we educate the next generation of scientists.
A number of reports in the last decade call for enhanced quantitative training of life scientists and say it will be a critical need for the life sciences in the future (e.g., National Research Council, 2003, 2005a, 2005b, 2008). Because research in the biological sciences is becoming increasingly quantitative, a greater ability to model biological phenomena using mathematical language is needed. Moreover, the vast collection of data now available to researchers through such fields as genomics has introduced a complexity and need for data analysis not previously relevant in the life sciences. Many of these changes have happened within the last decade, leaving even some life sciences faculty members ill equipped.
The committee considered the best mechanisms for empowering life scientists with the appropriate degree of mathematical sophistication and for providing educational settings in which students want to learn the math required to address problems of interest. The committee could have simply recommended that all students in the life sciences take one or more classes in advanced mathematics. But this would just produce students with a background in biology and mathematics but no assurance that they would be able to apply the mathematics they learn in one class to the biology they learn in another. Rather, biology and mathematics should be treated together and it would be even more effective if the physical sciences were also included. The committee is hopeful that the NSF program for interdisciplinary training for undergraduates in the biological and mathematical sciences2 will reveal best practices that will apply to a broad range of students.
Collaborative teaching efforts by life scientists and applied mathematicians, physicists, or engineers may be the best way to provide biology students with the quantitative and problem-solving skills they need and also help to bridge the physical and life sciences in the classroom setting. Such courses can also provide students with a model of how scientists and mathematicians approach problems, demonstrating the relevance of multidisciplinary approaches and the need for mathematical sophistication.
Enabling Interdisciplinary Research Starting at the Undergraduate Level
Enhanced quantitative training for biologists is an important first step in fostering researchers who can work at this intersection, but it is only one step. It will
Solicitation 08-510; http://www.nsf.gov/pubs/2008/nsf08510/nsf08510.htm.
also be critical to increase the exposure of physical scientists to the life sciences and vice versa.
The most basic way to realize this goal would be to have all physical science undergraduates study the principles of biology such as genetics and evolution and to have all biology majors receive appropriate preparation in physics, chemistry, and mathematics. In fact, biology majors at many undergraduate institutions have long been required to take specific courses in these other departments, although often there is no attempt to make these courses relevant to life science majors. However, this approach reinforces the impression that the scientific disciplines are discrete, isolated entities and leaves it to the student to draw connections about the relevance of other disciplines to biology.
An alternative approach, which is harder to establish but probably would have a more lasting impact, is to integrate applications, examples, and problems from other disciplines into core courses to increase relevance. For example, quantitative aspects could be given more emphasis in existing biology courses and materials from the life sciences could be incorporated into existing physical science and mathematics classes. It is just such an integration that a recent NRC report proposed (National Research Council, 2003).
More pertinent would be a single introductory course that incorporates elements of both the physical and life sciences and would introduce biology and physics majors to the basics of both disciplines. Although a small number of institutions offer a common introductory course (see Box 6-5 for an example), such experiments are far from universal. The committee encourages more institutions to design such courses while recognizing that developing such courses requires addressing a complex and at times conflicting set of goals. Any such course will need to ensure that the knowledge bases of the respective disciplines are presented coherently and at an appropriate depth. At the same time, the illustrations and thematic questions that attempt to integrate the various approaches to similar questions must be clearly presented and relate back to the original disciplines. Finally, time limitations undoubtedly will require making difficult decisions on which topics covered in traditional courses will be diminished or treated in a different way.
It also is important that these classes not be taught in isolation from the rest of the curriculum but be integrated into the set of courses offered by all involved departments. To meet these needs, faculty members from those departments must meet regularly to coordinate curriculum planning. Funding agencies also can play a role in these education efforts by not only facilitating the development of these courses but trying to evaluate the effectiveness of such strategies in increasing student familiarity and knowledge of both the physical and life sciences.
Introductory Interdisciplinary Science at the Evergreen State College
The Evergreen State College in Olympia, Washington, offers a 1-year-long interdisciplinary program entitled “Introduction to Natural Science: Life, the Universe, and Everything.” The course brings together unifying perspectives from physics and chemistry to provide a conceptual and experimental introduction to natural science. It takes a thematic approach, focusing on cycles and transformations of matter and energy in both living and nonliving systems, which allows students to see similar ideas emerging at a variety of levels.
The course is team-taught and always involves a chemist and a biologist, with a third faculty member from areas such as physics, computer science, or geology. Different areas of science are integrated throughout the course, including into exams that test knowledge in more than one subject area—especially areas that bring the disciplines together.
Enrolling approximately 100 students per year, the course combines lectures, problem-solving activities, laboratories, field trips, seminars, the reading of primary research literature, and independent scientific investigations by small groups in collaboration with one of the faculty members. It serves as preparation for more advanced courses in the physical and biological sciences, as well as in the health and environmental sciences.
RECOMMENDATION 5. At the undergraduate level:
Universities should establish science curriculum committees that include both life scientists and physical scientists to coordinate curricula between science departments and to plan introductory courses that prepare both those who would major in the life sciences and those who would enter the physical sciences.
Professional scientific societies should partner with peer societies across the life and physical sciences to organize workshops and provide resources that will facilitate multidisciplinary education for undergraduates.
Federal and private funding agencies should offer seed grants to academic institutions to develop new introductory courses that incorporate both the physical and life sciences and to professional societies for organizing workshops and developing resources for multidisciplinary education. They should also support research to identify best practices in such education.
The committee acknowledges that adding material to a curriculum or to an individual course may require institutions and faculty to make difficult decisions
about which existing topics to eliminate or treat in a different way. It is beyond the scope of this report to define the specific courses and subjects for a curriculum but the NRC report Bio2010: Transforming Undergraduate Education for Future Research Biologists may assist institutions in considering these issues (National Research Council, 2003).
Integrating Life and Physical Sciences for Graduate Students and Postdoctoral Researchers
Undergraduate training that brings together the life and physical sciences would be an important foundation for expanded multidisciplinary connections at the graduate and postdoctoral levels. Research habits of mind and the socialization of budding scientists as full members of the research community develop during Ph.D. and post-doctoral training. Ensuring their greater involvement in research across disciplinary boundaries and regular interaction with scientists from a variety of disciplines will make them comfortable with multidisciplinary research early in their careers.
As one example of supporting graduate training in emerging interdisciplinary fields, HHMI has partnered with the NIH to establish the Interfaces Initiative, which provides training grants to institutions to develop interdisciplinary graduate programs (see Box 6-6). Despite the promising experiment begun by HHMI and NIH, most graduate programs do not offer students an easy opportunity to work with researchers across the life–physical sciences interface. Most departments offer either their own doctoral program or an umbrella program that does not span the divide between the life and physical sciences.
Federal and private funding agencies provide support for a large number of doctoral students and postdoctoral researchers in the life and physical sciences. Many of the trainees are supported as research assistants on research grants awarded to principal investigators, while others are part of institutional training grants or supported by individual fellowships. The committee sees a role for leveraging this extramural support to encourage interdisciplinary training that spans the life and physical sciences.
RECOMMENDATION 6. Federal and private funding agencies should offer expanded training grants that explicitly include graduate students and post-doctoral researchers from fields across the life and physical sciences and that require the involvement of academic departments from both the physical and life sciences. Funding agencies should also offer administrative supplements to existing research grants that would enable a principal investigator in the life sciences to support a postdoctoral researcher with a background in the physical sciences, or vice versa.
HHMI-NIBIB Interfaces Initiative
The HHMI and the NIH’s National Institute of Biomedical Imaging and Bioengineering (NIBIB) have jointly developed and supported an interdisciplinary graduate research training program: the HHMI-NIBIB Interfaces Initiative. First awarded in 2005, the initiative’s 4-year training grants were established with the goal of teaching graduate students to work effectively across disciplinary lines. This initiative takes advantage of HHMI’s ability to catalyze the creation of new university programs and the ability of NIBIB to sustain such programs once formed.
The initiative supports institutional training grants rather than fellowships to individual predoctoral students because institutional grants require a greater degree of interaction by faculty from diverse fields. Faculty cooperation, in turn, can help drive institutional change and ongoing connections between disciplines.
Among the programs that have been supported are those with a focus in mathematical, computational, and systems biology (University of California, Irvine); multiscale analysis of biological structure and function (University of California, San Diego); and biophysical dynamics and self-organization (University of Chicago). Additional information about the HHMI-NIBIB Interfaces Initiatives is available at http://www.hhmi.org/grants/institutions/nibib.html.
The committee sees value in encouraging those who have received a doctorate in a traditional discipline, not just scientists training in interdisciplinary graduate programs, to consider applying their knowledge to research questions at the intersection of the physical and life sciences, and to give them a jump start on their research careers. One prominent program in this area is the Career Awards at the Scientific Interface program of the Burroughs Wellcome Fund (BWF), which has been offered since 2002 (see Box 6-7). The NIH has created the Pathway to Independence Award, which offers a similar two-phase postdoctoral/faculty award but with no specific consideration for physical scientists who have little background in the biomedical sciences. Career transition programs such as the BWF Career Awards and the NIH Pathway to Independence can encourage institutions to establish positions for creative early-career scientists that might not have otherwise existed. They also help young scientists to become established in their careers by reducing the funding pressure on them and fostering their transition to independent research careers (National Research Council, 2005c).
Several of the other career development awards offered by the NIH support scientists who propose to train in a new field. For example, the Mentored Research Scientist
Burroughs Wellcome Fund Career Awards at the Scientific Interface
The Career Awards at the Scientific Interface (CASI) program, sponsored by the Burroughs Wellcome Fund (BWF), offer 5 years of funding that bridges advanced postdoctoral training with the first 3 years of faculty service. The program accepts applications from individuals early in their postdoctoral careers who have a Ph.D. in mathematics, physics, chemistry, computer science, statistics, or engineering and whose work addresses biological questions. The program provides 1 to 2 years of salary and research support during the postdoctoral portion of the award, with the balance of the 5 years of support once the recipient has assumed a faculty position.
The program is unique in providing advice regarding career development to the fellows: Each awardee is counseled on the terms of the faculty offers he or she receives and is given advice on how that offer compares to others given to former fellows. The majority of the CASI fellows have gone on to top junior faculty positions and are developing into leaders in interdisciplinary fields such as systems biology and computational biology. More information about this program can be found at http://www.bwfund.org/pages/129/Career-Awards-at-the-Scientific-Interface/.
Development Award (K01) from the National Human Genome Research Institute is open to individuals with degrees in computer sciences, mathematics, chemistry, engineering, physics, and closely related scientific disciplines. The Mentored Quantitative Research Career Development Award (K25) from the NIBIB supports scientists and engineers with little or no experience in medicine or the life sciences to develop the relevant research skills that will allow them to conduct basic or clinical biomedical imaging or bioengineering research.
A number of additional programs support early-career scientists by providing an infusion of funds early in their research careers—but without an explicit career transition element. Proposals for these awards are often relatively short and frame a set of research questions, rather than a detailed experimental plan that contains significant amounts of preliminary data. The NIH director’s New Innovator Award complements the NIH director’s Pioneer Award (discussed in a preceeding section) and gives 5 years of funding that emphasizes innovation, with a focus on scientists who received their most recent doctoral degree 10 or fewer years ago but who have not yet received an R01 award.3 Appropriately, this program has included a number of scientists working at the intersection of the life and physical sciences. NSF’s
More information about this program can be found at http://nihroadmap.nih.gov/newinnovator/.
Faculty Early Career Development (CAREER) program provides 5 years of support to tenure-track but untenured faculty. Among the private-sector programs that currently or historically have supported new investigators are the Markey Scholar Awards from the Lucille P. Markey Charitable Trust, the Keck Distinguished Young Scholars in Biomedical Research from the W.M. Keck Foundation, the David and Lucille Packard Foundation Fellowships for Science and Engineering, the Beckman Young Investigator Program from the Arnold and Mabel Beckman Foundation, the Pew Scholars Program in Biomedical Sciences from the Pew Charitable Trusts, the Searle Scholars Program from the Kinship Foundation, the Damon Runyon Cancer Research Foundation Scholar Award, the Sloan Research Fellowships, the McKnight Scholar Awards from the McKnight Endowment Fund for Neuroscience, and the Klingenstein Fellowship Awards from the Esther A. and Joseph Klingenstein Fund. In addition, several universities offer time-limited independent research fellowships for early-career investigators, including the Carnegie Institution for Science, the Whitehead Institute for Biomedical Research, the San Francisco and Berkeley campuses of the University of California, and Harvard University.4
All of these awards that occur early in a scientist’s career can be critical to encouraging new faculty to take on innovative projects in multidisciplinary areas and should be encouraged. Despite these examples, however, the number of grant opportunities that support scientists switching fields of study is quite limited in both federally supported and private-sector programs.
RECOMMENDATION 7. Federal and private funding agencies should offer opportunities for both early-career and established investigators trained in one discipline to receive training in another and apply their experience and training to interdisciplinary problems. In particular, postdoctoral career awards should be established that facilitate the transition of a candidate prepared in a physical science field to apply that training to important questions in the life sciences and vice versa. Funding agencies should also provide expanded support for experienced investigators to receive training in a new field, perhaps in the form of sabbatical fellowships.
This recommendation is similar to those that have been offered before in reports including those of the NRC (National Research Council, 2005c), and the National Academies (2007).
For more information about these and other programs focused on new investigators, please see Chapter 2 of National Research Council, 2005c.
American Academy of Arts and Sciences, 2008. ARISE: Advancing Research in Science and Engineering: Investing in Early-Career Sciences and High-Risk, High-Reward Research. Cambridge, Mass: American Academy of Arts and Sciences.
National Academies, 2007. Rising Above the Gathering Storm: Energizing and Employing America for a Brighter Economic Future. Washington, D.C.: The National Academies Press.
National Institutes of Health, 2008. 2007-2008 Peer Review Self-Study, Final Draft. http://enhancing-peer-review.nih.gov/meetings/NIHPeerReviewReportFINALDRAFT.pdf.
National Research Council, 2003. Bio2010: Transforming Undergraduate Education for Future Research Biologists. Washington, D.C.: The National Academies Press.
National Research Council, 2005a. Advancing the Nation’s Health Needs: NIH Research Training Programs. Washington, D.C.: The National Academies Press.
National Research Council, 2005b. Mathematics and 21st Century Biology. Washington, D.C.: The National Academies Press.
National Research Council, 2005c. Bridges to Independence: Fostering the Independence of New Investigators in Biomedical Research. Washington, D.C.: The National Academies Press.
National Research Council, 2008a. Achievements of the National Plant Genome Initiative and New Horizons in Plant Biology. Washington, D.C.: The National Academies Press.
National Research Council, 2008b. Inspired by Biology: From Molecules to Materials to Machines. Washington, D.C.: The National Academies Press.
National Science Board, 2008. Enhancing Support of Transformative Research at the National Science Foundation. NSB-07-32. Arlington, Va.: National Science Foundation.