Preaching Convergence

Greater cross-disciplinary research efforts among the biological, physical, and engineering sciences are needed to accomplish the U.S.’s scientific and technological goals, according to a group of scientists from Massachusetts Institute of Technology. To move toward a “true disciplinary integration,” the group says it has developed a convergence model of research, which it defines as “the merging of distinct technologies, processing disciplines, or devices into a unified whole that creates a host of new pathways and opportunities.”

The MIT scientists made their case for this new research model at a forum held by the American Association for the Advancement of Science (AAAS) early this month. They presented a white paper with ideas and recommendations on what it would require, in terms of federal and academic support, for this research model to be effective in biomedical science. Speakers at the forum discussed how this model would affect federal science and engineering research funding and policy decisions. Others argued that what the MIT scientists propose is already happening.

“What we’re talking about is new concepts moving into the field of biology,” said Phillip A. Sharp, a professor of biology at MIT, who noted that the white paper cites biomedical research as the first area to benefit from the convergence of disciplines. “This includes the concepts of information theory, dynamic stability, and noise control, all of which are concepts common to engineering disciplines that are moving into life sciences.”

Other scientists in the group proposing the model gave examples of how convergence is being applied to biomedical research at MIT’s David H. Koch Institute for Integrative Cancer Research. Paula T. Hammond, a professor of chemical engineering, described the multidisciplinary efforts that went into the invention of a device to detect metastasizing cancer cells in human blood. Tyler Jacks, a professor of biology and director of the Koch Institute, spoke about the variety of convergence research projects in such areas as nanotechnology-based detection methods, new imaging agents for disease detection, and development of mathematical models to simplify complex biological phenomena.

The MIT paper cites the need for higher and more consistent federal support for research. Most biomedical research is supported by the National Institutes of Health, which has seen below-inflation budget increases over the past few years. As a result, NIH’s grant success rate remains low, making it challenging for new work to receive funding. But for the research that does get support, the paper claims, NIH’s investment has a significantly high rate of return to society in terms of saving and extending lives. According to the paper, the aging population of the U.S. makes the need for such investments even more critical.

Members of the government’s biomedical establishment who spoke at the forum were impressed with the progress MIT is making with increasing integration across research disciplines but took the view that convergence is already happening in some universities and federal agencies. They also noted some wrinkles in supporting cross-disciplinary work that still need to be ironed out, such as tenure, funding, and training, for the model to work.

“We are certainly aware of the call for NIH to do more work across its institutes,” Alan Guttmacher, director of the National Institute of Child Health & Human Development, told the forum. NIH, he said, is already moving in that direction with its work in translational research and in programs such as the National Children’s Study, which will require collaboration from many different disciplines. Still, Guttmacher noted that universities, where most of this research is performed, have a problem finding ways to reward and advance scientists who participate in convergence research.

Thomas Kalil, deputy director for policy at the White House Office of Science & Technology Policy, agreed that convergence-type research is important and told the forum that government policymakers should be considering the arguments made in the MIT paper when deciding how to fund research to meet the nation’s problems.

Guttmacher also sees the lack of a mechanism for funding of convergence research as a stumbling block. “A lot of people are in favor of innovative funding models—as long as you continue to fund their research,” he said. “This is going to require taking stock of our funding process and having everyone step back from their own self-interest a bit.”

Alan I. Leshner, chief executive officer of AAAS, said that the convergence model will “threaten the hell” out of the traditional academic disciplinary structure and funding mechanisms with which scientists are familiar.

One part of the funding structure that needs to change to support convergence is peer review. The MIT white paper is critical of NIH proposal reviews as becoming increasingly conservative and more selective. The present system discourages innovation and novel ideas in favor of just incremental progress in science, the paper states. Another problem with peer review of convergence research is finding reviewers with the necessary expertise.

To address these problems, the white paper recommends reforming the peer review process. For example, the authors say NIH needs more diversity in its peer review teams, which would increase “both the cutting-edge disciplinary depth and the array of disciplines represented.” This would make approval of convergence proposals more likely, the MIT researchers contend.

Other recommendations in the white paper include establishing a formal mechanism for building connections across stovepiped science-mission agencies within the government to establish a “convergence ecosystem” and encouraging universities to start teaching the next generation of convergence researchers through new curricula, apprenticeships, and training programs.