Skip to content

COVID-19 is an emerging, rapidly evolving situation.

Get the latest public health information from CDC:
Get the latest research information from NIH:
Get the latest shareable resources on coping with COVID-19 from NIMH:

Post by Former NIMH Director Thomas Insel: Roads Not Taken

By on

Julius Axelrod was one of NIMH’s greatest scientists and mentors for five decades until his death in 2004 at age 92. In addition to his many discoveries - which led to his 1970 Nobel Prize - Julie, as he was known, was famous for his aphorisms. One was his saying that “98 percent of the discoveries are made by 2 percent of the scientists.” While this might sound arrogant or elitist to some, data emerging over the past few years support the notion that much of our scientific endeavor involves following the herd and, importantly, that the herd grazes on not much more than 2 percent of the available land.

In a paper published last year in Nature entitled “Too many roads not taken,” Aled Edwards and his colleagues provide a nice example of this instinct.1 The class of enzymes called kinases are critical biological gatekeepers, and many are potentially involved in diseases of the brain. Although there are over 500 protein kinases known from the human genome, about 65 percent of the 20,000 kinase papers published in 2009 focused on the same 50 that were being studied in the early 1990’s. This narrow focus has persisted after many of the unexplored kinases showed up in unbiased genomic screens as potentially related to diseases.

The narrow focus of our science is even more conspicuous in this age of “-omics” – when we can study all of the genes, all of the transcripts, or all of the proteins without a hypothesis. Although we thought the fields of genomics, transcriptomics, and proteomics had defined the universe of roads not taken, recent publications from the ENCODE project reveal a dramatic expansion of this universe.2 ENCODE is a landmark effort that is mapping the working parts of the human genome and exposing just how limited our exploration has been to date. Researchers have focused mostly on the 2 percent of the genome that belongs to genes, but we now know that 80 percent of the genome is translated. In fact, the genome codes for a range of important biological signaling molecules, many of which are still being identified.

For the past 50 years, NIMH researchers have focused on a few pages of this vast text, assuming that dopamine and serotonin were most of what we needed to know about the biology of mental disorders. Where we have been is akin to colonial North America where the overwhelming majority of the population remained on a small fraction of the continent and the Midwest and West were frontiers explored by very few.

How do we encourage exploration of the vast frontiers of biology? How do we nudge the scientific field toward the unknown? In truth, much of what we currently do reinforces herd behavior. NIMH funding is guided largely by a system of peer review, and peer review tends to reward the familiar or, at best, small steps. But the problem is much more fundamental than this. Scientific training is based on an apprenticeship model, with the focus placed on following a mentor, not on breaking for a frontier. Furthermore, academic success requires publishing, which is most easily accomplished by remaining where the tools are good, the field is safe, and the territory is familiar. Not all kinases, for example, are equally easy to study. Successful scientists know to focus on problems that have a good chance of being solved. As the British biologist Sir Peter Medawar noted, “Science is the art of the soluble.” But the targets that are easiest to study are not always the most important. We have created a world of incentives for looking where the light is, even if that is not where the keys were lost.

This would not matter so much if our problems were not so important to solve. We simply cannot afford to have 98 percent of our scientists looking where the light is, staying within the safe zone. Given this, NIH has been working to build incentives to help attract the most intrepid scientists into the frontiers of science. NIH’s Pioneer Awards and New Innovator Awards are grants for innovative ideas that are opening new areas to research and creating new tools. Importantly, these awards are for the person, not the specific project, encouraging these scientists to pursue novel approaches to important problems. For example, a recent NIMH-supported Pioneer Award will allow Feng Zhang of MIT to develop a new approach to manipulating the genome and epigenome.3

NIMH is also trying to encourage innovation among young investigators through the BRAINS (Biobehavioral Research Awards for Innovative New Scientists) initiative. So far, we have awarded 28 early stage investigators with support to pursue an idea deemed high-risk but high-reward, helping to move them into areas not previously studied. Among the 28, Amit Etkin4 of Stanford is studying the neurobiology of psychotherapy, and Zhaolan Zhou5 of the University of Pennsylvania is defining the epigenetic signature of early life stress.

My own sense is that Julie Axelrod was partly right. There is a small group of scientific leaders who seek out new frontiers and set the pace for the vast majority of the field. But I am not convinced that tells the entire story. Each new breakthrough is based on a foundation laid by hundreds of scientists often working in distant fields. Today we find that discoveries are often the products of groups working together, rather than a lone investigator striving to be in the 2 percent club. At the same time, we do need to make more room for scientific pioneers - researchers who have completely new ideas and are willing to take risks to open up a new area of inquiry.

This need is important to consider during this period of austerity for NIMH funding. Tight budgets are not kind to risk takers. But, when budgets are tight, it is more important than ever to shore up support for the innovators who are not following familiar paths. At NIMH we are using the Pioneer and New Innovator initiatives, our own BRAINS awards program, and a policy of supporting grants that are trying new approaches even when the peer review scores are beyond our nominal payline. If there is any lesson to be learned from the many discoveries of this past year, it is that the continent we need to explore is both larger and more complex than we ever imagined. With so many roads not yet taken, who would want to follow the herd?


 1 Edwards AM, Isserlin R, Bader GD, Frye SV, Willson TM, Yu FH. Too many roads not taken. Nature. 2011 Feb 10;470(7333):163-5. PMID: 21307913

 2 The September 6, 2012 issue of Nature includes multiple articles pertaining to the ENCODE project. Information about the project and related publications are accessible via Nature’s ENCODE web pages:

 3 Additional information about this project may be accessed at:

 4 Additional information about this project may be accessed at:

 5 Additional information about this project may be accessed at: