Skip to main content
Premium Trial:

Request an Annual Quote

On Breeding the Next Generation of Bioinformaticists


Charles DeLisi says, “It’s time to abandon hedgehogs.”

Charles DeLisi is Metcalf Professor of Science and Engineering at Boston University, and also served as Dean of the College of Engineering from 1990-2000.

Earlier, he was Chair of Biomathematical Sciences at Mount Sinai Medical School, Director of the Department of Energy’s Health and Environmental Research Programs, Section Chief at NIH, and Theoretical Division Staff Scientist at Los Alamos National Laboratory. In 1999 he initiated the BU PhD program in Bioinformatics, which currently includes more than 40 doctoral students and 50 faculty from across the university. Among numerous honors, DeLisi was awarded the Presidential Citizens Medal in January by President Clinton for initiating the Human Genome Project.


I refer not to the hedgehog genes that are so important for differentiation, but to the way the word is used in Isaiah Berlin’s The Hedgehog and the Fox. The title of that essay runs deep, with many meanings, the most common taken from Archilochus — the Greek lyric-poet who portrayed the fox as knowing many things, and the hedgehog as knowing one big thing. Thus interpreted, we are not confronted with the choice between training hedgehogs and foxes — specialists and generalists — but with the challenge of creating their first- generation offspring, of educating a new breed of researcher.

The mid-21st century biologist’s use of the computer as a creative analytical engine, rather than a large data repository, will usher in an era of predictive biology. And her studied awareness of the social implications of the new technologies will pave a seamless continuum between university and industrial science.

The foment stirred by the draft human sequences — and the sequences of dozens of other genomes — has propelled us into new territory, only sparsely populated with researchers able to fully exploit the continually increasing data flood. The need to respond rapidly is compelling, but so too is the need to respond thoughtfully.

The educational community will do well to strive for programmatic diversity. Educational programs are almost always multidimensional, and few universities have the resources to afford full coverage of the range of possibilities. Each program must consider such matters as the applied/basic research spectrum; emphasis on development of new computational and mathematical methodologies, as opposed to biological applications; and distribution of resources between professional MS-level training and doctoral-level training. Even universities that have comprehensive programs will likely have differences in emphasis.

What must be avoided is entrainment to a dominant educational motif; the programmatic meme: the repeated replication of programs characteristic of a few prestigious schools. A wide range of responses to the choices before us will provide the nation with the pool of professionals required for a strong science base and a strong economy.

For any particular university, the response should be as dynamic and adaptive as resources and sound education permit. If history is a guide, we can expect to see sequelae to the genomic revolution, reverberations driving changes in science and society at a rate foreign to our past experience. The challenge is to develop programs that can anticipate change, without knowing precisely what shape it will take. This raises a complex set of issues and bears directly on the evolving nature of the relation between universities and industry.

The university’s main mission — to preserve and transmit our intellectual heritage while creating new knowledge that can alter paradigms and shake the foundations of our world view — demands the rigorous scrutiny of new ideas, and discourse that is not temporally constrained, but allowed to run a natural course. As a consequence, the time scale for change is typically longer than it is for most social establishments, and certainly longer than that of technologically based industry. These conditions, which bias in favor of developing a deep understanding of the world around us, are also fully commensurate with the best and most effective curricula.

Educational programs that track technology closely are costly, unstable, and fundamentally flawed, preparing students for the past rather than the future. Productivity in a world of technologies that cannot be anticipated is best prepared for by acquiring general problem-solving skills, by becoming adept at methods applicable in a wide range of contexts, by mastering the fundamental principles that provide the basis of all current technologies as well as those to be developed during the next several decades, and by attaining a mastery of fundamentals sufficient to allow self education. (Less easily taught is the ability to ask questions, the importance of which is conveyed by Picasso’s quip, “The trouble with computers is they only give questions.”) On those goals most people, I believe, will agree.

But we are now caught in a sea change that has produced a flood of new data, while we are nearly bereft of the large cadre of creative talent needed to convert it to knowledge at a rate commensurate with its importance. Under these circumstances, and indeed under most circumstances that we can foresee, some coupling between universities and industry will be necessary if higher education is to meet its obligation to society.

University science and engineering programs can be industrially linked in a number of ways: through advisory boards that have strong industrial representation; by offering industrial internships and rotations to students; by collaborations between faculty and industrial researchers who jointly mentor students; and by appointing industrial colleagues as adjunct faculty. Our own program uses these methods and others to ensure a flow between fundamental discovery and commercial products, and sound preparation for industrial leadership by those students who choose an industrial career.

When research projects are chosen with care, and appropriate safeguards are developed to minimize conflicts of interest, collaborative relations can broaden a student’s understanding of the relation between fundamental and applied research, and deepen her awareness of the sometimes tortuous route from scientific discovery to commercial product. We will then be educating scientists who are not only at ease in the several currently distinct academic disciplines that comprise bioinformatics, but who also possess a working familiarity with the complex social enterprise within which different forms of research act synergistically to change the world.

Opposite Strand is a forum for readers to express opinions and ideas about trends and issues in genomics. Submissions should be kept to 550 words and may be submitted to [email protected]

The Scan

White-Tailed Deer Harbor SARS-CoV-2 Variants No Longer Infecting Humans, Study Finds

A new study in PNAS has found that white-tailed deer could act as a reservoir of SARS-CoV-2 variants no longer found among humans.

Study Points to Benefits of Local Consolidative Therapy, Targeted Treatments in Cancer Care

In JCO Precision Oncology, researchers report that local consolidative therapy combined with molecularly targeted treatments could improve survival for some lung cancer patients.

Genetic Variants That Lower LDL Cholesterol Linked to Reduced Heart Disease Risk

Rare variants in two genes that lower LDL cholesterol are also associated with a decreased risk of coronary heart disease, according to a new JAMA Cardiology study.

Study Links Evolution of Longevity, Social Organization in Mammals

With the help of comparative phylogenetics and transcriptomics, researchers in Nature Communications see ties between lifespan and social organization in mammals.