10 No. 1
Science in the Seams
Computational and Life Sciences Initiative redefines disciplinary lines
High-performance computing at Emory
“Everybody understands or recognizes the combination of computational and life sciences as very promising. . . . Few people have been working in this combination of fields long enough to have established a leadership presence.”
“Rather than 'deconstructing' nature into its simplest parts . . . , the twenty-first century will likely be spent trying to understand, scientifically, the nature of complex interacting systems by “reconstructing” complexity.”
Thinking Outside the Pipeline
The impact of the unexpected in work-life issues
Creative Minds and "The Greatness Game"
Henry David Thoreau surely would have been surprised at having predicted the course of scientific inquiry for the next 125 years or so. For much of their histories, the keystone sciences of chemistry, biology, and physics distilled and dismantled the universe, then examined the fragments for clues about how it all worked. Biologists peered into the cell, the nucleus, and ultimately our biological destiny coiled up in DNA. The atom revealed its dense nucleus, which, with some energetic prodding, blew apart into fleeting particles with amusing names like lepton and quark. Chemists charted every known element and condensed the information in the periodic table.
“Over the last century, science has spent a lot of time breaking things down into parts to understand them,” says David Lynn, professor of biomolecular chemistry and chair of the chemistry department. “We’ve reached a point where we have enough foundational information to try to put it all back together again, to figure out how systems operate.”
Visionary scientists began to reconsider—or just ignore—disciplinary boundaries; branches of science once regarded as distinct and largely dissimilar converged, and no more so than in life sciences. “Natural science, social science, chemistry, and even psychology now start to have problems that hit between rather than within those disciplines,” Lynn adds. “The idea of interdisciplinary science has become preeminent in the way we think now, much more so than we did in the past. Science is interdisciplinary because problems are interdisciplinary.”
Breaking down barriers
But while individuals can deftly hopscotch among fields of study, their institutions are not so nimble. “To some extent we’ve gotten compartmentalized by the traditional structures of the university,” says Dennis Liotta, professor of chemistry. “We have departments, we have schools, and those have been functional units that served important administrative purposes. But as science has evolved we find them to be somewhat artificial barriers. They’re very useful for teaching undergraduates and to some extent graduate students. But if I want to collaborate on work with someone in the medical school, all of a sudden that intrinsic infrastructure is gone, and I have to work much harder on those kinds of projects.”
Emory needed a crucible in which its various and varied biological sciences and potent computer-based analytical tools could blend, re-form, and emerge with new and, it is hoped, even greater capabilities. What materialized, as envisioned and defined by a steering committee of Emory faculty and administrators, was the Computational and Life Sciences Initiative (CLS), a notion important enough to warrant an entire section in the Strategic Plan. To say that hopes are high is an understatement. From the Strategic Plan:
[The CLS] will optimize enterprise-wide technologies and unleash potential that may quickly move Emory to the next level of rank and reputation. . . . It is no exaggeration to say that this is the cutting edge of science early in the 21st Century. It is one of the areas where Emory can stake a claim to global leadership. . . . It will be this yin and yang of scientific discovery—the opportunity to create interplay between the fundamental and applied dimensions of new knowledge—that allows Emory to emerge as a destination university in the sciences.
Initial funding for the CLS comes from a portion of the sale of royalty rights to the anti-AIDS drug Emtriva, which Liotta developed in collaboration with a colleague from the medical school. That breakthrough arose from precisely the type of cross-fertilization of two university units envisioned by the CLS. Pursued independently and in isolation, the discovery might never have materialized.
The CLS will eventually occupy the equivalent of two floors of the new chemistry addition going up on Oxford Road, and the CLS executive committee is actively hunting for talent (applicant screening began May 1). Their five-year goal is to hire eight to ten new faculty and several postdoctoral candidates. A new Ph.D. program in computational science and informatics welcomes its first incoming class this fall. The faculty positions aren’t specified but will be determined by each appointee’s qualification and interests. Most of the newcomers will likely inhabit more than one department. Parallel initiatives in predictive health and neuroscience are in the works.
“We have a lot of strengths in the life sciences, but we also have certain gaps,” Liotta says. “This could be an opportunity for us to hire some high-quality individuals who can fill in those gaps and give us an opportunity to ask and answer some of the very big questions that address modern science.”
Emory is, of course, not alone in attacking interdisciplinary challenges, but the approaches differ. Harvard, for example, chose to take people from existing departments and to construct a new Department of Systems Biology. “In our assessment, such an approach may create the same type of limiting infrastructure as before,” says Lynn. “Emory chose to aggregate rather than isolate. We don’t know if we’re right or not, but we think the approach is more powerful.” The University of Chicago, Lynn adds, stands out as a school that creates centers to mix disciplines, such as social sciences, business, and economics. “They’ve been very successful at fashioning bridging structures that have succeeded. The CLS is modeled more on that type of bridging structure than on new departmental entities.
Computational science gives the CLS “oomph,” though its prominence in the mix might seem unusual at first. The field is typically associated with mathematically intensive domains such as high-energy physics, fluid mechanics, and molecular modeling, but the sheer volume of data that cascades from life sciences research nowadays overwhelms human analytic capacity. “There’s been a revolution in the amount of information we can acquire about biological systems and the human genome, gene sequences, and proteins,” says Lanny Liebeskind, Samuel Candler Dobbs Professor of Chemistry and Director of University Scientific Strategies.
“There’s so much information that it’s impossible for a human to interpret, analyze, and understand it, or to draw conclusions and look for patterns. It’s physically impossible without help from computers.” Really powerful computers. Think processing speeds of a trillion operations per second (a terraflop) and storage capacity in petabytes (one billion megabytes).
“There is an increasing convergence of life sciences and computational science, which is becoming more and more important for things like analysis of genes, simulations of blood flow in vessels, and identifying tumors in X-rays,” adds Vaidy Sunderam, Samuel Candler Dobbs Professor of Computer Science and co-chair, along with Lynn, of the CLS. “Computational science is going to become the underpinning of how life sciences is done. The experimental aspect—where you hypothesize something, perform a physical action and observe the result—will remain, but it’s increasingly complemented by theoretical things: you predict or you know something about how a system or organism is supposed to look and you model it mathematically, then translate that into a computer and actually simulate it to get some insight into how it works. It’s going to increase the pace and quality at which new discoveries are made.” At the same time, the fusion will inevitably produce advancements in the underlying fields of mathematics and computational techniques.
Much of the work conducted under the CLS umbrella will be basic science—stepping stones toward more tangible advances that lead to practical, “real-world” applications—that defies understanding by the untrained. (Consider a paper coauthored by Lynn in the Journal of Bacteriology in 2005: “Environmental pH Sensing: Resolving the VirA/VirG Two-component System Inputs for Agrobacterium Pathogenesis.”) Given Emory’s enviable track record and reputation in life sciences and medical research, the potential is tremendous: fabricate living tissues and organs; engineer photosynthesis and capture light for alternative energy sources; build biocomputers that are part microorganism, part circuitry; new vaccines and drugs; molecular “machines” that work from within cells to detect and repair damage; reveal the origins of living systems—how life began.
“There have been times in human history,” says Lynn, “when technology has shaken the foundation of our social structure—how we exist and function. There are many examples: The domestication of plants and animals, discovery of fossil fuels and nuclear energy, Darwinian evolution. You can argue we’re in the midst of another of those challenges, where we can define individuals by a sequence of base pairs, identify stem cells, and clone mammals to generate armies of identical individuals. The impact of humankind on the sustainability of the planet collides with social structure norms; upheavals emerge between religion and science. Imagine the CLS as a tree with roots that reach out to different departments to bring resources back to the center to attack problems. The bridging structure will attempt to combine theory and experiment in unique ways to address a series of problems that are preeminent in the world today.”—S.F.