Cameron Mura / Chemistry
Computer science students + Biochemistry students = Computation biology @ UVa
I think that to really know something is to be able to compute it. An ‘algorithm’ is no more than just that: a three-part information encapsulation, wherein (i) starting knowledge (‘input’) is fed into (ii) a method (‘program’), by which those data are transformed to yield (iii) new knowledge, in the form of novel, insightful results & predictions (‘output’). My own research area – biochemistry & structural biology – is increasingly becoming a computational science, whether it involves large-scale datasets that have become available over the past decade (e.g., the human genome), the wealth of data resulting from high-throughput determination of protein structures, or large-scale, massively parallel simulations that can now generate terabyte-sized ‘movies’ of the motion and dynamical behavior of proteins and other complex biological molecules, all in atomic detail! Indeed, this confluence of biochemistry and computational sciences over the past decade has led to the birth of entire new disciplines known as ‘computational biology’ and ‘bioinformatics’.
At present, UVa undergrads receive little to no exposure to this area, leaving them at a severe disadvantage in their understanding and appreciation of contemporary biochemical research (not to mention their own planning and thinking about post-graduation research opportunities). My dream idea is to alleviate this state of affairs by developing an informal discussion group comprised of a series of biweekly, 90-minute, lunch-time lectures on computational biology. This series will culminate with an end-of-semester field-trip to the National Institutes of Health (NIH; see below).
Approximately ten or twelve undergraduate students (third- and fourth-years) will be selected to participate, and the success of this endeavor will hinge upon the way the students are chosen: half of the undergrads will come from the Chemistry and Biology departments, while the other half will be drawn from the Computer Science (CS) department. The students will then be paired-up, one CS person with one non-CS person. The lunch-time lectures will alternate between me and the students: I’ll present the basic principles of a particular bioinformatics topic one week (chosen from recent topical research articles, published in high-visibility journals such as Nature or Science), and then one of the pairs of students will provide a critique of that topic the following week. After this ~10- 12-week series, all of us will make a field-trip to the NIH campus in Bethesda, Maryland to attend an instructional seminar in the area of computational biology, offered via one of the NIH’s institutes (the Center for Information Technology [CIT] regularly offers intense training seminars in areas such as molecular visualization, docking and drug design, genome analysis, etc.). There at the NIH, the students will get to see their newfound knowledge ‘in action’.
In addition to exposing the students to a whole new realm of science, this dream idea could have much longer-lasting impact on their lives – Students in the biochemical sciences are often intimidated by computation; reciprocally, those with affinity for physical or computational sciences are often ignorant about the rich complexity and mesmerizing beauty of biological systems. This dream idea would give both camps the intellectual confidence to fearlessly tackle any field of science – purely computational, purely biological, or anything in between. Perhaps most importantly, students would gain an appreciation for the highly interdisciplinary approach by which the modern biological sciences progress. Who knows, some of them may even end up deciding to be Biochem/CS double-majors!
In terms of budget, expenses would be incurred for the regularly repeating series of noon-time lunches (probably ~8 sessions x 10 students x $12/student = $960); a portion (or all) of the remainder of the $3k balance would be dedicated towards the NIH visit (two rental vans; registration costs).