Computing has changed biology--biology education must catch up.

541 EDUCATIONFORUM A dvances in computing have forever changed the practice of biological research. Computational biology, or bioinformatics, is as essential for biology in this century as molecular biology was in the last. In fact, it is diffi cult to imagine modern molecular biology without computational biology. For example, a diffi cult algorithmic puzzle had to be solved in order to successfully assemble the human genome sequence from millions of short pieces. However, the computational components of undergraduate biology education have hardly changed in the past 50 years. New courses for biologists should be more relevant to their discipline, complementing the standard mathematical courses that were originally designed for physicists and engineers. Bioinformatics and biology communities should work together so that education of biologists in the 21st century may become as sophisticated as the computational education of physicists or economists. For example, today's typical undergraduate economics curriculum may cover linear and integer programming, combinatorial algorithms, dynamic programming, game theory , and other computational concepts. These disciplines were in their infancy 40 years ago when the computational revolution started in economics. Because most biologists today (as most economists 40 years ago) do not know dynamic programming, for example, the idea of introducing such concepts into the biology curriculum may appear foreign and impractical. But the paradoxical result is that economics undergraduates may now be better prepared than biology graduate students to understand how DNA sequence alignment or gene prediction algorithms work (based on dynamic programming). The RECOMB Bioinformatics Education Conference (http://casb.ucsd.edu/ bioed/) explored ways to teach bioinfor-matics to undergraduate biology students. Attending biologists, computer scientists, and mathematicians from various branches of bioinformatics agreed that the time has come to shift the paradigm in biology education by adding new computational courses to standard curricula. This realization is not new: BIO2010, a National Research Council report (1), recommended substantial changes in the mathematics curricula for research-oriented biology undergraduates. Bialek and Botstein (2) and Pevzner (3) acknowledged the problem and outlined some creative approaches to its solution. However, the question of how best to deliver computational ideas to biologists remains. Because bioinformatics is a computational science, courses should strive to present the ideas that drive an algorithm's design and to explain the crux of a statistical approach, rather than merely to recount the algorithms and statistical techniques. It is critical that bioinformatics is taught as a science that explains computational ideas and shows how …