Carnegie Mellon University

Our Vision

Robert F. Murphy

Computational biology is a critically important and growing field that is essential to biomedical research.  The Computational Biology Department at Carnegie Mellon is part of the internationally-recognized School of Computer Science, and draws upon the incredible energy and expertise in the entire School.  It is an ideal place to be educated in this essential discipline.

The approach to computational biology that we take at Carnegie Mellon is unique in several ways.

First, we strongly believe that computational biology has important contributions to make in framing biological problems in computational terms (see “What is Computational Biology?”) and should not just be focused on helping biomedical researchers “understand their data.”  We place a lot of emphasis on novel ways of how viewing biology through a computational lens has led to research advances.

Second, as befits our being in a School of Computer Science, we emphasize the importance of developing computationally rigorous solutions to problems, which goes hand-in-hand with framing those problems well.  We try to avoid developing a piecemeal or ad hoc solution or relying on current thinking of how a biological process might work and instead try to find a theoretically sound way to compute what is needed directly from primary biomedical data.

Third, we provide an important grounding in both natural and computational sciences.  The role of computational biologists in framing problems requires knowledge of the fundamental principles of both.

Fourth, we believe strongly in the need for computational biology to drive biological experimentation.  There has been enormous discussion in the popular and scientific media about the need for automated analysis of “Big Bio Data” because of its overwhelming size.  While there are many opportunities created by large biomedical datasets, we stress that these datasets do not come close to being sufficient to develop accurate models of complex biological systems.  Most large datasets are currently acquired by choosing a small set of variables and sampling them very thoroughly.  This results in the acquisition of a lot of measurements that could have been predicted from others.  Furthermore, it is not feasible, either in terms of cost or time, to do this type of exhaustive sampling for all combinations of variables.  We believe in the future that iteratively doing modest sets of experiments chosen by computer models rather than individual investigators, a process called active machine learning, will enable more effective research.

Finally, a core value of Carnegie Mellon is its collaborative spirit. Collaborations between experimental and computational biologists is what drives the recent advances in the general area of systems biology. The Computational Biology Department fosters unique opportunities to students to be involved in such collaborations building on the great tradition of interdisciplinary work at CMU. Our students learn from and work with experimental faculty, and they participate in cutting edge research that is jointly performed by computational and experimental researchers.

Together, the embodiment of these principles in our degree programs helps students to develop as independent innovators who will help guide the future of biomedical research, and not just be able to apply today’s methods to today’s problems.