Dr. Hartemink's research interest is the development of new algorithms in statistical machine learning and artificial intelligence, and on the application of those methods to complex problems in computational genomics. Specific application areas include regulatory genomics and systems biology, although he is also interested in other domains. Current high-level projects include discovering principles and mapping networks of transcriptional regulation, understanding the role of chromatin organization in enacting this regulation, and revealing the mechanisms that control dynamic cellular processes, like the eukaryotic cell cycle.
Although these problems are quite diverse, a number of common themes appear repeatedly throughout his work: probabilistic representations, Bayesian statistics, fusion of information from multiple sources, optimization of joint objective functions, and learning in high-dimensional spaces without over-fitting. Many of these themes are variations on two simple ideas: careful attention to biology in the development of statistical models and the use of informative Bayesian priors to both regularize and guide automated learning.
Beyond GCB, Dr. Hartemink is the faculty director of the Office of Undergraduate Scholars and Fellows, and he previously directed the Computational Biology and Bioinformatics program. Dr. Hartemink has been at Duke since September 2001, after receiving his Ph.D. from the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology under the supervision of David Gifford, Tommi Jaakkola, and Rick Young.