Statistical Machine Learning and Computational Biology

Statistical machine learning is a field that combines algorithmic ideas with foundational concepts from probability and statistics. This combination makes statistical machine learning an essential tool for computational biology, in part because probabilistic notions are inherent in biology (arising, e.g., via thermodynamics, recombination and germline mutation) and in part because of the incomplete nature of most biological data sets. I will present several examples of applications of statistical machine learning to problems in biology, in the areas of protein functional annotation, protein structural modeling, protein structure prediction and multipopulation linkage and association analysis.