Statistical machine learning for computational biology

Abstract

The course is the second module of a three module course entitled "Bioinformatics: from sequence to structure". This course focuses on learning statistical models from biological data. Three problems are covered: gene finding, classification of gene expression data, and inferring regulatory networks from mRNA and proteomic data. The computational techniques covered include: HMMs, support vector machines, and structure learning with Bayesian networks. This course is made possible by a curriculum development grant from the NSF.