Course description

Machine learning is an important component in many bioinformatics problems: from gene finding, to protein function prediction. In this course we will focus on support vector machines (SVM) and other kernel-based classification. Kernel methods are an important class of machine learning methods that have a wide range of applicability, including in bioinformatics. Kernel methods lend themselves particularly well to the difficult aspects of biological data, which include high dimensionality (as in microarray data), lack of a vector based representation (as in sequence and structure data), and the need to combine heterogeneous sources of information. The course will provide a detailed overview of current research in kernel methods and their applications in bioinformatics. Students will present recent research papers, implement the methods described described in those papers, and test them using the PyML machine learning environment.

Textbook:

John Shawe-Taylor and Nello Cristianini. Kernel methods for pattern analysis, Cambridge, 2004.

Grading: