Lecture on Advanced Methods for Sequence Analysis WS07/08
In this lecture series, we present machine learning approaches to sequence analysis. Students will learn about state of the art methods such as the support vector machine and boosting and how to apply them in the context of biological sequences.
Students of Bioinformatics and Computer Science interested in (a) an overview of general and efficient algorithms for statistical learning used in computational biology, (b) sequence kernels for the problems such as promoter or splice site detection and (c) cutting edge methods for predicting structured outputs and obtaining interpretable classifiers. No specific knowledge will be required, but students are expected to have some mathematical maturity.
Lectures
- (23 Oct 07) Motivation and Overview
- (30 Oct 07) Support Vector Machines and Introduction to Kernels
- (6 Nov 07) Soft Margin SVMs and Convex Optimization
- (13 Nov 07) Boosting
- (20 Nov 07) Probabilistic Models
- (27 Nov 07) Statistical Learning Theory
- (4 Dec 07) Fast SVMs using String Kernels
- (18 Dec 07) Graph Kernels and their Applications
- (22 Jan 08) Multiple Kernel Learning
- (29 Jan 08) Structured Output Learning
- (6 Feb 08) Large Scale Applications and Summary (combined lecture)
Tutorial/Practical Sessions
- (8 Nov 07) Practical Session - Room A104, 14-17pm
- (29 Nov 07) Practical Session - Room C118a, Sand14, 15-18pm
- (10 Jan 08) Practical Session - Room A104, 14-17pm
- (14 Feb 08) Practical Session - Room A104, 14-17pm
For the computer practicals, we are using python. To install python (we recommend python2.5), see the following:
- Linux (usually installed) - http://www.python.org/download/
- OSX - http://www.pythonmac.org/packages/
- Windows - http://code.enthought.com/enthon/
Please also have the following packages installed:
- numpy - http://numpy.scipy.org/
- matplotlib - http://matplotlib.sourceforge.net/
- cvxopt - http://abel.ee.ucla.edu/cvxopt
- ipython (optional) - http://ipython.scipy.org
- shogun - http://www.shogun-toolbox.org
Official Information
People:
- Gunnar Raetsch, Cheng Soon Ong, Petra Philips
- Friedrich Miescher Laboratory of the Max Planck Society
- Max Planck Institute for Biological Cybernetics
- Contact: amsa07@tuebingen.mpg.de
Place:
- Lectures: Tue 14-16, am Sand, A302
- Practical Sessions: A104 (8.11, 10.1, 14.2, 14-17pm); C118a (29.11, 15-18pm)
Exam:
- Oral exam in mid February 2007, register by 31 January 2007.

