BGT 200/STA 270
Statistical Methods for Computational Biology
Fall 2003

Instructor: Edwin Iversen, Jr.
E-mail: iversen@stat.duke.edu
Phone: 681-8442
Office: 211E Old Chemistry
Office hours: TBA or by appointment

Course Description:

Introduction to methods of statistical inference and stochastic modeling underlying common tools in functional genomics and computational molecular biology. Statistical principles behind a range of problems in bioinformatics and genomics, including biological sequence analysis and structure prediction, database searching, gene expression analysis, statistical genetics, phylogenetic inference and genetic epidemiology. Applied data analysis techniques including regression analysis, clustering, principal components analysis (PCA), and data visualization. Advanced statistical models and algorithms including hidden Markov models (HMMs), the EM algorithm, Gibbs sampling and Markov chain Monte Carlo (MCMC).

Course assignments will involve hands-on data analysis of a range of biological datasets, and each lecture will include a practical component demonstrating use of standard statistical software. As a final project, each student will perform a more detailed analysis of a dataset of their choice, with instructor guidance.



Course meetings: Friday 11:50-2:10
Location: 025 Old Chem Bldg.

last updated 20 August 2003