STAT 420 - Statistics for Bioinformatics (3)

This course surveys the statistical methodology underlying current bioinformatics techniques. Topics to be covered include: dynamic programming, including the Needleman-Wunsch algorithm and Smith-Waterman algorithm; methods of inference, including maximum likelihood and Bayesian approach; Markov models, including Markov chains, hidden Markov models and inferences for these models; Monte-Carlo Markov chain methods, including Gibbs sampling and Metropolis-Hastings algorithm; extreme-value theory, including Gumbel distribution and significance of alignments; cluster analysis, including hierarchical methods, K-means method and determination of number of clusters; classification methods, including CART algorithm and QUEST algorithm; generalized linear models, including model types, inference and statistics for model fit; model validation, cross-validation; and predictive assessment.

Grading: Graded/Satisfactory Unsatisfactory/Audit
Course ID: 57061
Consent: No Special Consent Required
Components: Lecture
Prerequisite: MATH 152 and (STAT 350 or STAT 355) with a grade of ‘C’ or better.