Predoctoral Trainee

Beverly R. Seavey Computer Sciences Doctoral Training Program Dept. of Computer Sciences Faculty Supervisor: C. David Page Email: seavey@cs.wisc.edu 263-7626

Hidden Markov Models (HMMs) are widely used for modeling sequence data in bioinformatics. Increasingly, we are finding tasks in which a data point involves more than than one sequence component, each with features in addition to the sequence data. For proteins, relevant features could include isoelectric point, cellular localization, or predicted surface residues. Motivated by such data, we extend HMMs to contain additional variables, giving the model some Bayesian Net character. We apply this method to predicting, within a family of proteins and ligands, which pairs will bind strongly.

Resume

Resume 2002 (.doc format)

Publications

Seavey B and Page CD. A Hybrid HMM-Naive Bayes Model for Predicting Protein-Protein Interactions Based on SH3 Domains. Abstract, International Conference on Intelligent Systems for Molecular Biology, 2002, Edmonton, Alberta, CA.

Search for other publications by Beverly Seavey (Pub Med, Cite Seer)

Feedback, questions or accessibility issues.
Graduate Program  |  Postdoctorate Program  |  Short Term Research Program

File last modified: August 27, 2003
Copyright © 2003 The Board of Regents of the University of Wisconsin System.