Using Dirichlet Mixture Priors to Derive Hidden Markov Models for Protein Families

A Bayesian method for estimating the amino acid distributions in the states of a hidden Markov model (HMM) for a protein familyor the columns of a multiple alignment of that family is introduced. This method uses Dirichlet mixture densities as priors over aminoacid distributions. These mixture densities are determined from examination of previously constructed HMMs or multiple alignments. It is shown that this Bayesian method can improve the quality of HMMs produced from small training sets. Speci c experiments on the EF-hand motif are reported, for which these priors are shown to produce HMMs with higher likelihood on unseen data, and fewer false positives and false negatives in a database search task.

• Type: Conference
• Year: 1993
• Where: ISMB
• Authors: Michael Brown, Richard Hughey, Anders Krogh, I. Saira Mian, Kimmen Sjölander, David Haussler
• Keywords: Bayesian Method, Computational Biology, ISMB 1993, Mixture Densities, Multiple Alignments