Subject: Hidden Markov Model
Presenter: Alena Shmygelska
Paper: " HMMSTR: a Hidden Markov Model for Local Sequence-Structure Correlations in Proteins "
  by C. Bystroff, V. Thorsson, and D. Baker
Abstract HMMSTR: a Hidden Markov Model for Local Sequence-Structure Correlations in Proteins

"Local Structure" is a generic term for the conformations of short pieces of the protein chain, usually 3-20 residue pieces. In their paper authors describe a hidden Markov model, HMMSTR, that can identify sequence-structure relationship for local structures using I-sites library of sequence-structure motifs. Local structure predictions made by HMMSTR can be used for both homology modelling and ab initio protein folding.

Reference: C. Bystroff, V. Thorsson, and D. Baker. HMMSTR: a hidden Markov model for local sequence-structure correlations in proteins. J Mol Biol. 2000 Aug 4; 301(1):173-190.