Title: A clustering approach for estimating parameters of a profile hidden Markov model

Authors: Rosa Aghdam; Hamid Pezeshk; Seyed Amir Malekpour; Soudabeh Shemehsavar; Changiz Eslahchi

Addresses: School of Mathematics, Statistics and Computer Science, Shahid Beheshti University, G.C., Tehran, Iran ' School of Mathematics, Statistics and Computer Science and Center of Excellence in Biomathematics, College of Science, University of Tehran, Tehran 14155-6455, Iran; Bioinformatics Research Group, Institute for Research in Fundamental Sciences (IPM), Tehran, Iran ' Robert Cedergren Center for Bioinformatics and Genomics, Biochemistry Department, Universite de Montreal, 2900 Edouard-Montpetit, Montreal, QC, H3T 1J4, Canada ' School of Mathematics, Statistics and Computer Science and Center of Excellence in Biomathematics, College of Science, University of Tehran, Tehran 14155-6455, Iran ' Faculty of Mathematical Science, Shahid-Beheshti University, G.C., Tehran, Iran

Abstract: A Profile Hidden Markov Model (PHMM) is a standard form of a Hidden Markov Models used for modelling protein and DNA sequence families based on multiple alignment. In this paper, we implement Baum-Welch algorithm and the Bayesian Monte Carlo Markov Chain (BMCMC) method for estimating parameters of small artificial PHMM. In order to improve the prediction accuracy of the estimation of the parameters of the PHMM, we classify the training data using the weighted values of sequences in the PHMM then apply an algorithm for estimating parameters of the PHMM. The results show that the BMCMC method performs better than the Maximum Likelihood estimation.

Keywords: hidden Markov model; PHMM; profile HMM; Baum-Welch algorithm; BMCMC; Bayesian Monte Carlo Markov chain; entropy; clustering; parameter estimation; protein modelling; DNA sequences; multiple alignment; bioinformatics.

DOI: 10.1504/IJDMB.2013.054696

International Journal of Data Mining and Bioinformatics, 2013 Vol.8 No.1, pp.66 - 82

Received: 02 Oct 2010
Accepted: 03 Oct 2011
Published online: 20 Oct 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article