Alignment of multiple proteins with an ensemble of Hidden Markov Models. (15th January 2010)

• Record Type: Journal Article
• Authors: Song, Jia; Liu, Chunmei; Song, Yinglei; Qu, Junfeng; , Gurdeep S. Hura
• Is Part Of: International journal of data mining and bioinformatics. Volume 4:Number 1(2010)
• Page Start: 60
• Page End: 71
• Publication Date: 2010-01-15
• Subjects: profile HMMs -- hidden Markov models -- alignment accuracy -- progressive -- bioinformatics -- protein sequences -- multiple sequence alignment; Data mining -- Periodicals; Bioinformatics -- Periodicals; 006.312
• Languages: English
• ISSNs: 1748-5673

Abstract

In this paper, we developed a new method that progressively constructs and updates a set of alignments by adding sequences in a certain order to each of the existing alignments. Each of the existing alignments is modelled with a profile Hidden Markov Model (HMM) and an added sequence is aligned to each of these profile HMMs. We introduced an integer parameter for the number of profile HMMs. The profile HMMs are then updated based on the alignments with leading scores. Our experiments on BaliBASE showed that our approach could efficiently explore the alignment space and significantly improve the alignment accuracy.