The Research Of Hidden Markov Models And Applications In Biological Sequence Alignment

With the development of Human Genome Project(HGP),bioinformatics(the core discipline of life science)is also developing fast.In bioinformatics,sequence alignment is the most basic method and foundation.This paper focuses on the research of bio-sequences alignment based on the Hidden Markov Models(HMM).In this paper,first we introduce basic knowledge of bioinformatics.we also elaborate the definition of the bio-sequences alignment and some of the relevant concepts in sequence alignment.It emphatically introduces and analyzes HMM's foundational theories and basic concepts,and then summarizes out the three basic problems and algorithms of HMM applications.Then we introduce the sequence alignment based on HMM in detail.Finally,classical HMM are known to provide compelling performance among multiple sequence alignment algorithms,yet their stochastic nature does not help them cope with the existing dependence among the sequence element. To overcome the limitations of the classical HMM,the paper propose a new HMM- fuzzy Hidden Markov Models which is based on Sugeno fuzzy measures and Choquet integrals,by relaxing the classical HMM's independence assumptions.With the Fuzzy HMM,the fuzzy forward algorithm and the fuzzy backward algorithm and the fuzzy Viterbi algorithm are defined,and also recursive formulas are given about them and the reestimation formulas of the parameters.The example results shows that the algorithm for classical HMM is more superior to the algorithm for fuzzy HMM.