Hidden Markov Model Biological Information And The Improvement Of The Algorithm

The Hidden Markov Model(HMM)is a statistical method, which was introduced in 1970s. Each model consists of a Hidden Markov process and an observation process. There are three problems needed to be solved by using Hidden Markov Models, which are recognizing, decoding and training, The answers for those three problems consist of the theory of Hidden Markov Model. Previously, it was mainly used to recognize the speech, and Churchill put it to computational biology in 1989. With the development of Human Genome Project, bioinformatics, the core discipline of life science, has developed fast. Currently, HMM has a wider application in bioinformatics. This paper focus on the researchment of bioinformatic, which is based on the Hidden Markov Models.In this paper, firstly, the present research situation and developing trends of HMM are introduced, and basic knowledge of bioinformatics is described, which contains the background of biology development and main research contents. Secondly, it introduces the foundational theories and basic concepts of HMM emphatically, then summarizes out three basic problems and algorithms of HMM applications. Next, we analyse the application of the sequence alignment and CpG islands based on HMM in detail. Finally, among the caculation of the state transition probability and the output observation probability, the classical HMM always considers the current state rather than historical states, so this paper improves the markov assuption of state transition and output observation sequences, and establishes the new model.This paper gets the third-order forward-backward algorithm and ordinary circumstances, which is based on HMM and the second-order forward-backward algorithm of new model. In addition, this paper proposes second-order viterbi algorithm and introduces second-order EM algorithm.