Hidden Markov Model: Application towards genomic analysis

Hidden Markov Model (HMM) has become one of the interesting methods for the researchers, especially in bioinformatics where different analysis are carried out. These are widely used in science, engineering and many other areas such as bioinformatics, genomic mapping, computer vision, finance and economics, and in social science. HMMs require much smaller training sets, and that the examination of the inner structure of the model provides often a deeper understanding of the phenomenon. In this survey, we first describe the important algorithms for the HMMs, and provide useful comparisons, aiming at their advantages and shortcomings. We then consider the major g applications, such as annotations, gene alignment and profiling of sequences, DNA structure prediction, and pattern recognition. We also list some analysis on how to use HMM for DNA genomes. Finally, we conclude use and perspectives of HMMs in bioinformatics and provide a critical appraisal for the same.