Research On Transcription Factor Binding Sites Recognition Based On HMM

With the developing of bioinformatics many whole genome sequence have been sequenced. It is more and more important to do the research on transcription regulation. Transcription factor is an important transcription element. Identifying its binding site in DNA sequence—transcription binding sites(TFBS)is a hot point in nowadays research. Reliable prediction of transcription factor binding sites can be used to identify the target genes of transcription factors and infer the relationship between the positions of the binding sites and regulation activity of transcription factors. In order to use it to construct transcription regulation network, this will be a guideline for biological research. So this is a research full of value and challenge.In this dissertation, we propose an improved transcription factor binding sites finding method based on Hidden Markov Model. Hidden Markov Model is a strong probability model with math background, it had done very well in transcription factor binding sites finding. We combined Hidden Markov Model with the special features character of special transcription factor. Different classes of transcription factor have different sequence character in binding sites. We constructed new Hidden Markov Models just fits those transcription factor. We also combined restrict calculate in nucleic dependence to improve exactitudes in our algorithm.The algorithm is implemented by C++ .It is applied to the DNA sequence where TFBS maybe in and identify it. In the end of our paper we analyzed the experiment results. It can be proved our improved approach is reasonable and efficient. It got a higher accuracy rate.