A Novel Method For Classification Of Transcription Factors Based On Properties Of Transcription Factor Binding Sites

Transcription factors (TF) are proteins that control the first step of gene expression, the transcription of DNA into RNA sequences, which regulate gene transcription by specific binding of the regulatory regions of DNA sequences. The mechanism of transcriptional regulatory can be much better understood if the category of transcription factors is known. In this research we developed a novel method for classification of TFs based on properties of transcription factor binding sites (TFBSs) with SVMs, and achieved very good classification results.The main results in this paper are as follows:First, we get six properties of TFBSs to stand for TF. The properties are:the length of TFBSs for a TF, a new PWM value, the proportion of not conservative TFBSs, the proportion of no nucleosome of TFBSs, the proportion of conserved nucleosome of TFBSs, and the GC content of TFBSs.Second, we calculate the six properties of TFBSs for yeast and drosophila, and construct the factor for their TFs.Third, we classified the TFs of yeast and drosophila with SVMs based on the constructed factors. Finally, we get great classification results.In this article, we introduce a classification strategy for the prediction of the types of transcription factors. It has more biological significance to classify TFs using properties of their binding sites. The excellent results show that our method is effective and efficient.