Detection of DNA - Protein Binding Using Deep Learning

Transcription Factors (TF) are the crucial DNA-binding proteins that plays important role in the understanding of transcriptional regulation and detection of mutation mechanism. The prediction of possible DNA binding sites for transcription factors remains a challenging topic in computational biology because of the complexity of biological systems. Thus identification of Transcription Factor Binding Sites (TFBSs) using computational techniques has become an active field of research. In this paper, we propose a Convolutional Neural Network (CNN) model with k-mer encoding of DNA sequences for sequence feature extraction and identification of TFBSs. A series of experiments have been carried out using ChIP-seq dataset showing that our model has performed better than the other competing approaches.