Deep Learning for Protein Binding Sites Prediction from Protein Sequence

日期 : Jan 2018 - 现在

Proteins are functional units of a cell so identifying the binding sites of a protein is paramount for understanding many biological processes. It will help us to better understand potential protein functions, novel diseases as well as improve our understanding of molecular organisms. Although various computational methods for protein binding site prediction (based on protein sequence) have been proposed but their accuracy still needs improvement. Deep learning algorithms have been used in diverse areas such as Computer vision, Artificial intelligence etc., showing promising results. The main idea of this work is to predict binding sites with the use of deep learning algorithms from protein sequence data only. Initially, we would devise a manual feature based approach and compare its results with fully automated approach (Deep learning) for the prediction of protein binding sites