Prediction of Contact Maps in Proteins Based on Recurrent Neural Network with Bias Units

Prediction of inter_residue contact maps may be seen as a strategic step toward the solution of fundamental open problems in structural genomics. Predicting the contact map of a protein of unknown structure can give significant clues about the structure of and folding mechanism of that protein. In this paper, we focus on prediction of contact maps in proteins based on recurrent neural network with bias units and have gotten a better prediction results.