CGBVS-DNN: Prediction of Compound-protein Interactions Based on Deep Learning.

Journal: Molecular informatics
Published Date:

Abstract

Computational prediction of compound-protein interactions (CPIs) is of great importance for drug design as the first step in in-silico screening. We previously proposed chemical genomics-based virtual screening (CGBVS), which predicts CPIs by using a support vector machine (SVM). However, the CGBVS has problems when training using more than a million datasets of CPIs since SVMs require an exponential increase in the calculation time and computer memory. To solve this problem, we propose the CGBVS-DNN, in which we use deep neural networks, a kind of deep learning technique, instead of the SVM. Deep learning does not require learning all input data at once because the network can be trained with small mini-batches. Experimental results show that the CGBVS-DNN outperformed the original CGBVS with a quarter million CPIs. Results of cross-validation show that the accuracy of the CGBVS-DNN reaches up to 98.2 % (σ<0.01) with 4 million CPIs.

Authors

  • Masatoshi Hamanaka
    Graduate School of Medicine, Kyoto University, Shogoin-kawaharacho, city/>Sakyo-ku Kyoto, 606-8507, Japan.
  • Kei Taneishi
    Advanced Institute for Computational Science, RIKEN, 7-1-28, Minatojima-minami-machi, Chuo-ku, Kobe, Hyogo, 650-0047, Japan.
  • Hiroaki Iwata
    Division of School of Health Science, Department of Biological Regulation, Faculty of Medicine, Tottori University, 86 Nishi-cho, Yonago 683-8503, Japan.
  • Jun Ye
    Department of Electrical and Information Engineering, Shaoxing University, 508 Huancheng West Road, Shaoxing, Zhejiang 312000, PR China. Electronic address: yehjun@aliyun.com.
  • Jianguo Pei
    Software and Services Group, Intel Corporation, Shanghai, China.
  • Jinlong Hou
    Software and Services Group, Intel Corporation, Shanghai, China.
  • Yasushi Okuno
    Graduate School of Medicine, Kyoto University, Shogoin-kawaharacho, city/>Sakyo-ku Kyoto, 606-8507, Japan.