De novo drug design by iterative multiobjective deep reinforcement learning with graph-based molecular quality assessment.

Journal: Bioinformatics (Oxford, England)
Published Date:

Abstract

MOTIVATION: Generating molecules of high quality and drug-likeness in the vast chemical space is a big challenge in the drug discovery. Most existing molecule generative methods focus on diversity and novelty of molecules, but ignoring drug potentials of the generated molecules during the generation process.

Authors

  • Yi Fang
    Department of Neurosurgery, The Fuzhou General Hospital, Fuzhou, China.
  • Xiaoyong Pan
    Department of Veterinary Clinical and Animal Sciences, University of Copenhagen, Copenhagen, Denmark. xypan172436@gmail.com.
  • Hong-Bin Shen
    Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University, Shanghai, 200240, People's Republic of China. hbshen@sjtu.edu.cn.