iGRLDTI: an improved graph representation learning method for predicting drug-target interactions over heterogeneous biological information network.

Journal: Bioinformatics (Oxford, England)
Published Date:

Abstract

MOTIVATION: The task of predicting drug-target interactions (DTIs) plays a significant role in facilitating the development of novel drug discovery. Compared with laboratory-based approaches, computational methods proposed for DTI prediction are preferred due to their high-efficiency and low-cost advantages. Recently, much attention has been attracted to apply different graph neural network (GNN) models to discover underlying DTIs from heterogeneous biological information network (HBIN). Although GNN-based prediction methods achieve better performance, they are prone to encounter the over-smoothing simulation when learning the latent representations of drugs and targets with their rich neighborhood information in HBIN, and thereby reduce the discriminative ability in DTI prediction.

Authors

  • Bo-Wei Zhao
    The Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011, China.
  • Xiao-Rui Su
    The Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi, 830011, China.
  • Peng-Wei Hu
    IBM Research, Beijing, 100049, China.
  • Yu-An Huang
    Department of Computing, Hong Kong Polytechnic University, Hung Hom, Hong Kong, China.
  • Zhu-Hong You
    Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi, 830011, China. zhuhongyou@ms.xjb.ac.cn.
  • Lun Hu
    The Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Ürümqi, China.