Transfer learning for drug-target interaction prediction.

Journal: Bioinformatics (Oxford, England)
PMID:

Abstract

MOTIVATION: Utilizing AI-driven approaches for drug-target interaction (DTI) prediction require large volumes of training data which are not available for the majority of target proteins. In this study, we investigate the use of deep transfer learning for the prediction of interactions between drug candidate compounds and understudied target proteins with scarce training data. The idea here is to first train a deep neural network classifier with a generalized source training dataset of large size and then to reuse this pre-trained neural network as an initial configuration for re-training/fine-tuning purposes with a small-sized specialized target training dataset. To explore this idea, we selected six protein families that have critical importance in biomedicine: kinases, G-protein-coupled receptors (GPCRs), ion channels, nuclear receptors, proteases, and transporters. In two independent experiments, the protein families of transporters and nuclear receptors were individually set as the target datasets, while the remaining five families were used as the source datasets. Several size-based target family training datasets were formed in a controlled manner to assess the benefit provided by the transfer learning approach.

Authors

  • Alperen Dalkıran
    Department of Computer Engineering, Middle East Technical University, Ankara 06800, Turkey.
  • Ahmet Atakan
    Department of Computer Engineering, METU, Ankara, 06800, Turkey.
  • Ahmet S Rifaioğlu
    Department of Computer Engineering, Iskenderun Technical University, Hatay 31200, Turkey.
  • Maria J Martin
    European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, UK.
  • Rengül Çetin Atalay
    Faculty of Pulmonary and Critical Care Medicine, the University of Chicago, Chicago, IL, 60637, United States.
  • Aybar C Acar
    Cancer Systems Biology Laboratory, Graduate School of Informatics, METU, Ankara, 06800, Turkey.
  • Tunca Doğan
    European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, Cambridge, UK.
  • Volkan Atalay
    Department of Computer Engineering, METU, Ankara, 06800, Turkey. vatalay@metu.edu.tr.