Science Pool

Deep Learning Applied to Ligand-Based De Novo Drug Design

This article forms a book chapter in Artificial Intelligence in Drug Design.

In this review article, we provide the latest insights into de novo design approaches based on  artificial intelligence (AI) algorithms with a specific focus on ligand-based methods.

It includes:

  • a background to de novo design approaches developed prior to the use of AI
  • an overview of AI and commonly employed neural network architectures used in ligand based de novo design
  • a list of more than 100 deep generative models reported in the literature from 2017-2020
  • current applications of deep generative approaches in the drug discovery context
  • an insight into future applications of deep generative models in de novo drug design