An effective self-supervised framework for learning expressive molecular global representations to drug discovery

Abstract

How to produce expressive molecular representations is a fundamental challenge in artificial intelligence-driven drug discovery. Graph neural network (GNN) has emerged as a powerful technique for modeling molecular data. However, previous supervised approaches usually suffer from the scarcity of labeled data and poor generalization capability. Here, we propose a novel molecular pre-training graph-based deep learning framework, named MPG, that learns molecular representations from large-scale unlabeled molecules. In MPG, we proposed a powerful GNN for modelling molecular graph named MolGNet, and designed an effective self-supervised strategy for pre-training the model at both the node and graph-level. After pre-training on 11 million unlabeled molecules, we revealed that MolGNet can capture valuable chemical insights to produce interpretable representation. The pre-trained MolGNet can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of drug discovery tasks, including molecular properties prediction, drug-drug interaction and drug-target interaction, on 14 benchmark datasets. The pre-trained MolGNet in MPG has the potential to become an advanced molecular encoder in the drug discovery pipeline.

Cross-Adversarial Learning for Molecular Generation in Drug Design

“…De novo drug design (Hartenfeller and Schneider, 2010) through existing computer technology can speed up drug development and save research costs. The tasks involved in de novo drug design include molecular generation (Gómez-Bombarelli et al, 2016;Cao and Kipf, 2018;Jin et al, 2018;You et al, 2018;Madhawa et al, 2019;Popova et al, 2019;Zhang et al, 2019;Hong et al, 2020;Zang and Wang, 2020;Bagal et al, 2021), drug and drug interactions (DDI) (Li et al, 2021;Lin et al, 2021;Lyu et al, 2021;Zhao et al, 2021), disease associations (Ding et al, 2020;Lei and Zhang, 2020;Mudiyanselage et al, 2020;Lei X.-J. et al, 2021;Lei X. et al, 2021;Wang Y. et al, 2021;Lei and Zhang, 2021;Yang and Lei, 2021;Zhang et al, 2021), and so on.…”

Docking-based Virtual Screening with Multi-Task Learning

“…As the scaffold represents the core structure of a compound, our results indicate that the model trained with MTL captures more inherent characteristics via sharing knowledge across multiple tasks. By MTL, the learned model achieves a similar effect of representation learning by self-supervised learning [37]. Thus the high-quality representation learned by MTL leads to greater predictive power compared to single-task learning.…”

Accurate Protein-Ligand Complex Structure Prediction using Geometric Deep Learning

“…The similar idea could also be applied to drug development. However, existing methods proposed to design (or generate) novel drug candidates neglected to consider either the 3-D structures or the protein-drug interactions [58,59,60,61,62,63,64,65,66]. Here, we suggest that LigPose has the potential to build novel end-to-end drug-generation methods with both reőned structural details and conődent bioactivity prediction, to boost de novo drug design, which can be regarded as the future work.…”