Abstract:Abstract Effective molecular representation learning is of great importance to facilitate molecular property prediction, which is a fundamental task for the drug and material industry. Recent advances in graph neural networks (GNNs) have shown great promise in applying GNNs for molecular representation learning. Moreover, a few recent studies have also demonstrated successful applications of self-supervised learning methods to pre-train the GNNs to overcome the problem of insufficient labeled molecules. However, existing GNNs and pre-training strategies usually treat molecules as topological graph data without fully utilizing the molecular geometry information. Whereas, the three-dimensional (3D) spatial structure of a molecule, a.k.a molecular geometry, is one of the most critical factors for determining molecular physical, chemical, and biological properties. To this end, we propose a novel Geometry Enhanced Molecular representation learning method (GEM) for Chemical Representation Learning (ChemRL). At first, we design a geometry-based GNN architecture that simultaneously models atoms, bonds, and bond angles in a molecule. To be specific, we devised double graphs for a molecule: The first one encodes the atom-bond relations; The second one encodes bond-angle relations. Moreover, on top of the devised GNN architecture, we propose several novel geometry-level self-supervised learning strategies to learn spatial knowledge by utilizing the local and global molecular 3D structures. We compare ChemRL-GEM with various state-of-the-art (SOTA) baselines on different molecular benchmarks and exhibit that ChemRL-GEM can significantly outperform all baselines in both regression and classification tasks. For example, the experimental results show an overall improvement of 8.8% on average compared to SOTA baselines on the regression tasks, demonstrating the superiority of the proposed method.

Pre-training molecular representation model with spatial geometry for property prediction

Automated 3D Pre-Training for Molecular Property Prediction

Geometry-enhanced molecular representation learning for property prediction

3D Infomax improves GNNs for Molecular Property Prediction

Molecular Property Prediction Based on Graph Structure Learning

Multi-Modal Representation Learning for Molecular Property Prediction: Sequence, Graph, Geometry

SMG-BERT: integrating stereoscopic information and chemical representation for molecular property prediction

EMPPNet: Enhancing Molecular Property Prediction via Cross-modal Information Flow and Hierarchical Attention

AEGNN-M:A 3D Graph-Spatial Co-Representation Model for Molecular Property Prediction

MultiModal-Learning for Predicting Molecular Properties: A Framework Based on Image and Graph Structures

Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction

ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction

Self-Supervised Molecular Representation Learning With Topology and Geometry

Gram matrix: an efficient representation of molecular conformation and learning objective for molecular pretraining

Pre-training Molecular Graph Representation with 3D Geometry

Geometry-aware Line Graph Transformer Pre-training for Molecular Property Prediction

A knowledge-guided pre-training framework for improving molecular representation learning

Enhanced Graph Isomorphism Network For Molecular Admet Properties Prediction

Analyzing Learned Molecular Representations for Property Prediction

PointGAT: A quantum chemical property prediction model integrating graph attention and 3D geometry