Deep Molecular Representation Learning via Fusing Physical and Chemical Information
This work addresses molecular representation learning for deep learning applications in molecular science, presenting a novel hybrid method with strong specific gains.
The authors tackled the problem of molecular representation learning by introducing PhysChem, a neural architecture that fuses physical and chemical information, achieving state-of-the-art performance on the MoleculeNet benchmark and demonstrating effectiveness on SARS-CoV-2 datasets.
Molecular representation learning is the first yet vital step in combining deep learning and molecular science. To push the boundaries of molecular representation learning, we present PhysChem, a novel neural architecture that learns molecular representations via fusing physical and chemical information of molecules. PhysChem is composed of a physicist network (PhysNet) and a chemist network (ChemNet). PhysNet is a neural physical engine that learns molecular conformations through simulating molecular dynamics with parameterized forces; ChemNet implements geometry-aware deep message-passing to learn chemical / biomedical properties of molecules. Two networks specialize in their own tasks and cooperate by providing expertise to each other. By fusing physical and chemical information, PhysChem achieved state-of-the-art performances on MoleculeNet, a standard molecular machine learning benchmark. The effectiveness of PhysChem was further corroborated on cutting-edge datasets of SARS-CoV-2.