Molecular graph convolutions: moving beyond fingerprints

• DOI: 10.1007/s10822-016-9938-8
• Publication Date: 2016-08-24T08:46:36Z
• Authors: Steven Kearnes; Kevin McCloskey; Marc Berndl; Vijay Pande; Patrick Riley
• Keywords: 0301 basic medicine; FOS: Computer and information sciences; Computer Science - Machine Learning; 0303 health sciences; Molecular Structure; Machine Learning (stat.ML); Ligands; Machine Learning (cs.LG); 3. Good health; Machine Learning; 03 medical and health sciences; Pharmaceutical Preparations; Statistics - Machine Learning; Drug Design; Computer Graphics; Computer-Aided Design; Neural Networks, Computer

See "Version information" section<br/>Molecular "fingerprints" encoding structural information are the workhorse of cheminformatics and machine learning in drug discovery applications. However, fingerprint representations necessarily emphasize particular aspects of the molecular structure while ignoring others, rather than allowing the model to make data-driven decisions. We describe molecular "graph convolutions", a machine learning architecture for learning from undirected graphs, specifically small molecules. Graph convolutions use a simple encoding of the molecular graph---atoms, bonds, distances, etc.---which allows the model to take greater advantage of information in the graph structure. Although graph convolutions do not outperform all fingerprint-based methods, they (along with other graph-based methods) represent a new paradigm in ligand-based virtual screening with exciting opportunities for future improvement.<br/>