Molecular Graph Convolutions: Moving Beyond Fingerprints

2 Mar 2016  ·  Steven Kearnes, Kevin McCloskey, Marc Berndl, Vijay Pande, Patrick Riley ·

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.

PDF Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Graph Regression Lipophilicity Weave RMSE 0.715 # 4
Drug Discovery QM9 Molecular Graph Convolutions Error ratio 2.59 # 11


No methods listed for this paper. Add relevant methods here