Semantic Scholar Open Access 2016 1574 sitasi

Molecular graph convolutions: moving beyond fingerprints

S. Kearnes Kevin McCloskey Marc Berndl V. Pande Patrick F. Riley

Abstrak

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.

Penulis (5)

S

S. Kearnes

K

Kevin McCloskey

M

Marc Berndl

V

V. Pande

P

Patrick F. Riley

Format Sitasi

Kearnes, S., McCloskey, K., Berndl, M., Pande, V., Riley, P.F. (2016). Molecular graph convolutions: moving beyond fingerprints. https://doi.org/10.1007/s10822-016-9938-8

Akses Cepat

Lihat di Sumber doi.org/10.1007/s10822-016-9938-8
Informasi Jurnal
Tahun Terbit
2016
Bahasa
en
Total Sitasi
1574×
Sumber Database
Semantic Scholar
DOI
10.1007/s10822-016-9938-8
Akses
Open Access ✓