CaltechAUTHORS
Published July 8, 2020 | Version Submitted
Discussion Paper Open

Graph Neural Networks for the Prediction of Substrate-Specific Organic Reaction Conditions

Creators

Abstract

We present a systematic investigation using graph neural networks (GNNs) to model organic chemical reactions. To do so, we prepared a dataset collection of four ubiquitous reactions from the organic chemistry literature. We evaluate seven different GNN architectures for classification tasks pertaining to the identification of experimental reagents and conditions. We find that models are able to identify specific graph features that affect reaction conditions and lead to accurate predictions. The results herein show great promise in advancing molecular machine learning.

Additional Information

© 2020 by the author(s). To appear in the ICML 2020 Workshop on Graph Representation Learning and Beyond (GRLB). We thank the reviewers for their insightful comments and Prof Pietro Perona for mentorship guidance and helpful discussions on this work. Fellowship support was provided by the NSF (M.R.M., T.J.D. Grant No. DGE-1144469). S.E.R. is a Heritage Medical Research Investigator. Financial support from the Research Corporation Cottrell Scholars Program is acknowledged.

Attached Files

Submitted - 2007.04275.pdf

Files

2007.04275.pdf

Files (2.9 MB)

Name Size
md5:a4d6566221075a28535638d57f7959cb
2.9 MB Preview Download

Additional details

Identifiers

Eprint ID
106598
Resolver ID
CaltechAUTHORS:20201110-154207213

Related works

Describes
http://arxiv.org/abs/2007.04275 (URL)

Funding

NSF Graduate Research Fellowship
DGE-1144469
Heritage Medical Research Institute
Cottrell Scholar of Research Corporation

Dates

Created
2020-11-11
Created from EPrint's datestamp field
Updated
2023-06-02
Created from EPrint's last_modified field

Caltech Custom Metadata

Caltech groups
Heritage Medical Research Institute