Evaluating the roughness of structure–property relationships using pretrained molecular representations
Name
d3dd00088e.pdf
Description
Published version
Size
1.08 MB
Format
Adobe PDF
Checksum (MD5)
6ef535cc2480effbf5f1992842f11dd4
Author(s)
Graff, David E
Pyzer-Knapp, Edward O
Jordan, Kirk E
Shakhnovich, Eugene I
Coley, Connor W
Date Issued
2023
Journal
Digital Discovery
Publisher
Royal Society of Chemistry
Citation
Graff, David E, Pyzer-Knapp, Edward O, Jordan, Kirk E, Shakhnovich, Eugene I and Coley, Connor W. 2023. "Evaluating the roughness of structure–property relationships using pretrained molecular representations." Digital Discovery, 2 (5).
Version
Final published version
Abstract
Quantitative structure–property relationships (QSPRs) aid in understanding molecular properties as a function of molecular structure. When the correlation between structure and property weakens, a dataset is described as “rough,” but this characteristic is partly a function of the chosen representation. Among possible molecular representations are those from recently-developed “foundation models” for chemistry which learn molecular representation from unlabeled samples via self-supervision. However, the performance of these pretrained representations on property prediction benchmarks is mixed when compared to baseline approaches. We sought to understand these trends in terms of the roughness of the underlying QSPR surfaces. We introduce a reformulation of the roughness index (ROGI), ROGI-XD, to enable comparison of ROGI values across representations and evaluate various pretrained representations and those constructed by simple fingerprints and descriptors. We show that pretrained representations do not produce smoother QSPR surfaces, in agreement with previous empirical results of model accuracy. Our findings suggest that imposing stronger assumptions of smoothness with respect to molecular structure during model pretraining could aid in the downstream generation of smoother QSPR surfaces.
MIT Department
Massachusetts Institute of Technology. Department of Chemical Engineering
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Creative Commons Attribution
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1039/d3dd00088e
