Benchmarking geometry-based molecular representations for property prediction
Computational and Theoretical Chemistry · 2027-01 · 1267 · 116106
DOI: 10.1016/j.comptc.2026.116106 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computational and Theoretical Chemistry, 2027-01. Verfasst von Muhammad Shahzeb Ali und 1 weiteren Autor:innen. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The predictive performance of molecular machine learning models depends on how molecules are featurized, yet fair comparisons across methods are rare because studies often vary architectures and hyperparameters. We benchmark five geometry-based molecular featurization techniques that operate on three-dimensional atomic coordinates, i.e., Coulomb Matrix, Bag of Bonds, Atom-Centered Symmetry Functions (ACSF), Smooth Overlap of Atomic Positions (SOAP), and Weighted Views, across five benchmark datasets (ESOL, FreeSolv, BBBP, BACE, ClinTox). While Weighted Views employs a unique weighted aggregation architecture, we controlled for model capacity by adjusting network widths so that all methods had roughly the same number of trainable parameters. All representations are evaluated under stratified k-fold cross-validation with three independent seeds, and results are reported as the mean and standard deviation across the resulting test folds.
Zitieren
APA
Ali, M. S., & Bano, S. (2027). Benchmarking geometry-based molecular representations for property prediction. Computational and Theoretical Chemistry, 1267, 116106. https://doi.org/10.1016/j.comptc.2026.116106
Vancouver
Ali MS, Bano S. Benchmarking geometry-based molecular representations for property prediction. Computational and Theoretical Chemistry. 2027;1267:116106. doi:10.1016/j.comptc.2026.116106
BibTeX
@article{Ali2027,
author = {Ali, Muhammad Shahzeb and Bano, Shahtaj},
title = {Benchmarking geometry-based molecular representations for property prediction},
journal = {Computational and Theoretical Chemistry},
year = {2027},
volume = {1267},
pages = {116106},
doi = {10.1016/j.comptc.2026.116106}
}
EUVIMED