TGF-M: Topology-augmented geometric features enhance molecular property prediction
PLOS Computational Biology · 2025
DOI: 10.1371/journal.pcbi.1013004 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2025. Verfasst von Wei He und 6 weiteren Autor:innen. Beteiligte Einrichtung: Qingdao Institute of Software, College of Computer Science and Technology, China University of Petroleum (East China), Shandong Key Laboratory of Int…. Hauptthemen laut MEDLINE: Drug Design. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Project of China, National Natural Science Foundation of China, Taishan Scholarship.
Abstract (Originalsprache)
Accurate prediction of molecular properties is a key component of Artificial Intelligence-driven Drug Design (AIDD). Despite significant progress in improving these predictive models, balancing accuracy with computational complexity remains a challenge. Molecular topological and geometric features provide rich spatial information, crucial for improving prediction accuracy, but their extraction typically increases model complexity. To address this, we propose TGF-M (Topology-augmented Geometric Features for Molecular Property Prediction), a novel predictive model that optimizes feature extraction to enhance information capture and improve model accuracy, and reduces model complexity to lower computational cost. This approach enhances the model’s ability to leverage both topological and geometric features without unnecessary complexity. On the re-segmented PCQM4Mv2 dataset, TGF-M performs remarkably, achieving a low mean absolute error (MAE) of 0.0647 in the HOMO-LUMO gap prediction task with only 6.4M parameters. Compared to two recent state-of-the-art models evaluated within a unified validation framework, TGF-M demonstrates comparable performance with less than one-tenth of the parameters. We conducted an in-depth analysis of TGF-M’s chemical interpretability. The results further validate the method’s effectiveness in leveraging complex molecular topology and geometry during model learning, underscoring its potential and advantages. The trained models and source code of TGF-M are publicly available at https://github.com/TiAW-Go/TGF-M .
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Qingdao Institute of Software, College of Computer Science and Technology, China University of Petroleum (East China), Shandong Key Laboratory of Intelligent Oil and Gas Industrial Software, Qingdao, China.
- The Department of Stomatology, The First Medical Center, Chinese PLA General Hospital, Beijing, China.
Förderung
- National Key Research and Development Project of China
- National Natural Science Foundation of China
- Taishan Scholarship
- Shandong Provincial Natural Science Foundation
Zitieren
APA
He, W., Tian, X., Li, X., Han, P., Wang, S., Liu, L., & Song, T. (2025). TGF-M: Topology-augmented geometric features enhance molecular property prediction. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1013004
Vancouver
He W, Tian X, Li X, Han P, Wang S, Liu L, et al. TGF-M: Topology-augmented geometric features enhance molecular property prediction. PLOS Computational Biology. 2025. doi:10.1371/journal.pcbi.1013004
BibTeX
@article{He2025,
author = {He, Wei and Tian, Xu and Li, Xue and Han, Peifu and Wang, Shuang and Liu, Lin and Song, Tao},
title = {TGF-M: Topology-augmented geometric features enhance molecular property prediction},
journal = {PLOS Computational Biology},
year = {2025},
doi = {10.1371/journal.pcbi.1013004}
}
EUVIMED