Tabular Foundation Models for Environmental Quantitative Structure–Activity Relationship: Robust Prediction with Imperfect Data
Environmental Science & Technology · 04.08.2026 · 60 (30) · 21039-21050
DOI: 10.1021/acs.est.6c05983 ↗Kurz gefasst
Fachartikel in der Zeitschrift Environmental Science & Technology, online erschienen am 20.07.2026. Verfasst von Hanle Lin und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Chemical Engineering, East China University of Science and Technology, Shanghai200237, China. Hauptthemen laut MEDLINE: Quantitative Structure-Activity Relationship. Die Arbeit nennt 39 Literaturangaben. Gefördert von: National Natural Science Foundation of China (NSFC), Natural Science Foundation of Shanghai.
Abstract (Originalsprache)
Environmental quantitative structure–activity relationship (QSAR) modeling is frequently constrained by small sample sizes, missing values, and end point imbalance. To address these limitations, we propose a modular framework based on the TabPFN family that integrates zero-/few-shot prediction, missing robust probabilistic inference, and generative augmentation. Validated across hydroxyl radical rate prediction, singlet oxygen kinetics, and toxicity classification tasks, the framework demonstrates strong data efficiency. Data utility curves combined with a dynamic efficient-window identification strategy provide data-driven guidance for sample size selection. The framework also maintains stable predictive performance and preserves key feature identification under moderate missingness. Under extreme data scarcity (<100 samples) and severe class imbalance, the generative module improves certain conventional classifiers, outperforming GMM-based augmentation and achieving performance comparable to SMOTE while providing limited benefit for TabPFN itself. SHAP and PDP analyses further reveal chemically meaningful trends consistent with current mechanistic knowledge. Overall, this study provides a flexible QSAR workflow for reliable, interpretable, and cost-aware modeling under real-world data constraints.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Chemical Engineering, East China University of Science and Technology, Shanghai200237, China.
ORCID-Profile: Zheng Ma, Yayun Zhang
Förderung
- National Natural Science Foundation of China (NSFC)
- Natural Science Foundation of Shanghai
Zitieren
APA
Lin, H., Wen, H., Ma, Z., Zhao, Y., Zheng, Q., He, C., Cui, L., & Zhang, Y. (2026). Tabular Foundation Models for Environmental Quantitative Structure–Activity Relationship: Robust Prediction with Imperfect Data. Environmental Science & Technology, 60(30), 21039-21050. https://doi.org/10.1021/acs.est.6c05983
Vancouver
Lin H, Wen H, Ma Z, Zhao Y, Zheng Q, He C, et al. Tabular Foundation Models for Environmental Quantitative Structure–Activity Relationship: Robust Prediction with Imperfect Data. Environmental Science & Technology. 2026;60(30):21039-21050. doi:10.1021/acs.est.6c05983
BibTeX
@article{Lin2026,
author = {Lin, Hanle and Wen, Hao and Ma, Zheng and Zhao, Yan and Zheng, Qianwen and He, Chuyuan and Cui, Lekang and Zhang, Yayun},
title = {Tabular Foundation Models for Environmental Quantitative Structure–Activity Relationship: Robust Prediction with Imperfect Data},
journal = {Environmental Science & Technology},
year = {2026},
volume = {60},
number = {30},
pages = {21039--21050},
doi = {10.1021/acs.est.6c05983}
}
EUVIMED