Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction
Journal of Chemical Theory and Computation · 11.08.2026 · 22 (15) · 7811-7824
DOI: 10.1021/acs.jctc.6c01334 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemical Theory and Computation, online erschienen am 29.07.2026. Verfasst von Shunzhou Wan und 3 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Computational Science, University College London, London, WC1H 0AJ, United Kingdom. Hauptthemen laut MEDLINE: Drug Discovery, Artificial Intelligence, Proteins. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department for Science, Innovation and Technology, Science and Technology Facilities Council, University College London.
Abstract (Originalsprache)
Despite continuing hype about the role of AI in drug discovery, no “AI-discovered drugs” have so far received regulatory approval. Here we assess one of the latest AI-based tools in this domain. Boltz-2, a recently developed biomolecular foundation model, aims to bridge the gap between AI efficiency and physics-based precision through a joint “cofolding” approach. In this study, we provide an extensive evaluation of Boltz-2 using two large-scale data sets: 16780 compounds for 3CLPro and 21702 compounds for TNKS2. We compare Boltz-2 predicted structures with traditional docking and binding affinities with binding free energies derived from the physics-based ESMACS protocol. Structural analysis reveals significant global RMSD variations, indicating that Boltz-2 predicts multiple protein conformations and ligand binding positions rather than a single converged pose. Energetic evaluations exhibit only weak to moderate correlations across the global data sets. Furthermore, a focused analysis of the top 100 compounds yields no significant correlation between the Boltz-2 predictions and the binding free energies from fine-grained ESMACS, alongside frequently observed saturation-state errors in Boltz-2 predicted ligand structures. Our results show that Boltz-2 lacks the energetic resolution required for lead identification. These findings highlight the necessity of employing physics-based methods for the reliability and refinement of AI-derived models.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Centre for Computational Science, University College London, London, WC1H 0AJ, United Kingdom.
- Advanced Research Computing Centre, University College London, London, WC1H 0AJ, United Kingdom.
ORCID-Profile: Shunzhou Wan, Peter V Coveney
Förderung
- Department for Science, Innovation and Technology
- Science and Technology Facilities Council
- University College London
- University of Bristol
- China Scholarship Council
Zitieren
APA
Wan, S., Zhang, X., Xue, X., & Coveney, P. V. (2026). Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction. Journal of Chemical Theory and Computation, 22(15), 7811-7824. https://doi.org/10.1021/acs.jctc.6c01334
Vancouver
Wan S, Zhang X, Xue X, Coveney PV. Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction. Journal of Chemical Theory and Computation. 2026;22(15):7811-7824. doi:10.1021/acs.jctc.6c01334
BibTeX
@article{Wan2026,
author = {Wan, Shunzhou and Zhang, Xibei and Xue, Xiao and Coveney, Peter V},
title = {Reliability of AI Methods in Drug Discovery: Evaluation of Boltz-2 for Structure and Binding Affinity Prediction},
journal = {Journal of Chemical Theory and Computation},
year = {2026},
volume = {22},
number = {15},
pages = {7811--7824},
doi = {10.1021/acs.jctc.6c01334}
}
EUVIMED