Integrating Machine Learning-Based Pose Sampling with Established Scoring Functions for Virtual Screening
Uloženo v:
| Název: | Integrating Machine Learning-Based Pose Sampling with Established Scoring Functions for Virtual Screening |
|---|---|
| Autoři: | Thi Ngoc Lan Vu, Hosein Fooladi, Johannes Kirchmair |
| Zdroj: | J Chem Inf Model |
| Informace o vydavateli: | American Chemical Society (ACS), 2025. |
| Rok vydání: | 2025 |
| Témata: | Machine Learning, Molecular Docking Simulation, 106005 Bioinformatik, Preclinical/methods, Protein Conformation, 102001 Artificial intelligence, 102001 Artificial Intelligence, Drug Evaluation, 106005 Bioinformatics, 301207 Pharmazeutische Chemie, Article, 301207 Pharmaceutical chemistry, Proteins/chemistry |
| Popis: | Physics-based docking methods have long been the cornerstone of structure-based virtual screening (VS). However, the emergence of machine learning (ML)-based docking approaches has opened up new possibilities for enhancing VS technologies. In this study, we explore the integration of DiffDock-L, a leading ML-based pose sampling method, into VS workflows by combining it with the well-established Vina and Gnina scoring functions. We assess this integrated approach in terms of its VS effectiveness, pose sampling quality, and complementarity to traditional physics-based docking methods, such as AutoDock Vina. Our findings from the DUDEZ benchmark dataset show that DiffDock-L performs competitively in both VS performance and pose sampling in cross-docking settings. In most cases, it generates physically plausible and biologically relevant poses, establishing itself as a viable alternative to physics-based docking algorithms. Additionally, we found that the choice of scoring function significantly influences VS success. |
| Druh dokumentu: | Article Other literature type |
| ISSN: | 1549-960X 1549-9596 |
| DOI: | 10.26434/chemrxiv-2025-96kzg-v2 |
| DOI: | 10.26434/chemrxiv-2025-96kzg |
| DOI: | 10.1021/acs.jcim.5c00380 |
| Přístupová URL adresa: | https://pubmed.ncbi.nlm.nih.gov/40343848 |
| Rights: | CC BY URL: http://creativecommons.org/licenses/by/4.0/This article is licensed under CC-BY 4.0 |
| Přístupové číslo: | edsair.doi.dedup.....96f451a419a87135013f46869c38c86b |
| Databáze: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstrakt: | Physics-based docking methods have long been the cornerstone of structure-based virtual screening (VS). However, the emergence of machine learning (ML)-based docking approaches has opened up new possibilities for enhancing VS technologies. In this study, we explore the integration of DiffDock-L, a leading ML-based pose sampling method, into VS workflows by combining it with the well-established Vina and Gnina scoring functions. We assess this integrated approach in terms of its VS effectiveness, pose sampling quality, and complementarity to traditional physics-based docking methods, such as AutoDock Vina. Our findings from the DUDEZ benchmark dataset show that DiffDock-L performs competitively in both VS performance and pose sampling in cross-docking settings. In most cases, it generates physically plausible and biologically relevant poses, establishing itself as a viable alternative to physics-based docking algorithms. Additionally, we found that the choice of scoring function significantly influences VS success. |
|---|---|
| ISSN: | 1549960X 15499596 |
| DOI: | 10.26434/chemrxiv-2025-96kzg-v2 |