Predicting the binding of small molecules to proteins through invariant representation of the molecular structure

• Verfasst von: Beccaria, Riccardo [VerfasserIn]
• Verfasst von: Lazzeri, A. [VerfasserIn]
• Verfasst von: Tiana, Guido [VerfasserIn]
• Titel: Predicting the binding of small molecules to proteins through invariant representation of the molecular structure
• Verf.angabe: R. Beccaria, A. Lazzeri, and G. Tiana
• E-Jahr: 2024
• Jahr: September 9, 2024
• Umfang: 10 S.
• Illustrationen: Illustrationen
• Fussnoten: Gesehen am 08.10.2024
• Titel Quelle: Enthalten in: Journal of chemical information and modeling
• Ort Quelle: Washington, DC : American Chemical Society, 2005
• Jahr Quelle: 2024
• Band/Heft Quelle: 64(2024), 17 vom: Sept., Seite 6758-6767
• ISSN Quelle: 1549-960X
• DOI: doi:10.1021/acs.jcim.4c00752
• Datenträger: Online-Ressource
• Sprache: eng
• K10plus-PPN: 190508241X

Abstract

We present a computational scheme for predicting the ligands that bind to a pocket of a known structure. It is based on the generation of a general abstract representation of the molecules, which is invariant to rotations, translations, and permutations of atoms, and has some degree of isometry with the space of conformations. We use these representations to train a nondeep machine learning algorithm to classify the binding between pockets and molecule pairs and show that this approach has a better generalization capability than existing methods.