Pocket-Surface Discrete Differential Geometry as a Leakage-Robust Feature Class for Protein–Ligand Binding Affinity Prediction
Molecules · 01.06.2026 · 31 (11) · 1899
DOI: 10.3390/molecules31111899 ↗Kurz gefasst
Fachartikel in der Zeitschrift Molecules, online erschienen am 01.06.2026. Verfasst von Mehmet Ali Balcı und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics, Faculty of Science, Mugla Sitki Kocman University, 48000 Mugla, Turkey. Hauptthemen laut MEDLINE: Proteins. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Türkiye Bilimsel ve Teknolojik Araştırma Kurumu, The Scientific and Technological Research Council of Türkiye.
Abstract (Originalsprache)
Protein–ligand binding affinity prediction underpins structure-based drug discovery, yet random partitions of public benchmarks overestimate generalisation due to protein-family and ligand leakage, and the marginal value of explicit pocket-geometry descriptors over atom-level graph neural networks remains unclear. We computed a 59-dimensional discrete differential geometry descriptor on the ligand-aware solvent-excluded surface of 3285 PDBBind v2020 complexes, combining curvature distributions, the leading sixteen Laplace–Beltrami eigenvalues and a ten-point heat-kernel signature, and evaluated it in gradient-boosted tree pipelines across progressively stricter split regimes and two leak-proof external benchmarks, together with four mechanistically distinct injection strategies in a SchNet-style graph neural network. The descriptor lifted Pearson correlations by 0.111 on cluster-disjoint testing, 0.258 on LP-PDBBind DataSAIL S2 and 0.365 on CASF-2016, while in isolation reaching 0.456 to 0.594 on external benchmarks, on a par with X-Score and AutoDock Vina (version 1.2). TreeSHAP attribution localised the dominant signal to the heat-kernel signature. The four graph neural network injection strategies produced no statistically significant lift, indicating that distance-based message passing on atomic coordinates already captures much of the geometric content. Pocket-surface discrete differential geometry, therefore, offers an interpretable, leakage-robust and lightweight feature class for early-stage virtual screening, and motivates hybrid mesh-to-atom architectures.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Mathematics, Faculty of Science, Mugla Sitki Kocman University, 48000 Mugla, Turkey.
- Department of Mathematics, Faculty of Science, Ege University, 35100 Izmir, Turkey.
- Translational Oncology Department, Oncology Institute, Dokuz Eylul University, 35340 Izmir, Turkey.
- Oncology Department, Institute of Health Sciences, Dokuz Eylul University, 35340 Izmir, Turkey.
ORCID-Profile: Mehmet Ali Balcı, Erbil Çetin, Gizem Calibasi-Kocal, Ömer Akgüller
Förderung
- Türkiye Bilimsel ve Teknolojik Araştırma Kurumu
- The Scientific and Technological Research Council of Türkiye
Zitieren
APA
Balcı, M. A., Çetin, E., Calibasi-Kocal, G., & Akgüller, Ö. (2026). Pocket-Surface Discrete Differential Geometry as a Leakage-Robust Feature Class for Protein–Ligand Binding Affinity Prediction. Molecules, 31(11), 1899. https://doi.org/10.3390/molecules31111899
Vancouver
Balcı MA, Çetin E, Calibasi-Kocal G, Akgüller Ö. Pocket-Surface Discrete Differential Geometry as a Leakage-Robust Feature Class for Protein–Ligand Binding Affinity Prediction. Molecules. 2026;31(11):1899. doi:10.3390/molecules31111899
BibTeX
@article{Balc2026,
author = {Balcı, Mehmet Ali and Çetin, Erbil and Calibasi-Kocal, Gizem and Akgüller, Ömer},
title = {Pocket-Surface Discrete Differential Geometry as a Leakage-Robust Feature Class for Protein–Ligand Binding Affinity Prediction},
journal = {Molecules},
year = {2026},
volume = {31},
number = {11},
pages = {1899},
doi = {10.3390/molecules31111899}
}
EUVIMED