Persistent Directed Flag Laplacian (PDFL)-Based Machine Learning for Protein–Ligand Binding Affinity Prediction
Open Access
- 5 April 2025
- journal article
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 21 (8) , 4276-4285
Abstract
Directionality in molecular and biomolecular networks plays an important role in the accurate representation of the complex, dynamic, and asymmetrical nature of interactions present in protein-ligand binding, signal transduction, and biological pathways. Most traditional techniques of topological data analysis (TDA), such as persistent homology (PH) and persistent Laplacian (PL), overlook this aspect in their standard form. To address this, we present the persistent directed flag Laplacian (PDFL), which incorporates directed flag complexes to account for edges with directionality originated from polarization, gene regulation, heterogeneous interactions, etc. This study marks the first application of PDFL, providing an in-depth analysis of spectral graph theory combined with machine learning. In addition to its superior accuracy and reliability, the PDFL model offers simplicity by requiring only raw inputs without complex data processing. We validated our multikernel PDFL model for its scoring power against other state-of-the-art methods on three popular benchmarks, namely PDBbind v2007, v2013, and v2016. The computational results indicate that the proposed PDFL model outperforms competitors in protein-ligand binding affinity predictions, suggesting that PDFL is a promising tool for protein engineering, drug discovery, and general applications in science and engineering.Keywords
All Related Versions
Funding Information
- Bristol-Myers Squibb (65109)
- Michigan State University
- National Institutes of Health (R01AI164266, R01GM126189)
- National Science Foundation (DMS-2052983, IIS-1900473)
This publication has 43 references indexed in Scilit:
- Persistent Laplacian projected Omicron BA.4 and BA.5 to become new dominating variantsComputers in Biology and Medicine, 2022
- Persistent spectral graphInternational Journal for Numerical Methods in Biomedical Engineering, 2020
- MathDL: mathematical deep learning for D3R Grand Challenge 4Journal of Computer-Aided Molecular Design, 2019
- DG‐GL: Differential geometry‐based geometric learning of molecular datasetsInternational Journal for Numerical Methods in Biomedical Engineering, 2019
- Mathematical deep learning for pose and binding affinity prediction and ranking in D3R Grand ChallengesJournal of Computer-Aided Molecular Design, 2018
- Improving scoring-docking-screening powers of protein-ligand scoring functions using random forestJournal of Computational Chemistry, 2016
- Improving AutoDock Vina Using Random Forest: The Growing Accuracy of Binding Affinity Prediction by the Effective Exploitation of Larger Data SetsMolecular Informatics, 2015
- An entropy-based persistence barcodePattern Recognition, 2015
- Topological Persistence and SimplificationDiscrete & Computational Geometry, 2002
- On the use of size functions for shape analysisBiological Cybernetics, 1993