Publications
Preprint · arXiv · May 14, 2022
High Performance of Gradient Boosting in Binding Affinity Prediction
Dmitrii Gavrilev, Nurlybek Amangeldiuly, Ivanov, Sergei, Evgeny Vladimirovich Burnaev
Abstract
Prediction of protein-ligand (PL) binding affinity remains the key to drug discovery. Popular approaches in recent years involve graph neural networks (GNNs), which are used to learn the topology and geometry of PL complexes. However, GNNs are computationally heavy and have poor scalability to graph sizes. On the other hand, traditional machine learning (ML) approaches, such as gradient-boosted decision trees (GBDTs), are lightweight yet extremely efficient for tabular data. We propose to use PL interaction features along with PL graph-level features in GBDT. We show that this combination outperforms the existing solutions.
Computational Drug Discovery MethodsMachine Learning in BioinformaticsProtein Degradation and Inhibitorsprotein-ligand binding affinity predictiongradient boostinggradient boosting decision tree
More by Nurlybek Temirbay
- From Knowledge to Action: Outcomes of the 2025 Large Language Model (LLM) Hackathon for Applications in Materials Science and ChemistryarXiv · 2026 · Citations: 5
- Engagement assessment in project-based education: a machine learning approach in team chat analysisEducation and Information Technologies · 2023 · Citations: 31
- Improving Collaborative Design Learning: Which Feedback System is More Effective; Can Ai Take the Wheel?SSRN Electronic Journal · 2023
- DERIVATION OF STRUCTURAL FEATURES THAT AFFECT DRUG METABOLISM BY CYP2C9материалы конференции · 2021
