XLPFE: A Simple and Effective Machine Learning Scoring Function for Protein–Ligand Scoring and Ranking

Abstract

Prediction of protein–ligand binding affinities is a central issue in structure-based computer-aided drug design. In recent years, much effort has been devoted to the prediction of the binding affinity in protein–ligand complexes using machine learning (ML). Due to the remarkable ability of ML methods in nonlinear fitting, ML-based scoring functions (SFs) can deliver much improved performance on a selected test set, such as the comparative assessment of scoring functions (CASF), when compared to the classical SFs. However, the performance of ML-based SFs heavily relies on the overall similarity of the training set and the test set. To improve the performance and transferability of an SF, we have tried to combine various features including energy terms from X-score and AutoDock Vina, the properties of ligands, and the statistical sequence-related information from either the binding site or the full protein. In conjunction with extreme trees (ET), an ML model, we have developed XLPFE, a new SF. Compared with other tested methods such as X-score, AutoDock Vina, ΔvinaXGB, PSH-ML, or CNN-score, XLPFE achieves consistently better scoring and ranking power for various types of protein–ligand complex structures beyond the CASF, suggesting that XLPFE has superior transferability. In particular, XLPFE performs better with metalloenzymes. With its faster speed, improved accuracy, and better transferability, XLPFE could be usefully applied to a diverse range of protein–ligand complexes.

Ligand binding affinity prediction with fusion of graph neural networks and 3D structure-based complex graph

“…Accurate prediction of protein-ligand binding affinity remains one of the grand challenges of computational chemistry and biology. [1][2][3] With the ever increasing amount of high-resolution experimentally determined protein-ligand structures, 4 the binding affinity prediction methods have switched from physicsbased [5][6][7][8][9][10][11] to empirical scoring functions [12][13][14][15] and knowledgebased, 16,17 and in the last decade to machine learning [18][19][20][21][22][23][24][25][26][27] and deep learning based methods. [28][29][30][31][32][33][34][35][36][37][38] Especially, deep learning is an end-to-end method that is ideally suited to find hidden nonlinear relationships 39 between 3D protein-ligand complex structures and binding affinity.…”

Iterative Knowledge-Based Scoring Function for Protein–Ligand Interactions by Considering Binding Affinity Information

“…Comparison with Other Scoring Functions. In recent years, many scoring functions have been developed, such as traditional scoring functions DLIGAND2 75 and KORP-PL 76 and several latest machine learning-based scoring functions, such as PIGNet, 77 △ Lin_F9 XGB, 78 PLANET, 79 RTMScore, 80 Sfcnn, 81 XLPFE, 82 graphDelta, 83 and EGNA. 84 We have also compared our ITScoreAff with those functions, and the detailed results are provided in Supporting Information, Tables S6 and S7.…”

Enhancing Generalizability in Protein–Ligand Binding Affinity Prediction with Multimodal Contrastive Learning

“…These functions include force field-based, − empirical − to knowledge-based, − machine learning-based methodologies, , or hybrid methodologies. − Given their prowess in navigating intricate, high-dimensional nonlinear challenges, machine learning techniques have found synergy with scoring functions. Traditional machine learning scoring functions typically rely on hand-crafted features, , employing algorithms like RF, gradient-booted tree (GBT), or SVM for refinement . Yet, there has been a paradigm shift toward leveraging end-to-end neural networks, such as convolutional neural networks (CNN) and graph neural networks (GNN), for feature extraction.…”