3D Convolutional Neural Networks and a CrossDocked Dataset for Structure-Based Drug Design
Preprint
- 20 February 2020
- preprint
- Published by American Chemical Society (ACS) in ChemRxiv
Abstract
One of the main challenges in drug discovery is predicting protein-ligand binding affinity. Recently, machine learning approaches have made substantial progress on this task. However, current methods of model evaluation are overly optimistic in measuring generalization to new targets, and there does not exist a standard dataset of sufficient size to compare performance between models. We present a new dataset for structure-based machine learning, the CrossDocked2020 set, with 22.5 million poses of ligands docked into multiple similar binding pockets across the Protein Data Bank and perform a comprehensive evaluation of grid-based convolutional neural network models on this dataset. We also demonstrate how the partitioning of the training data and test data can impact the results of models trained with the PDBbind dataset, how performance improves by adding more, lower-quality training data, and how training with docked poses imparts pose sensitivity to the predicted affinity of a complex. Our best performing model, an ensemble of 5 densely connected convolutional newtworks, achieves a root mean squared error of 1.42 and Pearson R of 0.612 on the affinity prediction task, an AUC of 0.956 at binding pose classification, and a 68.4% accuracy at pose selection on the CrossDocked2020 set. By providing data splits for clustered cross-validation and the raw data for the CrossDocked2020 set, we establish the first standardized dataset for training machine learning models to recognize ligands in non-cognate target structures while also greatly expanding the number of poses available for training. In order to facilitate community adoption of this dataset for benchmarking protein-ligand binding affinity prediction, we provide our models, weights, and the CrossDocked2020 set at https://github.com/gnina/models.Keywords
All Related Versions
- Published version: Journal of Chemical Information and Modeling, 60 (9), 4200.
This publication has 36 references indexed in Scilit:
- Cross‐docking benchmark for automated pose and ranking prediction of ligand bindingProtein Science, 2019
- Most Ligand-Based Classification Benchmarks Reward Memorization Rather than GeneralizationJournal of Chemical Information and Modeling, 2018
- Protein–Ligand Scoring with Convolutional Neural NetworksJournal of Chemical Information and Modeling, 2017
- Forging the Basis for Developing Protein–Ligand Interaction Scoring FunctionsAccounts of Chemical Research, 2017
- Overview of Methods and Strategies for Conducting Virtual Small Molecule ScreeningCurrent Protocols in Chemical Biology, 2017
- Optimal strategies for virtual screening of induced-fit and flexible target in the 2015 D3R Grand ChallengeJournal of Computer-Aided Molecular Design, 2016
- Benchmarking methods and data sets for ligand enrichment assessment in virtual screeningMethods, 2015
- AutoDock Vina: Improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreadingJournal of Computational Chemistry, 2009
- Effects of inductive bias on computational evaluations of ligand-based modeling and on drug discoveryJournal of Computer-Aided Molecular Design, 2007
- Ensemble docking of multiple protein structures: Considering protein structural variations in molecular dockingProteins: Structure, Function, and Bioinformatics, 2006