Graph Neural Networks for Identifying Protein-Reactive Compounds
Preprint
- 7 November 2023
- preprint
- Published by American Chemical Society (ACS) in ChemRxiv
Abstract
The identification of protein-reactive electrophilic compounds is critical to the design of new covalent modifier drugs, screening for toxic compounds, and the exclusion of reactive compounds from high throughput screening. In this work, we employ traditional and graph machine learning algorithms to classify molecules being reactive towards proteins or nonreactive. For training data, we built a new dataset, ProteinReactiveDB, comprised primarily of covalent and noncovalent inhibitors from DrugBank, BindingDB, and CovalentInDB databases. To assess the transferability of the trained models, we created a custom set of covalent and noncovalent inhibitors, which was constructed from recent literature. Baseline models were developed using Morgan fingerprints as training inputs, but they performed poorly when applied to compounds outside the training set. We then trained various Graph Neural Networks (GNNs), with the best GNN model achieving an Area Under the Receiver Operator Characteristic (AUROC) curve of 0.84, precision of 0.92, and recall of 0.73. We also explore the interpretability of these GNNs using Gradient Activation Mapping (GradCAM), which shows regions of the molecules GNNs deem most relevant when making a prediction. These maps indicated that our trained models can identify electrophilic functional groups in a molecule and classify molecules as protein-reactive based on their presence.Keywords
All Related Versions
Funding Information
- Natural Sciences and Engineering Research Council of Canada (RGPIN-05795-2016)
This publication has 68 references indexed in Scilit:
- MolGraph: a Python package for the implementation of molecular graphs and graph neural networks with TensorFlow and KerasJournal of Computer-Aided Molecular Design, 2024
- Generative machine learning for de novo drug discovery: A systematic reviewComputers in Biology and Medicine, 2022
- Molecular fingerprint and machine learning to accelerate design of high‐performance homochiral metal–organic frameworksAIChE Journal, 2021
- A compact review of molecular property prediction with graph neural networksDrug Discovery Today: Technologies, 2020
- Covalent inhibitor reactivity prediction by the electrophilicity index—in and out of scopeJournal of Computer-Aided Molecular Design, 2020
- Leveraging Compound Promiscuity to Identify Targetable Cysteines within the KinomeCell Chemical Biology, 2019
- Discovery of novel arylethenesulfonyl fluorides as potential candidates against methicillin-resistant of Staphylococcus aureus (MRSA) for overcoming multidrug resistance of bacterial infectionsEuropean Journal of Medicinal Chemistry, 2019
- Targeted Covalent Inhibitors for Drug DesignAngewandte Chemie International Edition, 2016
- Automated computational screening of the thiol reactivity of substituted alkenesJournal of Computer-Aided Molecular Design, 2015
- An NMR Spectroscopic Method to Identify and Classify Thiol‐Trapping Agents: Revival of Michael Acceptors for Drug Discovery?Angewandte Chemie International Edition, 2010