Predicting the binding of small molecules to nuclear receptors using machine learning

Abstract

Nuclear receptors (NRs) are important biological targets of endocrine-disrupting chemicals (EDCs). Identifying chemicals that can act as EDCs and modulate the function of NRs is difficult because of the time and cost of in vitro and in vivo screening to determine the potential hazards of the 100 000s of chemicals that humans are exposed to. Hence, there is a need for computational approaches to prioritize chemicals for biological testing. Machine learning (ML) techniques are alternative methods that can quickly screen millions of chemicals and identify those that may be an EDC. Computational models of chemical binding to multiple NRs have begun to emerge. Recently, a Nuclear Receptor Activity (NuRA) dataset, describing experimentally derived small-molecule activity against various NRs has been created. We have used the NuRA dataset to develop an ensemble of ML-based models to predict the agonism, antagonism, binding and effector binding of small molecules to nine different human NRs. We defined the applicability domain of the ML models as a measure of Tanimoto similarity to the molecules in the training set, which enhanced the performance of the developed classifiers. We further developed a user-friendly web server named ‘NR-ToxPred’ to predict the binding of chemicals to the nine NRs using the best-performing models for each receptor. This web server is freely accessible at http://nr-toxpred.cchem.berkeley.edu. Users can upload individual chemicals using Simplified Molecular-Input Line-Entry System, CAS numbers or sketch the molecule in the provided space to predict the compound’s activity against the different NRs and predict the binding mode for each.

Sensitive image-based chromatin binding assays using inducible ERα to rapidly characterize estrogenic chemicals and mixtures

“…To understand how the chemical structures of the EPA reference chemicals may inform observed assay responses, we used the NR-ToxPred tool to predict (anti-) estrogenic activity of tested compounds. The NR-ToxPred tool uses the Nuclear Receptor Activity (NuRA) dataset and machine-learning based models to predict agonist- and antagonist-like binding to ERα based on chemical structures ( Ramaprasad et al., 2022 ). The predicted likelihood of binding values tends to decrease as in vivo activity decreases ( Figure S5 A).…”

“…Prediction of chemical binding to ERα was performed using the NR-ToxPred online tool ( http://nr-toxpred.cchem.berkeley.edu/predict ) ( Ramaprasad et al., 2022 ). Chemical Pubmed CID numbers were used to retrieve simplified molecular-input line-entry system (SMILES) structures used as input into NR-ToxPred.…”

Screening of the Antagonistic Activity of Potential Bisphenol A Alternatives toward the Androgen Receptor Using Machine Learning and Molecular Dynamics Simulation

“…Recently, Ramaprasad et al proposed an ML-based web server model (NR-ToxPred) to predict the chemical binding of nine different NRs using the NuRA data set and achieved good performance . But NR-ToxPred were trained based on the traditional ML algorithms with molecular fingerprinting and do not use molecular descriptors as input or compare with the performance of GNN models.…”

Prediction of the Interactions of a Large Number of Per- and Poly-Fluoroalkyl Substances with Ten Nuclear Receptors

“…Recently, we developed NR-ToxPred (http://nr-toxpred. cchem.berkeley.edu/), 54 Shortlisting and Classification of PFASs. For each NR, docking results were combined for each PFAS chemical.…”