In Silico Prediction of ERRα Agonists Based on Combined Features and Stacking Ensemble Method

• Published in: ChemMedChem 2024-10, Vol.19 (20), p.e202400298-n/a
• Main Authors: Xu, Jiahao, Huang, Zejun, Duan, Hao, Li, Weihua, Zhuang, Jingyan, Xiong, Le, Tang, Yun, Liu, Guixia
• Format: Article
• Language: English
• Subjects: Agonists; Algorithms; Combined features; Computer Simulation; Data sampling; Diabetes mellitus (non-insulin dependent); ERRalpha Estrogen-Related Receptor; ERRα agonists; Estrogens; Humans; Machine learning; Metabolic disorders; Molecular modelling; Molecular Structure; Prediction models; Receptors, Estrogen - antagonists & inhibitors; Receptors, Estrogen - chemistry; Receptors, Estrogen - metabolism; Stacking ensemble method; Type 2 diabetes
• ISSN: 1860-7179; 1860-7187; 1860-7187
• DOI: 10.1002/cmdc.202400298

Estrogen‐related receptor α (ERRα) is considered a very promising target for treating metabolic diseases such as type 2 diabetes. Development of a prediction model to quickly identify potential ERRα agonists can significantly reduce the time spent on virtual screening. In this study, 298 ERRα agonists and numerous nonagonists were collected from various sources to build a new dataset of ERRα agonists. Then a total of 90 models were built using a combination of different algorithms, molecular characterization methods, and data sampling techniques. The consensus model with optimal performance was also validated on the test set (AUC=0.876, BA=0.816) and external validation set (AUC=0.867, BA=0.777) based on five selected baseline models. Furthermore, the model's applicability domain and privileged substructures were examined, and the feature importance was analyzed using the SHAP method to help interpret the model. Based on the above, it's hoped that our publicly accessible data, models, codes, and analytical techniques will prove valuable in quick screening and rational designing more novel and potent ERRα agonists. Discovering ERRα agonists is a promising direction for the treatment of metabolic diseases like type 2 diabetes. By merging combined features and stacking ensemble method, the consensus model with optimal performance was constructed based on the data collected from different sources. Our explainable model and open‐source data could provide valuable guidance for the discovery and modification of ERRα agonists.