FetoML: Interpretable predictions of the fetotoxicity of drugs based on machine learning approaches

Abstract

Pregnant females may use medications to manage health problems that develop during pregnancy or that they had prior to pregnancy. However, using medications during pregnancy has a potential risk to the fetus. Assessing the fetotoxicity of drugs is essential to ensure safe treatments, but the current process is challenged by ethical issues, time, and cost. Therefore, the need for in silico models to efficiently assess the fetotoxicity of drugs has recently emerged. Previous studies have proposed successful machine learning models for fetotoxicity prediction and even suggest molecular substructures that are possibly associated with fetotoxicity risks or protective effects. However, the interpretation of the decisions of the models on fetotoxicity prediction for each drug is still insufficient. This study constructed machine learning-based models that can predict the fetotoxicity of drugs while providing explanations for the decisions. For this, permutation feature importance was used to identify the general features that the model made significant in predicting the fetotoxicity of drugs. In addition, features associated with fetotoxicity for each drug were analyzed using the attention mechanism. The predictive performance of all the constructed models was significantly high (AUROC: 0.854-0.974, AUPR: 0.890-0.975). Furthermore, we conducted literature reviews on the predicted important features and found that they were highly associated with fetotoxicity. We expect that our model will benefit fetotoxicity research by providing an evaluation of fetotoxicity risk for drugs or drug candidates, along with an interpretation of that prediction. Author summaryDrugs are often necessary for the treatment of diseases in pregnant females. However, some drugs can potentially cause fetotoxicities, such as teratogenicity and abortion. Therefore, it is essential to study fetotoxicity, but traditional toxicity testing demands time, money, and labor. To modernize these testing methods, in silico approaches for predicting the fetotoxicity of drugs are emerging. The proposed models so far have successfully predicted the fetotoxicity of drugs and proposed some fetotoxicity-related substructures, but the interpretation of the models determination is still insufficient. In this study, we proposed FetoML to predict the fetotoxicity of drugs based on machine learning and provide the substructures that the model focused on in predicting fetotoxicity for each drug. We confirmed the significant predictive performance and interpretability of the model through a quantitative performance evaluation and literature review. We expect FetoML to benefit fetotoxicity studies of drugs by modernizing the paradigm of fetotoxicity testing and providing insights to researchers.