What question did this study set out to answer?

The study aims to improve toxicity prediction for chemical compounds using advanced machine learning techniques.

March 31, 2026Environmental Toxicology and Chemistry1 citations

Cross-chemical and cross-species toxicity prediction: benchmarking and a novel 3D-structure-based deep learning model

RYRuying YuanFujian Medical University JSJoseph ShawIndiana University Bloomington HTHaixu TangKiel University

Key Points

The study aims to improve toxicity prediction for chemical compounds using advanced machine learning techniques.
Curated a dataset of aquatic toxicity containing over 50,000 records from multiple sources.
Developed a Python library called tox-learn for dataset cleaning and machine learning methods.
Benchmarking of various predictive models with different train-test splitting strategies.
Introduced a 3D structure-based deep learning model named 3DMol-Tox for toxicity predictions.
Machine learning method, molecular fingerprint, and train-test strategy significantly affect prediction performance.
Incorporating species information generally improved toxicity predictions.
The new 3D structure model achieved comparable regression accuracy to existing 2D methods, with improved classification accuracy.

Abstract

Abstract Prediction of a compound’s toxicity is a key step toward realizing animal-free testing of chemical compounds. Recent advances have yielded significant progress in computational toxicity prediction, including machine learning methods that utilize chemical fingerprints and deep learning-based latent representations. However, challenges remain, primarily due to the lack of clean training datasets and the inconsistent model performance. To address these challenges, we curated a comprehensive dataset of aquatic toxicity from seven data sources, which contains 50,603 records for 5,889 compounds across 2,285 different species, much larger than similar datasets used in previous studies. We also developed tox-learn, a Python library featuring tools for automated dataset cleaning, machine learning methods and performance evaluation. The library places special emphasis on avoiding overestimation of prediction accuracy caused by improper train-test data splitting. Based on this toolbox, we benchmarked various predictive models using different train-test splitting strategies on the curated dataset. Our results showed that the choice of machine learning method, molecular fingerprint, and train-test splitting strategy all significantly affect performance. We demonstrated that incorporating species information generally improved predictions, although the degree of improvement depended on how this information was represented. In addition, we developed a new 3D structure–based deep-learning model, 3DMol-Tox, which achieves regression accuracy comparable to the best 2D-structure based model (GPBoost) while exhibiting consistently higher within-one-bin (W1B) classification accuracy. Finally, we analyzed the impact of different train–test splitting strategies and provide recommendations based on our benchmarking, such as using structure-aware splitting to mitigate information leakage, a common issue that inflates reported model performance.

Ask AI

Helpful

Bookmark

View Full Paper