Machine Learning for Toxicity Prediction in Low-Sample Molecular Classes
Barajas, C.; Dunphy, L.; Mullany, L.; Tiburzi, O.; Lloyd, E.
Show abstract
Deep learning models such as Chemprop have advanced quantitative molecular property prediction, but their reliance on large training sets limits use in data-scarce domains. We propose a framework that fine-tunes a general baseline model trained on publicly available data on small, class-specific datasets. The resulting models retain the baseline's generalization ability while gaining class-specific accuracy and produce probabilistic outputs that capture uncertainty in the training data. We demonstrate the approach on three toxicity classes defined by a common core structure, target, or mode of action: (i) organophosphates, (ii) androgen receptor antagonists, and (iii) estrogen receptor beta antagonists. Each fine-tuned model outperforms classical machine-learning methods and the EPA TEST tool. The probabilistic nature of the predictions enables prioritization of compounds for experimental validation and seamless integration with data streams of varying quality, supporting iterative decision-making in chemical safety and drug discovery.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Predicting Toxicity and Bioactivity of the Chemical Exposome: A Case Study for the Blood Exposome Database 97%
- Merging Bioactivity Predictions from Cell Morphology and Chemical Fingerprint Models Using Similarity to Training Data 92%
- DrugDiff - small molecule diffusion model with flexible guidance towards molecular properties 92%
Similar papers in this journal
Similar papers in this journal
Similar papers in this journal
- Enabling Transparent Toxicokinetic Modeling for Public Health Risk Assessment 93%
- Distinguishing classes of neuroactive drugs based on computational physicochemical properties and experimental phenotypic profiling in planarians 93%
- Repurposing Therapeutics for COVID-19: Rapid Prediction of Commercially available drugs through Machine Learning and Docking 92%
Similar papers in this journal
- AnimalGAN: A Generative Adversarial Network Model Alternative to Animal Studies for Clinical Pathology Assessment 95%
- Predicting compound activity from phenotypic profiles and chemical structures 92%
- CurveCurator: A recalibrated F-statistic to assess, classify, and explore significance of dose-response curves 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.