Uncertainty-aware machine learning to predict non-cancer human toxicity for the global chemicals market
Technical University of Denmark · University of Cambridge · +5 more institutions
Abstract
Humans are exposed to thousands of chemicals, yet limited toxicity data hinder effective management of their impacts on human health. High-performing machine learning models hold potential for addressing this gap, but their uncharacterized prediction performance across the wider range of chemicals undermines confidence in their results. We develop uncertainty-aware models to predict reproductive/developmental and general non-cancer human toxicity effect doses. Our well-calibrated models provide uncertainty estimates aligned with observed prediction errors and chemical familiarity. We predict toxicity with 95% confidence intervals for >100,000 globally marketed chemicals and identify toxicity and uncertainty…
Citation impact
- FWCI
- 79.45
- Percentile
- 100%
- References
- 78
Authors
6Topics & keywords
- Chemical toxicity
- Human health
- Transparency (behavior)
- Risk assessment
- Predictive modelling
- Range (aeronautics)
- Gender equality