Assessing chemical toxicity across Eukaryota using multimodal transformers
Käll, S.; Gustavsson, M.; Svedberg, P.; Dalman, M.; Henriksson, J.; Backhaus, T.; Kristiansson, E.
Show abstract
Biodiversity is globally threatened by chemical pollution, yet toxicity data remain unavailable for millions of species and tens of thousands of chemicals, severely limiting our ability to assess ecological impacts. Here we present TRIDENT-2, a multimodal artificial intelligence model for predicting chemical toxicity across evolutionarily diverse eukaryotic species. Trained on 560,780 toxicity assays spanning 82,775 chemicals, 6,793 species, and multiple exposure scenarios, TRIDENT-2 accurately predicts toxicity across Eukaryota with an average median absolute error ranging from 1.76 to 3.80. By jointly learning from chemical, biological, and experimental information, it remains accurate across broad chemical and taxonomic distances, allowing for toxicity assessment for species and chemicals beyond the current experimental evidence. Our findings demonstrate that artificial intelligence can help overcome longstanding data limitations in ecotoxicology, paving the way for improved decision-making and reducing chemical impacts on biodiversity and ecosystems.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Genetic variability in pathways associates with pesticide-induced nervous system disease in the United States 92%
- Multi-behavioral phenotyping in early-life-stage zebrafish for identifying disruptors of non-associative learning 92%
- Wastewater-based estimation of the effective reproductive number of SARS-CoV-2 89%
Similar papers in this journal
- AnimalGAN: A Generative Adversarial Network Model Alternative to Animal Studies for Clinical Pathology Assessment 92%
- Altruistic disease signalling in ant colonies 91%
- Coupling comprehensive pesticide-wide association study to iPSC dopaminergic screening identifies and classifies Parkinson-relevant pesticides 91%
Similar papers in this journal
- Missing genomic resources for the next generation of environmental risk assessment 93%
- Impacts of an urban sanitation intervention on fecal indicators and the prevalence of human fecal contamination in Mozambique 91%
- Variation in the transcriptome response and detoxification gene diversity drives pesticide tolerance in fishes 90%
Similar papers in this journal
- Mutagenic and carcinogenic potency determinations for NDMA support the cumulative dose assumption underpinning the less-than-lifetime Threshold of Toxicological Concern 93%
- High-throughput PBK modelling for dermal exposure: a pragmatic approach to predict systemic pharmacokinetics 92%
- Leveraging integrative toxicogenomic approach towards development of stressor-centric adverse outcome pathway networks for plastic additives 92%
Similar papers in this journal
- Gaussian graphical modeling of the serum exposome and metabolome reveals interactions between environmental chemicals and endogenous metabolites involved in critical metabolic processes 92%
- DNA epigenetic marks are linked to reproductive aberrations in amphipods 91%
- First national survey of terrestrial biodiversity using airborne eDNA 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.