Hasty Briefsbeta

Bilingual

Hybrid experimental-QSAR-artificial intelligence framework for chronic ecological risk assessment of emerging pollutants in coastal environments - PubMed

8 hours ago
  • #coastal pollutants
  • #ecological risk assessment
  • #QSAR-AI framework
  • A hybrid framework combines experimental data, QSAR tools, and AI to assess chronic ecological risks of emerging pollutants in coastal environments.
  • Experimental results identified acetaminophen and caffeine as high-risk substances, with diclofenac posing negligible risk.
  • QSAR screening highlighted caffeine, losartan, and benzoylecgonine as priority pollutants.
  • VEGA analysis showed reliable predictions for about 65-70% of compound-taxon combinations.
  • AI model TRIDENT moderated extreme QSAR outputs, confirming high risks for caffeine, acetaminophen, and losartan.
  • Algae were the most sensitive taxonomic group, followed by crustaceans and fish.
  • The framework enhances ecological risk assessment in data-limited tropical coastal areas and is internationally transferable.