API src

By integrating previously overlooked drivers AI boosts bioaccumulation assessment in fish

Description: The increasing use of chemicals has led to the integration of the bioconcentration factor (BCF) into chemical regulation. A machine learning model trained on chemical properties and test conditions for 962 chemicals enabled us to estimate experimental BCFs in 16 fish species with up to 90 % accuracy. We showed that the BCF is not, as generally assumed, a fixed chemical-specific criterion, but increases with lower exposure concentration and longer exposure duration. A review of 165 regulatory studies submitted to EU authorities showed that about half of the chemicals that should have been classified as bioaccumulative according to worst-case criteria defined in our study were not so classified based on the experimental BCFs submitted. We therefore propose the implementation of two new objective metrics, BCF50 for minimum requirement bioaccumulation screening and BCF90 as a realistic worst-case surrogate, into regulation. © 2026 by the Authors

Global identifier:

Doi(
    "10.60810/openumwelt-8690",
)

Types:
Text(
    Publication,
)

Tags: Fisch ? Bioakkumulation ? Biokonzentrationsfaktor ? Chemikalien ? Künstliche Intelligenz ? Studie ? 500 Naturwissenschaften und Mathematik::540 Chemie ? Erfassung ? Organics ?

License: Creative Commons Namensnennung 4.0

Language: Englisch/English

Organisations

Persons

Issued: 2026-01-01

Last harvest: 04.09.2026 01:32

Status

Quality score

Accessed 1 times.