Accounting for Experimental Variability in DEB-TKTD Modelling: Cross-Laboratory Calibration and Validation with Daphnia magna

Variability in standard toxicity tests between laboratories complicates the calibration and validation of toxicokinetic-toxicodynamic (TKTD) models. We assessed the robustness of a Dynamic Energy Budget (DEB)-TKTD model for Daphnia magna exposed to azoxystrobin under time-variable conditions, using four Good Laboratory Practice-compliant datasets from two laboratories employing harmonized Organisation for Economic Cooperation and Development (OECD) guideline 211 designs but different D. magna clones. Two datasets were used for calibration and two for validation. To address differences in control growth and reproduction, physiological parameters were estimated separately for each dataset, while TKTD parameters were obtained from joint and independent calibrations. Joint calibration identified a consistent physiological mode of action-hazard on reproduction-with strong statistical support (ΔAIC ≥ 10) and yielded TKTD parameters that reproduced both laboratories' data. Model validation met EFSA's NRMSE < 50% criterion for combined datasets (NRMSE: 4.0% for growth; 42.4% for reproduction). However, when evaluated individually, one validation dataset failed the reproduction criterion (NRMSE > 50%) due to markedly higher apparent sensitivity (EC10 up to 5.9× lower than others), despite remaining within the range of historically documented inter-laboratory variability. Simulated constant-exposure dose-response curves enabled direct comparison across experiments with differing exposure pulse profiles, revealing this dataset as an outlier without undermining overall model robustness. These results demonstrate that DEB-TKTD models can accommodate substantial inter-laboratory variability when physiological differences are explicitly modelled, and that combining datasets enhances parameter robustness. Our results also suggest that EFSA's model validation criteria should be revised to account for the level of experimental variability inherent to sublethal endpoints.

Authors

Institutions

Publication Details

Journal
Environmental Toxicology and Chemistry
Published
2026-11-17
DOI
https://doi.org/10.1093/etojnl/vgag294
Primary Topic
Environmental Toxicology and Ecotoxicology
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Accounting for Experimental Variability in DEB-TKTD Modelling: Cross-Laboratory Calibration and Validation with Daphnia magna

Erik B. Muller, Liubov Zakharova, Roman Ashauer, Neil Sherborne et al.
Environmental Toxicology and Chemistry
Environmental Toxicology and Ecotoxicology
article

Accounting for Experimental Variability in DEB-TKTD Modelling: Cross-Laboratory Calibration and Validation with Daphnia magna

Erik B. Muller, Liubov Zakharova, Roman Ashauer, Neil Sherborne, Carmen Börschig, Marie Trijau, Benoît Goussen, C. Romoli, Anja Coors, Philipp Emnet
article en

Abstract

Variability in standard toxicity tests between laboratories complicates the calibration and validation of toxicokinetic-toxicodynamic (TKTD) models. We assessed the robustness of a Dynamic Energy Budget (DEB)-TKTD model for Daphnia magna exposed to azoxystrobin under time-variable conditions, using four Good Laboratory Practice-compliant datasets from two laboratories employing harmonized Organisation for Economic Cooperation and Development (OECD) guideline 211 designs but different D. magna clones. Two datasets were used for calibration and two for validation. To address differences in control growth and reproduction, physiological parameters were estimated separately for each dataset, while TKTD parameters were obtained from joint and independent calibrations. Joint calibration identified a consistent physiological mode of action-hazard on reproduction-with strong statistical support (ΔAIC ≥ 10) and yielded TKTD parameters that reproduced both laboratories' data. Model validation met EFSA's NRMSE < 50% criterion for combined datasets (NRMSE: 4.0% for growth; 42.4% for reproduction). However, when evaluated individually, one validation dataset failed the reproduction criterion (NRMSE > 50%) due to markedly higher apparent sensitivity (EC10 up to 5.9× lower than others), despite remaining within the range of historically documented inter-laboratory variability. Simulated constant-exposure dose-response curves enabled direct comparison across experiments with differing exposure pulse profiles, revealing this dataset as an outlier without undermining overall model robustness. These results demonstrate that DEB-TKTD models can accommodate substantial inter-laboratory variability when physiological differences are explicitly modelled, and that combining datasets enhances parameter robustness. Our results also suggest that EFSA's model validation criteria should be revised to account for the level of experimental variability inherent to sublethal endpoints.

Environmental Toxicology and Chemistry
University of California, Santa Barbara (US), Syngenta (United Kingdom) (GB), ECT Oekotoxikologie (Germany) (DE), University of York (GB), Syngenta (Switzerland) (CH)
Environmental Toxicology and Ecotoxicology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon