Development of a data-driven approach to Adverse Outcome Pathway network generation: a case study on the...

Area of Science:

• Toxicology and Chemical Risk Assessment
• Computational Biology and Bioinformatics
• Endocrinology

Background:

• Adverse Outcome Pathways (AOPs) are crucial for understanding toxicological mechanisms and integrating new data into risk assessments.
• AOP Networks (AOPNs) offer a more comprehensive representation of complex biological systems than individual AOPs.
• Current methods for generating AOPNs lack standardization, necessitating systematic approaches.

Purpose of the Study:

• To develop a structured search strategy for identifying relevant AOPs within the AOP-Wiki.
• To create an automated, data-driven workflow for generating AOPNs.
• To apply and demonstrate the approach through a case study focusing on Estrogen, Androgen, Thyroid, and Steroidogenesis (EATS) pathways.

Main Methods:

• A priori development of a search strategy using effect parameters from ECHA/EFSA guidance.
• Manual curation of AOP-Wiki content to identify and exclude irrelevant pathways.
• Automated data processing, filtering, and formatting using a computational workflow.
• Application of the workflow to generate an EATS-focused AOPN.

Main Results:

• A structured search strategy and an automated workflow for AOPN generation were successfully developed.
• A case study yielded an AOPN map for EATS modalities, facilitating further research.
• The computational approach is available as an R-script for regenerating and filtering AOP networks.

Conclusions:

• The developed approach provides a standardized method for AOPN generation, enhancing the integration of mechanistic toxicological data.
• This work supports mechanism-based identification of endocrine disruptors by mapping relevant AOPs.
• The freely available R-script promotes reproducible research and application in chemical risk assessment.