Integrative Model Network for Chemical Risk Assessment

Integrative chemical risk assessment plays a key role in understanding the health and environmental impacts of chemical exposures. A crucial step in this process is the efficient integration of models and data across various disciplines, including exposure science, epidemiology, and toxicology. This integration supports decision-making in both research and regulatory contexts. 

Effective implementation of integrative risk assessment requires user-friendly workflows that harmonise and connect diverse modeling tools and datasets. However, the lack of standardised approaches, data harmonisation, and interoperability among tools poses significant challenges. PARC aims to address these gaps by developing a network of interoperable modeling tools and data resources that can be seamlessly used in workflows for chemical risk assessment.

The integrative model network is designed with long-term sustainability in mind. By aligning with existing initiatives, such as the PARC Chemical Risk Assessment (CRA) Hub and other potential external e-infrastructures, the network aims to:

  • support harmonised, scalable, and cross-sector integrative risk assessments across Europe, in alignment with international guidance and regulatory requirements such as GDPR, transparency regulations, and regulatory (e.g. EFSA, ECHA) methodologies, as well as NGRA approaches from PARC,
  • provide a structured overview of selected modeling tools and workflows suitable for addressing risk assessment questions
  • develop and operationalise user-friendly, generic workflows that can span multiple domains and link modeling tools and datasets to support complex integrative risk assessments,
  • enhance the interoperability and reusability of selected modeling tools to enable their efficientm practical implementation in integrative assessments,
  • foster cross-sector collaboration,
  • reduce redundancy in tool development and application.
PARC model network

 

Iterative implementation and case studies 

The development of the integrative model network follows a pragmatic, iterative approach, combining both bottom-up and top-down strategies:

  • Bottom-up: Practical workflows and model links are developed within PARC case studies, addressing immediate needs in integrative assessments.
  • Top-down: A general conceptual and technical framework for model integration is continuously refined based on insights gained from ongoing applications in case studies.

Current workflows include:

  • HBM-based risk assessment (including mixtures): leveraging measured exposures from human biomonitoring studies to support (mixture) exposure risk assessment.
  • Aggregate exposure assessment: combining exposure assessment models and toxicokinetic models to aggregate exposures from multiple sources and routes.
  • New Approach Methodologies (NAMS): tools that allow the assimilation of NAMs data for hazard metrics to be used in chemical risk assessment.
  • Quantification of uncertainty: developing approaches to assess and systematically propagate uncertainty across interconnected models and modeling tools.
     

Workflows and modeling tools

A model inventory is currently under development to improve the findability and accessibility of modeling tools and workflows within the model network. In alignment with the PARC CRA Hub, this inventory includes metadata schemas tailored to harmonise tool and workflow descriptions, enhancing their FAIRness (Findable, Accessible, Interoperable, Reusable). 

Sustainability and future outlook

By enabling transparent, consistent, and FAIR workflows, the PARC integrative model network is poised to become a cornerstone of modern chemical risk assessment, bridging research and regulatory practices through a shared, interoperable digital ecosystem. The current version of the model network features a curated selection of modeling tools that support various aspects of risk assessment. As the network evolves, additional tools and functionalities will be integrated to support a broader range of use cases and user needs.