An End Point-Specific Framework for Read-Across Analog Selection for Human Health Effects.

Area of Science:

• Computational toxicology
• Chemical safety assessment
• Read-across methodology

Background:

• The read-across analog approach compares chemical structures to predict toxicity, filling data gaps in safety assessments.
• Advancements focus on chemical grouping and clustering to improve toxicity predictions.
• Analog selection requires considering various properties like physicochemical, toxicokinetic, and toxicodynamic factors, emphasizing mechanisms of action.

Purpose of the Study:

• To present six end point-specific, rule-based schemes for selecting read-across analogs.
• To establish transparent workflows for data gap filling in chemical safety assessment.
• To accelerate the safety assessment process by systematically driving analog selection.

Main Methods:

• Integration of computational chemistry and toxicology principles.
• Development of six distinct rule-based schemes tailored to specific human health endpoints (genotoxicity, skin sensitization, etc.).
• Creation of end point-specific workflows for analog selection and data gap imputation.

Main Results:

• Description of six novel, end point-specific schemes for read-across analog selection.
• Establishment of transparent workflows supporting analog selection rationale for specific endpoints.
• Demonstration of a systematic framework to drive analog selection.

Conclusions:

• The proposed framework systematically drives read-across analog selection for specific endpoints.
• These end point-specific schemes enhance transparency and rigor in chemical safety assessments.
• The integration of computational approaches accelerates the overall safety evaluation process.