An expert system rulebase for identifying contact allergens

Area of Science:

• Toxicology
• Computational Chemistry
• Dermatology

Background:

• Current in vitro methods are insufficient for prospective identification of skin sensitizers (contact allergens).
• Expert systems can integrate chemical structure-toxicity knowledge.
• The guinea pig maximization test (GPMT) is a standard method for assessing skin sensitization potential.

Purpose of the Study:

• To develop a computational expert system for the prospective identification of skin sensitizers.
• To derive structure-activity rules for skin sensitization based on existing data.
• To enhance the strategic approach for identifying contact allergens.

Main Methods:

• Utilized a historical database of 294 substances tested using the guinea pig maximization test (GPMT).
• Grouped substances by likely reaction mechanisms with skin proteins.
• Derived empirical structure-activity rules for chemical functionalities lacking identified mechanisms.
• Incorporated 40 derived rules into the DEREK expert system.

Main Results:

• Successfully derived 40 structure-activity rules for skin sensitization.
• Integrated these rules into the DEREK expert system.
• Established a foundation for prospective identification of contact allergens.

Conclusions:

• The DEREK expert system, equipped with 40 structure-activity rules, represents a significant advancement in the prospective identification of skin sensitizers.
• This computational approach aids in the strategic identification of contact allergens, potentially reducing reliance on animal testing.
• Further development of such expert systems is crucial for improving in vitro toxicological assessments.