Searching for an enhanced predictive tool for mutagenicity.

Area of Science:

• Computational toxicology
• Medicinal chemistry
• Genotoxicity assessment

Background:

• The Ames test is a standard for assessing chemical mutagenicity.
• Predicting mutagenic potential computationally aids drug discovery and safety evaluation.
• Existing models often rely on single datasets, limiting predictive power.

Purpose of the Study:

• To evaluate the mutagenic potential of organic compounds using the MCASE program.
• To develop and refine computational models for predicting mutagenicity based on chemical structure.
• To identify structural features (biophores) responsible for mutagenic activity.

Main Methods:

• Collected experimental Ames test mutagenicity data for 2513 chemicals.
• Utilized the Multiple Computer Automated Structure Evaluation (MCASE) program for analysis.
• Compiled a general mutagenicity dataset and 15 individual Salmonella test strain datasets.
• Developed multiple-database mutagenicity models.

Main Results:

• Achieved good correlations between chemical structure and mutagenic activity.
• Demonstrated significant improvement in prediction accuracy with increased data.
• Identified specific biophores associated with mutagenic activity across different strains.
• The multiple-database model outperformed single-database models.

Conclusions:

• The MCASE program is a valuable tool for predicting the mutagenic potential of organic compounds.
• Computational models incorporating multiple datasets enhance prediction accuracy.
• Identification of biophores provides insights into mechanisms of mutagenicity.
• This approach can guide the development of safer chemicals.