Classification of compounds by cluster analysis of Ames test data

Area of Science:

• Toxicology
• Computational Biology
• Genetics

Background:

• The Ames test is a standard method for assessing chemical mutagenicity.
• Analyzing dose-response curves is crucial for understanding compound effects.
• Existing methods may not fully account for compound toxicity in mutagenicity assessments.

Purpose of the Study:

• To develop and validate a computational method for analyzing Ames test dose-response curves.
• To integrate compound toxicity into mutagenicity assessments.
• To classify compounds based on their mutagenic activity and dose-response characteristics.

Main Methods:

• Utilized a theoretical equation m(x) = (B + Sx)exp(-Tx) to model dose-response.
• Employed the least-squares method with the SALS program package to determine parameters B (background), S (mutagenicity), and T (toxicity).
• Applied cluster analysis to the determined parameters for compound classification.

Main Results:

• The combined least-squares and cluster analysis method successfully separated mutagenic and non-mutagenic compounds.
• Cluster analysis into six groups revealed characteristic parameter values for each group.
• The method provides a robust classification of compounds based on mutagenic activity and dose-response curve nature.

Conclusions:

• The developed procedure is highly effective for classifying compounds according to their mutagenic activity.
• Integrating toxicity parameters enhances the accuracy of mutagenicity assessment.
• This approach offers a valuable tool for toxicological and drug discovery research.