In vitro drug sensitivity-gene expression correlations involve a tissue of origin dependency.

Area of Science:

• Chemogenomics
• Systems Biology
• Pharmacology

Background:

• Chemogenomics aims to link drug activity with biological variables.
• Clustering drug activity profiles and gene expression data yields different groupings.
• These groupings, however, still reflect the drugs' mechanisms of action.

Purpose of the Study:

• To explain the discrepancies observed between different clustering methods in chemogenomics.
• To propose a novel interpretation of these differences in terms of weighted drug-drug distances.

Main Methods:

• Analysis of cell line drug activity profiles.
• Correlation analysis between drug activity and gene expression profiles.
• Developing a weighted distance metric for drug comparisons.

Main Results:

• Discrepancies in clustering arise from differential weighting of drug-drug distances.
• The weights are determined by the specific cell lines where drug activity profiles diverge.
• This provides a unified view of different chemogenomic clustering approaches.

Conclusions:

• The observed differences in chemogenomic clustering are not contradictory but complementary.
• A weighted distance approach reconciles various clustering outcomes.
• This enhances the understanding of drug mechanisms and biological associations.