Extraction of tacit knowledge from large ADME data sets via pairwise analysis.

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Pharmacokinetics

Background:

• Pharmaceutical companies gather extensive ADME data across projects.
• Cross-project data mining can uncover broader structure-activity relationships (SAR).
• Pairwise analysis is effective for identifying subtle structural changes impacting activity.

Purpose of the Study:

• To detail the process of full pairwise analysis for high-throughput ADME assays.
• To present strategies for applying pairwise analysis in prospective compound design.
• To analyze activity patterns arising from molecular transformations.

Main Methods:

• Full pairwise analysis of high-throughput ADME assays (microsomal clearance, permeability, efflux, lipophilicity).
• Development of strategies for prospective application in compound design.
• Analysis of compound pairs sharing molecular transformations.

Main Results:

• Identification of various SAR-relevant molecular transformations: bioisosteres, additives, multiplicatives, and switches.
• Demonstration of pairwise analysis's ability to detect significant activity changes from small structural modifications.
• Validation of prospective application strategies for compound design.

Conclusions:

• Pairwise analysis is a valuable tool for extracting cross-project knowledge from ADME data.
• Understanding transformation patterns enhances predictive SAR modeling.
• This approach can optimize compound discovery and development by guiding structural modifications.