Matcher: An Open-Source Application for Translating Large Structure/Property Data Sets into Insights for Drug Design

Area of Science:

• Medicinal Chemistry
• cheminformatics
• Drug Discovery

Background:

• Matched molecular pair (MMP) analysis is crucial for understanding structure-activity relationships in drug discovery.
• Existing MMP analysis tools struggle with large datasets (>10,000 compounds), lacking flexibility and requiring computational expertise.
• There is a need for user-friendly, powerful tools to analyze complex chemical structure-property data.

Purpose of the Study:

• To introduce Matcher, an open-source application designed for efficient MMP analysis of large chemical datasets.
• To provide a no-code solution with novel search algorithms and automated visualization for MMP transformations.
• To enable researchers to easily explore structure-property relationships and accelerate drug discovery decisions.

Main Methods:

• Developed Matcher, an open-source application featuring novel search algorithms and automated querying-to-visualization.
• Integrated a chemical sketcher for user-defined control over MMP transformation searches based on fragment and environment structure.
• Demonstrated Matcher's utility with a public ChEMBL dataset of ~20,000 small molecules, analyzing CYP3A4 and hERG inhibition data.

Main Results:

• Matcher offers unprecedented control over MMP analysis, allowing flexible search and clustering of transformations.
• The application provides seamless navigation between MMPs, statistics, property graphs, and raw experimental data.
• Users can reproduce and share analyses using unique, shareable links within the Matcher interface.

Conclusions:

• Matcher democratizes MMP analysis, making large structure-property datasets more accessible and transparent.
• The tool significantly accelerates data-driven decision-making in drug discovery by simplifying complex analyses.
• Matcher is freely available, open-source, and containerized for easy deployment and use.