Reconciling multiple connectivity scores for drug repurposing.

Area of Science:

• Computational biology
• Pharmacology
• Bioinformatics

Background:

• Drug repurposing relies on reversing disease molecular signatures.
• The 'connectivity map' study (2006) introduced disease-drug connectivity scores.
• Numerous studies have since proposed variations in calculating these scores.

Purpose of the Study:

• To systematically reconcile diverse disease-drug similarity metrics and connectivity scores.
• To define these metrics using consistent notation and terminology.
• To provide a unified scheme for computational drug development.

Main Methods:

• Reconciliation of multiple disease-drug similarity metrics (e.g., ES, css, Cosine).
• Systematic definition of various connectivity scores (e.g., CS, RGES, NCS).
• Consistent notation and terminology applied across all metrics.

Main Results:

• A coherent definition of multiple connectivity scores and their relationships.
• Clarified insights into disease-drug similarity calculations.
• Established a unified framework for comparing different scoring approaches.

Conclusions:

• The unified scheme facilitates easier comparison and investigation of computational drug development methods.
• Provides a foundation for integrating newer methods transparently.
• A live document and GitHub repository are available for community contribution.