CCMDR

CCMDR maps clinical laboratory-derived drug effects to disease signatures to prioritize drug repurposing candidates using electronic health records (EHRs) and national survey data.

Key Features:

• Clinical Connectivity Mapping: Relates drug-associated laboratory result changes to disease-associated laboratory signs to identify clinical connections between drugs and diseases.
• Clinical Drug Effect Vectors: Constructs vectors representing associations between drugs and laboratory results using a continuous self-controlled case series model applied to longitudinal EHR data.
• Clinical Disease Sign Vectors: Generates vectors capturing relationships between diseases and laboratory results using the Wilcoxon rank sum test on large-scale national survey data.
• Repurposing Possibility Score: Computes a score for each drug–disease pair via a dot product-based scoring function between clinical disease sign vectors and clinical drug effect vectors.
• Comprehensive Evaluation: Assesses the framework on 392 drugs across six chronic diseases: asthma, coronary heart disease, congestive heart failure, heart attack, type 2 diabetes, and stroke.

Scientific Applications:

• Drug repurposing candidate identification: Prioritizes existing drugs for new indications by matching drug-induced laboratory effect vectors to disease sign vectors.
• Translational bridging: Uses clinical laboratory results to connect preclinical signals with real-world clinical phenotypes for translational research.
• Explainable association discovery: Provides laboratory-result–based evidence to reveal known and novel drug–disease associations.

Methodology:

Integrates longitudinal EHR laboratory results and national survey laboratory data to build clinical drug effect vectors via a continuous self-controlled case series model and clinical disease sign vectors via Wilcoxon rank sum tests, then computes repurposing possibility scores with a dot product-based function and evaluates results on 392 drugs across six chronic diseases.