Integrative transcriptome-based drug repurposing in tuberculosis

Tuberculosis (TB) remains the leading cause of death from infectious disease, with rising antibiotic resistance highlighting the need for host-directed therapeutics (HDTs). Transcriptome-based connectivity mapping offers a promising strategy by identifying drugs that reverse disease gene expression signatures, but current approaches are limited by reliance on single datasets, platforms, or scoring methods.

Here, we present a unified computational framework that systematically integrates heterogeneous transcriptomic data and multiple connectivity methods for robust drug prioritization. Our approach combines 28 TB gene expression signatures spanning microarray and RNAseq platforms, diverse cell types, and infection conditions, and applies multi-method connectivity scoring to identify consistent disease-drug reversal signals. By aggregating signals across datasets and methods, the framework captures dominant TB signatures while mitigating platform and biological variability. Using this integrative strategy, we prioritized 64 FDA-approved drugs as candidate HDTs, including previously reported host-directed therapeutic candidates such as statins and tamoxifen. Downstream pathway and network analyses further revealed enrichment in TB-relevant mechanisms and identified key bridging genes (e.g., IL-8, CXCR2) as potential therapeutic targets.

This work establishes transcriptome-based connectivity mapping as a viable approach for systematic HDT discovery in bacterial infections and provides a robust computational framework applicable to other infectious diseases. Our findings offer immediate opportunities for experimental validation of prioritized drug candidates and mechanistic investigation of identified druggable targets in TB pathogenesis.