Study questions network-pharmacology targets in dilated cardiomyopathy using single-nucleus RNA sequencing

A secondary analysis of cardiac datasets reports that candidate genes nominated via network pharmacology for dilated cardiomyopathy do not show cell-resolved transcriptional enrichment above the genome-wide background, underscoring the limitations of literature-derived drug target nomination.

The authors reported that a direct-evidence intersection of targets nominated for Yangxin Decoction in dilated cardiomyopathy (DCM) yielded 61 candidate genes.

According to the study, bulk cardiac datasets showed no enrichment of these candidates above the transcriptome-wide background, and AKT1 was not significant in any bulk cohort.

The researchers analyzed single-nucleus RNA sequencing data from 52 DCM and 18 control donors, finding cardiomyocyte-resolved changes in AKT1 (log2 fold change of 0.351, adjusted P = 3.32 × 10^-4) and STAT3 (which replicated at -0.771 in an independent cohort), but noted that the AKT1 fold-change was small.

An unbiased genome-wide comparison revealed that the nominated 61-gene set was not significantly enriched in cardiomyocytes, showing a 46.7% match versus a 41.3% background (P = 0.281), with AKT1 ranking 2,182 out of 14,586 genes.

The study demonstrates that network-pharmacology methods can nominate drug targets that lack robust, cell-resolved alteration in diseased tissues or enrichment above a random genome-wide background.

The findings suggest that therapeutic mechanisms nominated through literature networks require rigorous, cell-type-resolved transcriptomic validation to avoid pursuing false-positive translational pathways in complex cardiac diseases.

For developers of age-related cardiovascular therapeutics, the study underscores the limitations of relying on unvalidated network biology or bulk tissue assays for target identification.

The results suggest that asset-validation pipelines targeting cardiomyopathy must incorporate single-nucleus resolution to distinguish genuine cell-type-specific therapeutic mechanisms from background transcriptional noise.