Pangenome-resolved deconvolution of gut metatranscriptomes reveals species-specific microbial activity programs in Parkinson’s disease

Gut microbiome alterations are associated with Parkinson’s disease (PD), yet their species-level functional consequences remain poorly understood. Here, we develop a pangenome-resolved framework integrating paired metagenomic and metatranscriptomic data to reconstruct species-level transcriptional profiles. Applied to two independent cohorts comprising 160 individuals and 7,548 profiles across 215 species, this approach revealed reduced genome-wide transcriptional breadth and increased inter-individual heterogeneity in PD. Genomic diversity and transcriptomic variation were significantly associated in 61 of 95 tested species. We identified 2,271 pangenome-resolved features associated with disease status, duration, or severity, with 39% of significant gene-level associations involving accessory genes. Species-specific responses included reduced metabolic and biosynthetic expression in commensals, including Coprococcus aceti and Streptococcus thermophilus, alongside increased stress-adaptation, mucin-degradation, and metal-homeostasis functions in other taxa. Cross-cohort classification performance varied substantially between species. These findings establish pangenome-resolved metatranscriptomics as a framework for identifying lineage-specific functional responses within microbiomes.