Cross-species etiologically informed stratification enables T cell receptor-based diagnosis of Parkinson’s Disease

Developing peripheral blood-based diagnostic models for idiopathic Parkinson’s disease (iPD), particularly those leveraging the T-cell receptor (TCR) repertoire, has long been considered infeasible because patient-derived TCRs appear to lack convergent sequence motifs. We reasoned that this apparent absence of shared TCR features likely reflects both insufficient sample sizes and unaccounted immune heterogeneity within the iPD population. To overcome these barriers, we reconstructed TCR repertoires from the two largest iPD cohorts currently available and used three mechanistically distinct mouse models as etiologically informed anchors to molecularly stratify patients. Within these data-driven subtypes, we trained subtype-specific, multimodal multi-instance classifiers, achieving Area Under Curve (AUC) exceeding 0.8 for both subtypes. Our findings underscore the critical role of disease stratification in enabling TCR repertoire–based modeling in neurodegenerative diseases. By leveraging immune trajectories from distinct mouse models to stratify patients, we achieved immune repertoire-based diagnosis of Parkinson’s disease via deep learning.