Integrating Long-Read Structural Variant Analysis with single-nucleus RNA-seq to Elucidate Gene Expression Effects in Disease

Structural variants (SVs) are a major source of genetic diversity, yet how they impact cell types in complex brain diseases remains largely unexplored, partially due to limitations of short-read sequencing. Here, we addressed this fundamental question in Parkinson's disease (PD). generating long-read whole-genome sequencing (WGS) data for 100 post-mortem brain samples from a PD cohort, constructing a high-confidence catalog of 74,552 SVs. To resolve their functional impact, we integrated single-nucleus RNA-sequencing data from two brain regions from the same samples and focused functional analyses on SVs proximal to genes previously nominated as cis-regulated, potential causal targets of PD-associated GWAS loci. Using expression quantitative trait locus and allele-specific expression analyses, we uncovered SVs significantly associated with expression in specific cell types as well as effects shared across cell types. This study demonstrates the power of uniting long-read WGS with transcriptomics to uncover SVs underlying complex disease architecture with cell type resolution.