Improved Annotation of Asthma Gene Variants with Cell Type Deconvolution of Nasal and Lung Expression Quantitative Trait Loci

Asthma is a genetically complex inflammatory airway disease associated with more than 200 SNPs. However, the functional effects of many asthma-associated SNPs in lung and airway epithelial samples are unknown. Here, we aimed to conduct expression quantitative trait loci (eQTL) analysis using a meta-analysis of nasal and lung samples. We hypothesize that incorporating cell type proportions of airway and lung samples enhances eQTL analysis outcomes. Nasal brush (n = 792) and lung tissue (n = 1,087) samples were investigated separately. Initially, a general eQTL analysis identified genetic variants associated with gene expression levels. Estimated cell type proportions were adjusted based on the Human Lung Cell Atlas. In addition, the presence of significant interaction effects between asthma-associated SNPs and each cell type proportion was explored and considered evidence for cell type–associated eQTL. In nasal brush and lung parenchyma samples, 44 and 116 asthma-associated SNPs were identified as eQTL. Adjusting for cell type proportions revealed eQTL for an additional 17 genes (e.g., FCER1G, CD200R1, and GABBR2) and 16 genes (e.g., CYP2C8, SLC9A2, and SGCD) in nose and lung, respectively. Moreover, we identified eQTL for nine SNPs annotated to genes such as VASP, FOXA3, and PCDHB12 displayed significant interactions with cell type proportions of club, goblet, and alveolar macrophages. Our findings demonstrate increased power for identifying eQTL among asthma-associated SNPs by considering cell type proportion of the bulk RNA-sequencing data from nasal and lung tissues. Integration of cell type deconvolution and eQTL analysis enhances our understanding of asthma genetics and cellular mechanisms, uncovering potential therapeutic targets for personalized interventions.