AbstractBackgroundLarge genome-wide association studies (GWAS) and other genetic studies have revealed genetic loci that are associated with chronic obstructive pulmonary disease (COPD). However, the proteins responsible for COPD pathogenesis remain elusive. We used integrative-omics by combining genetics of lung function and COPD with genetics of proteome to identify proteins underlying lung function variation and COPD risk.MethodsWe used summary statistics from the GWAS of human plasma proteome from the INTERVAL cohort (n=3,301) and integrated these data with lung function GWAS results from the UK Biobank cohorts (n=400,102) and COPD GWAS results from the ICGC cohort (35,735 cases and 222,076 controls). We performed in parallel: a proteome-wide Bayesian colocalization, and a proteome-wide Mendelian Randomization (MR) analyses. Next, we selected proteins that colocalized with lung function and/or COPD risk and explored their causal association with lung function and/or COPD using MR analysis (P<0.05).ResultsWe found 537, 607, and 250 proteins that colocalized with force expiratory volume in one second (FEV1), FEV1/forced vital capacity (FVC), or COPD risk, respectively. Of these, 1,051 were unique proteins. The sRAGE protein demonstrated the strongest colocalization with FEV1/FVC and COPD risk, while QSOX2, FAM3D and F177A proteins had the strongest associations with FEV1. Of these, 37 proteins that colocalized with lung function and/or COPD, also had a significant causal association. These included proteins such as PDE4D, QSOX2 and RGAP1, amongst others.ConclusionIntegrative-omics reveals new proteins related to lung function. These proteins may play important roles in the pathogenesis of COPD.
Network Analysis of Genome-Wide Association Studies for Chronic Obstructive Pulmonary Disease in the Context of Biological Pathways
