Ensemble genomic analysis in human lung tissue identifies novel genes for chronic obstructive pulmonary disease

Jarrett D Morrow,John Quackenbush,Nathaniel Marchetti,George R Washko,Craig P Hersh,Weiliang Qiu,Xiaobo Zhou,Raphael Bueno,Bartholome Celli,John Platig,Kimberly Glass,Edwin K Silverman,Gerard J Criner,Dawn L Demeo,Michael H Cho

doi:10.1186/s40246-018-0132-z

Abstract

BackgroundGenome-wide association studies (GWAS) have identified single nucleotide polymorphisms (SNPs) significantly associated with chronic obstructive pulmonary disease (COPD). However, many genetic variants show suggestive evidence for association but do not meet the strict threshold for genome-wide significance. Integrative analysis of multiple omics datasets has the potential to identify novel genes involved in disease pathogenesis by leveraging these variants in a functional, regulatory context.ResultsWe performed expression quantitative trait locus (eQTL) analysis using genome-wide SNP genotyping and gene expression profiling of lung tissue samples from 86 COPD cases and 31 controls, testing for SNPs associated with gene expression levels. These results were integrated with a prior COPD GWAS using an ensemble statistical and network methods approach to identify relevant genes and observe them in the context of overall genetic control of gene expression to highlight co-regulated genes and disease pathways. We identified 250,312 unique SNPs and 4997 genes in the cis(local)-eQTL analysis (5% false discovery rate). The top gene from the integrative analysis was MAPT, a gene recently identified in an independent GWAS of lung function. The genes HNRNPAB and PCBP2 with RNA binding activity and the gene ACVR1B were identified in network communities with validated disease relevance.ConclusionsThe integration of lung tissue gene expression with genome-wide SNP genotyping and subsequent intersection with prior GWAS and omics studies highlighted candidate genes within COPD loci and in communities harboring known COPD genes. This integration also identified novel disease genes in sub-threshold regions that would otherwise have been missed through GWAS.

Highlights

Expression quantitative trait locus analysis tests the association between genetic variants and gene expression and can point to relevant single nucleotide polymorphisms (SNPs) and genes within Genome-wide association studies (GWAS) loci [15,16,17] using the observation that trait-associated SNPs are likely to be eQTLs/eQTL SNPs (eSNPs) [17] and/or have gene regulatory implications [18]
This integrative method extracts the additional genetic and genomic signals contained in the sub-threshold SNPs by combining evidence across genotyping, gene expression and DNA methylation datasets and highlights novel genes and loci within regions that may not have been identified through GWAS
We identified eQTLs using the gene expression and imputed genotyping data and integrated them with prior GWAS and omics studies using an ensemble approach of statistical and network methods (Fig. 1)

Summary

Results

Gene expression data were available for lung tissue samples from 86 severe COPD cases (mean FEV1 26.4% predicted) and 31 controls with normal spirometry, all Caucasians (Additional file 1: Table S1). To observe overall genetic control of gene expression in a disease context, we intersected all cis-eQTL results with the nominally significant GWAS SNPs (p < 0.05) [8] and plotted the p values from the two sets (Fig. 2). We further sought to place our 50 Sherlock-derived genes in the context of overall genetic control of gene expression using network methods, since co-regulated genes may have shared function This process has the potential to reveal additional COPD genes of interest. LBF logarithm of Bayes factor *Gene identified in the cis-eQTL-GWAS intersection in (Additional file 1: Table S4). Seven communities were validated based on nominally significant (meta-p < 0.05) differential expression and differential methylation results (Table 2) These communities contain the Sherlock-derived genes CDH23, CHRNA5, HNRNPAB, IREB2, PCBP2, ZNF652, ACVR1B, and RPL23A Six of the nine remaining communities, which were lacking joint evidence, had either nominally significant differential expression or differential methylation

Conclusions

Background

Discussion

Methods