Genetic association studies have uncovered novel disease variants for many complex diseases, but translating association results into biologically actionable insights has been proportionally slower, in part due to the challenges of designing follow-up experiments. We frequently collaborate with wet-lab researchers and have found that, when designing validation experiments, they are most interested in disease- and tissue-specific findings related to genes. We created Reducing Associations by Linking Genes And omics Results (REALGAR), a user-friendly R Shiny app to help guide experimental validation. Using asthma as a disease model, REALGAR combines several omics datasets for which users are able to view gene-centric results in a tissue-specific, disease-focused manner. Specifically, REALGAR inputs consist of GWAS results, including those from the EVE and GABRIEL asthma consortia, microarray and RNA-Sequencing results we obtained by applying RAVED to data in the Gene Expression Omnibus, and an asthma-related ChIP-Seq dataset from ENCODE. REALGAR can help users better understand GWAS findings and re-prioritize GWAS hits for further experimentation.
To use REALGAR, a user selects inputs of interest—gene name or SNP ID, tissue(s), phenotype(s) and which GWAS results to display. REALGAR output includes forest plots and genome tracks. Forest plots show gene fold-changes of differential expression for the datasets selected, with colors corresponding to q-values. The forest plots include a combined meta-analysis score, which REALGAR computes on-the-fly based on the gene expression datasets selected. Genome tracks show gene transcripts, glucocorticoid-receptor binding sites and SNPs with GWAS results. The plots and underlying results can be downloaded from the app.