Gene set variation analysis of mRNA expression in bronchial biopsies from the U-BIOPRED asthma study

Improving our understanding of the molecular pathways involved in asthma may facilitate better stratification of endotypes. The aim of the present study was to apply Gene Set Variation Analysis to microarray data from the U-BIOPRED cohort, in order to estimate the variation of specific gene sets in asthma. Endobronchial biopsies from 150 subjects distributed between mild-moderate, smoker and non-smoker severe asthma, and healthy controls were analysed. We utilised gene signatures of distinct immune cell type activation, of response to therapy, from different tissues and diseases. Signature enrichment in patient transcriptomes can be represented as a dot plot. The natural killer (NK) signature is equally distributed across cohorts except for an evident split in the non-smoker severe asthma group revealing potential patient subgroups (Figure 1, upper panel). The Th2 signature is elevated in subsets of each cohort (Figure 1, lower panel). Enrichment scores can be subjected to further classical statistical analysis to explore association with clinical features. We conclude that the use of gene signature variation for patient stratification provides an alternative to classical differential gene and pathway expression analysis, which aim to identify differences between predefined patient subgroups. Our preliminary results demonstrate immune pathology subsets in asthma patients. This work was funded by IMI.