Integrating airway microbiome and blood proteomics data to identify multi-omic networks associated with response to pulmonary infection

Abstract

Host response to airway infections can vary widely. Cystic fibrosis (CF) pulmonary exacerbations provide an opportunity to better understand the interplay between respiratory microbes and the host. This study aimed to investigate the observed heterogeneity in airway infection recovery by analyzing microbiome and host response (i.e., blood proteome) data collected during the onset of 33 pulmonary infection events. We used sparse multiple canonical correlation network (SmCCNet) analysis to integrate these two types of -omics data along with a clinical measure of recovery. Four microbe–protein SmCCNet subnetworks at infection onset were identified that strongly correlate with recovery. Our findings support existing knowledge regarding CF airway infections. Additionally, we discovered novel microbe–protein subnetworks that are associated with recovery and merit further investigation.