Abstract:Despite recent advances in chronic obstructive pulmonary disease (COPD) research, few studies have identified the potential therapeutic targets systematically by integrating multiple-omics datasets. This project aimed to develop a systems biology pipeline to identify biologically relevant genes and potential therapeutic targets that could be exploited to discover novel COPD treatments via drug repurposing or de novo drug discovery. A computational method was implemented by integrating multi-omics COPD data from unpaired human samples of more than half a million subjects. The outcomes from genome, transcriptome, proteome, and metabolome COPD studies were included, followed by an in silico interactome and drug-target information analysis. The potential candidate genes were ranked by a distance-based network computational model. Ninety-two genes were identified as COPD signature genes based on their overall proximity to signature genes on all omics levels. They are genes encoding proteins involved in extracellular matrix structural constituent, collagen binding, protease binding, actin-binding proteins, and other functions. Among them, 70 signature genes were determined to be druggable targets. The in silico validation identified that the knockout or over-expression of SPP1, APOA1, CTSD, TIMP1, RXFP1, and SMAD3 genes may drive the cell transcriptomics to a status similar to or contrasting with COPD. While some genes identified in our pipeline have been previously associated with COPD pathology, others represent possible new targets for COPD therapy development. In conclusion, we have identified promising therapeutic targets for COPD. This hypothesis-generating pipeline was supported by unbiased information from available omics datasets and took into consideration disease relevance and development feasibility.

Multi-Omics Analysis Identified Drug Repurposing Targets for Chronic Obstructive Pulmonary Disease