Identifying Subtypes of COPD Using Metabolomic and Genomic Approaches - Project Summary Chronic obstructive pulmonary disease (COPD) is a rising cause of mortality worldwide. Disease heterogeneity is one major barrier to understanding and therapeutic development for COPD. Understanding the molecular basis for this heterogeneity and determining the causes and consequences of the disease is a major challenge. Genetic variants, present since birth, have the potential to serve as a causal anchor for disease-related pathways and potential COPD subtypes. However, genetic variants often have small effect sizes and poorly understood functional effects. Metabolomics has emerged as a critical biomarker for lung disease but has not been assayed in lung and blood at a cohort level nor been connected to COPD genetic risk variants. In this proposal, we plan an integrative genomic approach to identify genetically driven pathways of COPD, informed by metabolomics, transcriptomics, and proteomics. First, we will curate existing and generate new genome-wide association studies for COPD-related phenotypes and implement statistical and machine learning methods to identify groups of variants displaying similar multi-phenotype association profiles. Second, to determine the molecular profiles of these groups of genetic variants, we will use existing transcriptomics, proteomics, and whole-genome sequencing data along with newly generated lung and blood metabolomics data in 1,000 subjects from the Lung Tissue Research Consortium (LTRC). We will also identify relationships between the new metabolites data, COPD and related phenotypes, and genetic variants and perform additional targeted metabolomics in 500 COPDGene subjects. Finally, we will assess the potential of these groups of variants to identify COPD subtypes, gene-environment interactions, and potential drug targets. Altogether, our integrative analysis will produce genetically informed molecular pathways and identify more specific groups of patients for therapy. In addition, genetic association, metabolomic, and methodologic resources generated for this project will be of value to the lung disease and complex trait genetics community.