Risk Prediction and Subtyping

COPD is a heterogeneous disease, likely comprised of multiple subtypes. We have used machine learning methods to identify COPD subtypes using lung function, imaging, and omics data. We have also created polygenic risk scores for COPD, pulmonary fibrosis, and asthma, and multiomic risk scores for COPD. Our risk scores can identify high risk groups that are likely to develop these chronic lung diseases and those who will experience significant disease progression.