Metabolic syndrome is a cluster of conditions involving various metabolic disorders, significantly increasing the risk of hyperlipidemia, diabetes, hypertension, obesity and cardiovascular disease. The global prevalence of metabolic syndrome is on the rise, with over one-third of the U.S. population affected. Given its complex cause of onset, it is likely that metabolic syndrome interacts with a wide range of other diseases.
In this study, we apply a generative deep learning model, Variational Autoencoder(VAE), to analyze proteomics and metabolomics data from individuals with metabolic syndrome. Our goal is to explore potential associations between metabolic syndrome and other diseases through multi-omics integration.
We have identified candidate biomarkers associated with metabolic syndrome using external cohort data and are currently planning to validate these findings in the UK Biobank dataset.
