Engineering genomic features to better understand the molecular basis of immune mediated diseases and their treatment

Genetic and transcriptional studies of immune-mediated diseases (IMD) have demonstrated complex sharing of features between these diseases, suggested new treatments, and collectively generated data on 100,000s of IMD patients. However, successes have been limited because huge sample sizes are required to robustly assess the millions of genetic variants or thousands of genes measured. Disease prognosis and treatment success in IMD remain variable and unpredictable. I will mine these data across diseases, to generate orders of magnitude fewer summary features describing shared IMD factors in a more parsimonious manner. I will use molecular measurements and clinical data to understand the gene products the features represent, and their relevant contexts (cell type, disease state/subtype). Through this targeted feature engineering, I aim to describe the relationships between different IMD, both common and rare, and identify new therapeutic treatment targets or diseases/disease-subtypes where existing treatments may be newly applicable.