Health Data Science Lab Overview
Many human diseases are influenced by a complex combination of genetic, environmental, and lifestyle factors. The Health Data Science team develops approaches to better understand and prevent these complex diseases by collecting, analyzing, and translating multi-omic data into actionable clinical insights for physicians and patients. Our group capitalizes on and integrates our joint expertise in multiple fields, including clinical medicine, translational science, data and computer science, genetics (medical, pharmacogenomics, and wellness), statistics, and computational biology.
Actionable Multi-omic Disease Models: We are building a clinical knowledge base for multi-omic disease risk evaluation and disease prevention with an emphasis on patient/physician education and actionability. This knowledge base incorporates the best scientific evidence and clinical guidelines for genetic risk factors as well as phenotypic measurements derived from blood, saliva, and stool.
Multi-omic Clinical Trials: We are collaborating with academic and medical partners to collect unique deeply phenotyped multi-omic datasets in a wide range of wellness and disease cohorts. We have thousands of longitudinal multi-omic data points on healthy individuals who participated either in a Scientific Wellness clinical trial or commercial wellness program. In addition, some of the disease cohorts include breast cancer survivors, early stage Alzheimer’s disease, colorectal cancer, and type 2 diabetes. We are passionate about building these datasets to better characterize wellness-to-disease transitions and develop novel applications in P4 and systems medicine.
Phenome-wide Characterization of Genetic Predispositions: We are developing polygenic scores for a wide range of chronic diseases, including type 2 diabetes, inflammatory bowel disease, and cardiovascular disease. These scores are being used to stratify ‘healthy’ populations by disease predisposition and to study the manifestation of genetic risk for insights into disease prevention and treatment.
Multimodal Analysis of Human Health and Disease: Complex diseases and traits are driven by and influence a multitude of biological, environmental, and lifestyle factors. By applying computational and statistical approaches to longitudinal multi-omic data, we systematically study how these factors interact to affect human health. We seek to identify predictive and causal relationships that can help lead to better disease prevention, diagnosis, and treatment.