Aims: Our research aims to use raw data from UK Biobank to systematically investigate the complex relationships of lifestyles, physical and biological measures, biomarkers and genetic susceptibility with metabolic diseases (e.g. hyperuricemia, diabetes, kidney disease) and cancer by using traditional statistical methods and machine learning algorithms.
Scientific rationale: The risk of metabolic diseases and cancer is determined by both environmental and genetic factors. However, there is still a paucity of data regarding the effects of gene-environment interaction on these diseases. Moreover, the mechanism of this risk remains unclear. In addition, traditional statistical methods often have inherent limitations in modeling the complex relationships between various risk factors and the clinical outcomes, whereas the advantages of machine learning techniques may help fill in this gap.
Project duration: This project will last for 60 months.
Public health impact: Findings from this research may provide significant healthcare benefits of adhering to a healthy lifestyle in individuals with metabolic diseases or cancer. Better understanding of the main risk factors of these diseases will enable physicians to identify high risk patients for early intensification and individualization of treatment to prevent these diseases.