Genes, Environment and Gene-Environment Interactions in Chronic Diseases and comorbidities: From Association to Causation
Chronic diseases such as cancer, diabetes, and cardiovascular diseases have imposed a huge social and economic burden. There are a variety of genetic and environmental (e.g., air pollution, behavior habits) risk factors associated with chronic diseases. Previous studies can only tell us that genetic and environmental factors are associated with these diseases, but they cannot conclude that the diseases are caused because of genes and environmental factors. In addition, previous studies have focused only on genetic or environmental factors, but exposure to both may have different results (e.g., greater susceptibility to certain diseases).
Deep learning techniques have shown excellent performance in the early diagnosis and prediction of chronic diseases using multidimensional data. We hypothesized that the state-to-art prediction models can accurately identify high-risk populations and provide novel tools for the early prevention and control of chronic diseases and comorbidities. In addition, the causal models can answer whether and how simultaneous or separate exposure to genetic and environmental factors can lead to the onset of disease.
We plan to perform our study within a duration of three years. Our study will provide novel models for the early risk prediction of chronic diseases. Moreover, causal methods will identify modifiable risk factors and biological pathways for the intervention of chronic diseases. The proposed new methods have important implications for the comprehensive risk management of chronic diseases and fall into the field of public health promotion.