Research Question
What are the interactive patterns between multi-omic factors and environmental factors in influencing the risk, progression, and pathological mechanisms of chronic diseases? How can these findings be used to improve the performance of disease prediction models and the precision of targeted prevention measures?
Research Objectives
1.To integrate multi-omic data (genomic, transcriptomic, epigenomic, proteomic, metabolomic) with clinical phenotypes of chronic diseases for identifying causal risk factors of certain chronic diseases.
2.To incorporate multi-omic data into machine learning algorithms (e.g., GLMNET, XGBoost, multi-modal deep learning) for constructing prediction models of chronic disease.
3.To integrate genetic data with environmental exposures for investigating gene-environment interactions related to disease risk, progression, and prognosis.
4.To use multi-type imaging data to assess disease-related structural and functional changes and compare trajectory differences between chronic disease subgroups and healthy control populations.
5.To use gut microbiota sequencing data for characterizing the complex cross-talks between gut microbiota and lifestyle factors related to disease risk, progression, and prognosis.
Scientific Rationale
This project leverages the multi-dimensional resources of the UK Biobank (including multi-omic data, imaging data, environmental exposure data, electronic health records, and long-term follow-up data) to explore the interaction mechanisms between multi-omic factors and environmental factors in chronic diseases. It aims to deepen the understanding of disease pathogenesis, and enhance the accuracy of disease prediction and the effectiveness of prevention measures.
