This UK Biobank study will compare how multi-scale exposures (genomic, clinical, environmental) differentially impact cancer vs non-cancer outcomes. Key questions: 1) What is the impact of single or shared exposures (e.g., obesity, smoking) on distinct outcomes; 2) Whether genetic or omics biomarkers can distinguish disease-specific pathways; 3) What mechanisms may determine outcome specificity. We hope to: 1) Compare exposure-outcome associations between different exposures and cancer (e.g., lung, colorectal) or non-cancer diseases (e.g., cardiovascular diseases, Alzheimer’s) using UK Biobank’s linked EHRs, questionnaires data, laboratory test data, etc.; 2) Identify special sites and differential proteins for cancer or non-cancer outcomes based on genomic and omic data; 3) Develop AI models for competing risk prediction. While cancer and non-cancer diseases share common risk factors (e.g., aging, inflammation), their etiological pathways diverge fundamentally. The abundant population, laboratory, genomics and omics data in UK Biobank provide sufficient feasibility for the above goals, informing targeted prevention strategies.
