Aim: Develop a comprehensive aging clock model using multi-omics data from the UK Biobank, including clinical, proteomic, and metabolomic datasets. This model based on machine learning will predict physiological age and organ senescence, and assess disease risk from a single blood test.
Research Questions:1. How can multi-omics data be integrated to accurately forecast an individual’s biological age and organ aging? 2. What is the model’s accuracy and generalizability across different databases and independent clinical sample repositories? 3. How can the model be optimized and simplified for practical medical application through a minimally invasive blood test?
Objectives:1. Construct an aging clock model based on integrated multi-omics datasets. 2. Validate the model’s predictive accuracy in external datasets and clinical samples. 3. Refine the model for practical use in clinical settings with the finger-blood test technology we developed.
Scientific Rationale:Emerging evidence suggests that multi-omics data provide a holistic view of the aging process, capturing changes in gene expression, protein function, and metabolic pathways. By analyzing these datasets, we aim to identify biomarkers of aging that can be used to assess an individual’s physiological age and organ health. And we are already advancing low-cost technology for detecting target biochemical substances in fingerstick blood tests, which will support the technical feasibility of our model.. Our ultimate goal is to create a user-friendly, clinically applicable model that requires only a drop of blood, offering insights into organ senescence and disease risk prediction.
