1. Research Questions
Q1. What are the comorbid biological mechanisms of myopia, obesity, poor posture, and mental health disorders in children and adolescents, and what are the key biomarkers?
Q2. How can machine-learning models integrate multi-omics data to accurately predict disease risk in children and adolescents?
2. Objectives
We will build an integrated “multi-omics + long-term health-record” data resource and develop and then validate a multi-morbidity risk-prediction system for children and adolescents.
By pinpointing the key genes, proteins, and metabolic pathways that drive the co-occurrence of the four target conditions, we will map a cross-disease molecular network that supplies the evidence base for “one prevention for multiple diseases.”
Within a formal causal-inference framework, we will estimate the true effects of genetic variants on disease onset and progression and quantify how genes, environments, and molecules interact with one another.
3. Scientific Rationale
Multi-omics profiles capture the full continuum from health to disease with high sensitivity and specificity, while machine-learning excels at distilling hidden patterns from these high-dimensional data. Together, they greatly accelerate the discovery of robust biomarkers and the construction of accurate risk-prediction models.
