Hypothesis
Our hypotheses are that combat-related traumatic injury is associated with premature biological ageing, and that age gaps can be used to predict a 3-year change in clinical outcomes.
Summary
Alongside Prof. Janet Lord’s team (Birmingham) we aim to publish a paper on premature ageing in the ADVANCE cohort. We want to use published proteomic and DNA methylation ageing clocks and for this, we will need access to the SomaScan data (already accessed) and the DNA methylation data from Prof. Lord’s team.
Aims:
- Score the ADVANCE for their biological age using clinical clocks (PhenoAge), proteomic clocks (from https://doi.org/10.1038/s41586-023-06802-1) and DNA methylation clocks.
- Investigate associations between the difference in predicted and chronological age (age gap) and cardiometabolic risk factors.
- Assess the predictive ability of baseline age gaps to predict a 3-year change in clinical outcomes.
We will use: univariate statistics (including paired analysis for FU1 data) and multivariate statistics (including ML algorithms like PLS, LASSO, XGBoost etc). We will also use externally trained statistical models of ageing to score ADVANCE participants.
Keywords
Somascan, DNA Methylation, Premature Aging, Machine Learning