Data for 'Physical fitness is negatively associated with DNA methylation-based risk of aging-related diseases'

This is the code and data for the article titled 'Physical fitness is negatively associated with DNA methylation-based risk of aging-related diseases'. To run this code, you have two options: the first option is to run all the code steps using the notebook 'full_code.ipynb'. the second option is through multiple python code snippets. To run this code using the multiple python code snippets, you can follow this order: 1) calculate_predictor_episcore_correlations.py 2) map_to_diseases.py 3) patient_level_disease_risk_estimation.py 4) top_10_patients.py This code expects the following input files: 1) Generated episcores for the patients (.xlsx file) 2) Episcore_mapping (.csv file) 3) The 5 fitness predictors (VO₂max, GripStrength, JumpMax, BMI, and cognition) and confounders which are Age and Sex (coded where 1 = male and 0 = female) 4) Episcore-disease hazard ratio values from Gadd et al reference paper supplementary file, sheet 1J And it outputs the following data files: 1) Significant correlations found of fitting OLS model to the predictor-episcore dataset after adjusting for Sex and Age, Association is considered significant if p-value < 0.05 ('significant_results.xlsx'). 2) The results of mapping previous table to the diseases ('Predictors_diseases.xlsx'), note that this table is used to produce the tripartite figures. 3) The results of Patient-Level Disease Risk Estimation for patients across 10 diseases, and categorizing fitness predictors as low, normal and high ('df_high_risk.xlsx'). 4) The top 10 patients with the highest number of flagged diseases ('top_10_high_risk_patients_all_data.xlsx').