luisamariariverarivera/Rwanda_Aging_Public

Rwanda Epigenetic Aging # Rwanda_Aging_Public **Description:** This repository contains code to reproduce Uwizeye et al (2024) Prenatal exposure to genocide accelerates epigenetic aging as measured in second-generation clocks among young adults. Preprint available: medrxiv.org **Data Availability:** Raw data are not available due to restrictions in our consent form. Data for the purposes of reproduction only is available by request to luisa.maria.rivera@gmail.com. Our DNA methylation preprocessing pipeline and cell-type deconvolution is not presented here but is available upon request. Epigenetic age estimates were generated using the Biolearn tool from the Biomarkers of Aging consortium available at: bio-learn.github.io. **Files:** The "Processing.R" file is the basic data processing file. - Joins our various phenotypic and age estimate files, and contains the cell-type PCA used to control from cell-type heterogeneity - Calcuates aging residuals where appropriate (e.g. not for GrimAgeAccel and DunedinPACE, which calculate residuals automatically) The "Analysis.R" file contains the analysis in the main text. - We begin by running a series of regression models investigating the effect of exposure group on aging estimates with and without adjustment for postnatal ACES. - We extract effect sizes and plot these to demonstrate the increased sensitivity of second generation clocks compared with first generation clocks to the prenatal environment. The "revision_analyses.R" script contains analysis requested by reviewers after peer review. - We compare exposure group BMI and SES - We use the epismoker package to estimate smoking prevalence in participants from DNA methylation data. - We conduct sensititivity analyses with interaction by sex and with additional immune cell type principle components.