Blood-based multivariate methylation risk score
for cognitive impairment and dementia

Background · Methods · Shiny · Software · Contact

This repository contains the scripts used for the project "Blood-based multivariate methylation risk score for cognitive impairment and dementia".

Background

As DNA methylation may act as the molecular link between lifestyle/environment and the biological processes governing health and disease, blood-derived DNA methylation data might be utilized for the early identification of persons at risk of developing dementia. Therefore, our research aim is to establish a robust model for predicting a person's midlife dementia risk.

Methods

The applied methodology consists of four main steps:

1. Model generation using the DNA methylation data of the EXTEND and EMIF-AD cohorts.
2. Model validation in the EMIF-AD, PPMI, and ADNI cohorts.
3. Model interpretation (here).
4. Model extension with genetic and cerebral spinal fluid biomarkers in the EMIF-AD cohort.

Shiny

You can also generate the multivariate risk scores for cognitive impairment and dementia yourself in an R Shiny application. You can do this by downloading the current GitHub repository and running the following command:

runApp('PATH/Shiny') ("PATH" refers to the path to the GitHub directory)

Please note that it takes a long time to launch the app for the first time because large models (> 10 GB) are being downloaded from Zenodo.

Software

• R version 4.1.2
• RStudio version 2021.09.02+382

Contact

Feel free to contact me via email: [email protected]