DNA methylation is widely used to model physiological phenotypes, such as aging and type II diabetes. The Epigenetic Pacemaker, EPM, is an implementation of a fast conditional expectation maximization algorithm that models epigenetic states under and evolutionary framework. The EPM was first introduced by Snir et al. as an extension of the Universal Pacemaker (UPM) model of genome evolution. In contrast to regression bases approaches, the EPM does not assume a linear relationship between the epigenetic state and a trait of interest. As a result the EPM can model non-linear epigenetic trait associations directly without transformation of the phenotype of interest.
Reference: https://epigeneticpacemaker.readthedocs.io/en/latest/
Home Page has the description for general EPM model. More info could be found at: https://epigeneticpacemaker.readthedocs.io/en/latest/
Before starting, users could find the desired format for input files, shown in the Sample File Format section. The first line should be the sample IDs. The second line should be the age of the sites, and the following rows being different sites and their values. Notes that there should not be empty lines at the end of the file. Users can also use the sample data in Sample File Download section to get familiar with this web interface.
After getting used to this web interface, users can start uploading their own files in the Test Your Dataset section under the Start tab. A file should be uploaded and PCC value should be defined for a successful plot. When users upload their files and enter a PCC value and click the Plot button, the main panel will show the EPM output for their values. The Summary section contains how many sites have been selected based on the PCC value, and the number of individuals after selection. Users can find the information for the selected sites under Table, and download the results of Epigenetic Pacemaker when they click the Download button.