Hello community,

I have been working for a few months with the data analysis of my sequenced reads in R. Each sample is a different count matrix. Each matrix contains 3 columns (Chromosome, Genomic Position, Read Counts). The rows are more than 400.000 and each row represents a methylated CpG site.

The comparison of normalised read counts per CpG position between samples, will give me the amount of methylation upon this site. Not as a percentage but as a correlation between the different samples.

This comparison should be a linear model (I guess) like the differentially expressed genes in RNA-seq so I have been reading and scratching the surface of R package "limma" and the function "voom" and "limma-trend" as well as the package "edgeR". However, I have no experience in bioinformatics and I can not understand how to use them properly.

Can anyone guide me in order to perform such a comparison between my samples?