Computational protocols to analyse DamiD-deq data
The R markdown shows an example of how differential methylation may be called from counts files and then translated into peaks.
The exact analysis does not have to follow the same methods, but other tools could be used to obtain similar results (for example, DESeq2, Limma...). Make sure to observe the same file strucutre for the peak-calling scripts (which has no error handling or input checks, sorry).
The peak calling script aggregates neighbouring sites of methylation and opposing direction into pairs and keeps aggregating as long as they are signficantly differentially methylated. It allows for a single unmethylated tag as long as it has the same directionality.
Finally, the penetration of methylated tags at the core of a peak (1kb around its center) is calculated to allow filtering on this metric down the track.