Difference betweent peaks sets

[NoemieL]

Hi,
I am interested to find the list of peaks that are related to the call peak summary plots generated by ArchR. I would like to know the difference between the sets of peaks I obtained with the following different commands because non of them are given the same results:
Peak-Call-Summary2.pdf

1. markerList2_GRangesList <- getMarkers(markersPeaks, cutOff = "FDR <= 1", returnGR = TRUE) #to get the full list of peaks with markersPeaks <- getMarkerFeatures(ArchRProj = ArchR, useMatrix = "PeakMatrix", groupBy = "group",bias = c("TSSEnrichment", "log10(nFrags)"),testMethod = "wilcoxon")
   I get total of peaks divided in group X (98081 peaks) and Y (87268 peaks). And non common peaks.

2. PeakSet_X=readRDS(gzfile("Save-ArchR/PeakCalls/X-reproduciblePeaks.gr.rds"), refhook = NULL)
   give 147654 peaks for group X
   PeakSet_Y=readRDS(gzfile("Save-ArchR/PeakCalls/Y-reproduciblePeaks.gr.rds"), refhook = NULL)
   give 147869 peaks for group Y
   and 117285 common peaks.

3. peakSet=getPeakSet(ArchR_peak) give 185349 peaks with:
   peakSet[-grep("Y",peakSet$GroupReplicate),] giving 98081 peaks for group X
   peakSet[grep("Y",peakSet$GroupReplicate),] giving 87268 peaks for group Y and non common peaks.

1 and 3 are concordant but I don't get the common peaks that were suggested in the call peak summary plot, why? were could I find them? For 2 I have the common peaks but when combining both data I have less peaks (147654+147869-117285=178238) than the total in 1 or 3, why? I am mostly interested to get file 1 with all the peaks identify to be able to apply different filters on FDR and Log2FC.

Thank you very in advances for you clarification