Iam using ChiPpeakAnnotation to annotate my MeDip-seq people and i works great so far using the premade annotations from ensemble (TSS, Exon ...). However now I would like to annotate peaks with a specific set of genes. To make my costume annotation I picked all genes which iam interested in from the premade TSS ensemble annotation and made a bed file (Name, Chr, Start, Ende). The following script i used to annotated the peaks:
TSS <- annotatePeakInBatch(mydata.ranged, AnnotationData=annoTSS, output="both", multiple=T, maxgap= 100, PeakLocForDistance= "middle")
temp = as.data.frame(TSS)
pie(table(temp [as.character(temp$fromOverlappingOrNearest) == "Overlapping" | (as.character(temp$fromOverlappingOrNearest) == "NearestStart" & !temp$peak %in% temp[as.character(temp$fromOverlappingOrNearest) == "Overlapping",]$peak),]$insideFeature))
The problem is that for all my different samples the the pie-chart looks the same which made me a bit suspicious. I think something is wrong with my costume made annotation right now the rangedData file looks like this:
RangedData with 410 rows and 2 value columns across 20 spaces
space ranges | strand score
<factor> <IRanges> | <character> <numeric>
ENSMUSG00000048960 1 [10993465, 11303682] | + 1
ENSMUSG00000025932 1 [14168954, 14310200] | + 1
ENSMUSG00000037509 1 [34678188, 34813309] | + 1
ENSMUSG00000026109 1 [50900647, 51187270] | + 1
ENSMUSG00000025978 1 [55170159, 55226782] | + 1
Should be sufficient the annotation of my peaks right ?
Thanks for any suggestions.