Chipseq peak calling and peak frequency region

Question

[mavy]

I am working on Chip seq data, I have implemented the chipseq pipeline with spike in method and called peaks using MACS2.

The same data has been analyzed before in my lab, and my peaks show a similar pattern as the previous ones . By this I meant that the samples showing more peaks before are showing more in mine too. (which I consider that my work till peak calling is correct)

Now I have to do the downstream analysis, for which I am using CHipseeker R package.

I am plotting the number of peaks around the TSS, it shows the maximum peak frequency at the TSS , whereas the previous results indicate that they are minimum at the TSS and also proven in the text.

I am completely clueless about what I am doing wrong or what's causing this difference.

I am using the .broadpeak file from its output as the bed files input in the Chipseeker package. Am I using the wrong bed file ??

txdb <- TxDb.Hsapiens.UCSC.hg38.knownGene

promoter <- getPromoters(TxDb=txdb, upstream=2000, downstream=2000)

tagMatrixList <- lapply(files, getTagMatrix, windows=promoter)

plotAvgProf(tagMatrixList, xlim=c(-3000, 3000), conf=0.95,resample=500, facet="row")

Will appreciate any help regarding this

[Arup Ghosh]

If you are working with Histone modification data the broadpeak file is the right choice for TF binding events choose narrowPeak.