Hi Jianhong, 

I recently try using your ChIPpeakAnno and came across some questions that I hope can get your help.

I used findOverlapsOfPeaks from ChIPpeakAnno to find overlaps from 3 peak data and plot the results:

ol1 <- findOverlapsOfPeaks(a,b,c)

ol2<- findOverlapsOfPeaks(a,b,d)

ol3 <- findOverlapsOfPeaks(a,b,e)

makeVennDiagram(ol1,NameOfPeaks=c('A','B','C'))

makeVennDiagram(ol2,NameOfPeaks=c('A','B','D'))

makeVennDiagram(ol3,NameOfPeaks=c('A','B','E'))

So I would expect overlaps between A and B remain the same in the 3 conditions. However as you can see from attachment there are slight differences between them. So I'm wondering how this would happen.

Thanks for your time. Looking forward to your reply.

Best

Sujun

(Posting the reply from author)

Hi Sujun,

Thank you for selecting ChIPpeakAnno in your research. If you don't mind, please try to get help from https://support.bioconductor.org, thus others could also benefit from your questions.

You may want to have a try the following code to see what happened in your data:

ol1 <- findOverlapsOfPeaks(a, b, c, connectedPeaks="keepAll")

ol2 <- findOverlapsOfPeaks(a, b, d, connectedPeaks="keepAll")

ol3 <- findOverlapsOfPeaks(a, b, e, connectedPeaks="keepAll")

makeVennDiagram(ol1, connectedPeaks="keepAll")

makeVennDiagram(ol2, connectedPeaks="keepAll")

makeVennDiagram(ol3, connectedPeaks="keepAll")

This link may give you some details: C: Curious behaviour of ChIPpeakAnno

Let me know if you have any questions.

Yours Sincerely,

Jianhong Ou