Question

Quantifying peaks corresponding to a purticular feature

0

Entering edit mode

snigdhapola • 0

@snigdhapola-7746

Last seen 9.6 years ago

CRRao AIMSCS, India

Hi,

how to quantify the peaks annotated to a particular genomic region in a particular window size.

Thanks,

Snigdha.

chipseeker • 1.4k views

ADD COMMENT • link 9.6 years ago snigdhapola • 0

score 2 · Answer 1 · 2015-06-10

2

Entering edit mode

Guangchuang Yu ★ 1.2k

@guangchuang-yu-5419

Last seen 9 weeks ago

China/Guangzhou/Southern Medical Univer…

A possible solution is define the genomic region as a GRanges object and use the findOverlaps from GenomicFeatures package to get the peaks that overlap with the genomic region. Then you can count the number of peaks.

Any idea?

ADD COMMENT • link 9.6 years ago Guangchuang Yu ★ 1.2k

score 0 · Answer 2 · 2015-06-11

0

Entering edit mode

snigdhapola • 0

@snigdhapola-7746

Last seen 9.6 years ago

CRRao AIMSCS, India

It is after we get a count of no of peaks corresponding to a TF or histone mark, say around 3-4kb of TSS region of a particular gene, do we go by taking an average of -log10(pvalue) of the peaks we obtained in the region or any other way ?

ADD COMMENT • link 9.6 years ago snigdhapola • 0

0

Entering edit mode

how do you calculate the pvalue?

ADD REPLY • link 9.6 years ago Guangchuang Yu ★ 1.2k

0

Entering edit mode

No. If we get more than one peak in a given window size, around a gene, for a particular histone modification or transcription factor, any idea if we have to consider peak with max p-value or an average of the p-values of all the peaks in that window size ?

ADD REPLY • link 9.6 years ago snigdhapola • 0