Annotate each peak to gene starts within 10 kb of the peak with python ?
1
0
Entering edit mode
8.5 years ago
Kevin_Smith ▴ 10

How can I do this with peaks in a bed file and genes in a txt file ? Also I will like to see the nearest gene to each peak.

Thank you very much

ChIP-Seq sequence gene sequencing alignment • 3.6k views
ADD COMMENT
0
Entering edit mode
8.5 years ago

Via command-line toolkit, to map a set of names of overlapping genes to each peak within 10kb:

$ bedmap --echo --echo-map-id-uniq --range 10000 peaks.bed genes.bed > annotation-answer.bed

To get the nearest gene to each peak:

$ closest-features --closest peaks.bed genes.bed > nearestgene-answer.bed

To use with Python, call with subprocess or similar library.

ADD COMMENT
0
Entering edit mode

I'm using pybedtools, do you know how I can use your code with this module?

Thanks!

ADD REPLY
0
Entering edit mode

To use with Python, call with subprocess or similar library.

ADD REPLY
0
Entering edit mode

I'm new with python , I'm using pybedtool. I don't know how to do it. Do you can give me more specific instructions?

ADD REPLY
0
Entering edit mode

For example:

import subprocess
subprocess.call('bedmap --echo --echo-map-id-uniq --range 10000 peaks.bed genes.bed > annotation-answer.bed', shell=True)

More details are available via the documentation. Search engines will also point you to a good review of the subprocess library, among others.

ADD REPLY
0
Entering edit mode

Using pybedtools you can do something like the following (I simply adapted the code found in pybedtools documentation):

from pybedtools import BedTool

peaks = BedTool('peaks.bed.gz')
genes = BedTool('hg19.gff')

nearby = genes.closest(peaks, d=True, stream=True)

for gene in nearby:
    if int(gene[-1]) < 10000:
        print gene.name
ADD REPLY

Login before adding your answer.

Traffic: 2350 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6