PeakSeq

[seq_GA]

Anyone has used Peakseq tool for chip-seq experiments?
How to run the parallel version of compile.py? I get to see the following error when I try to run as below inside the directory where all *.fa files are saved for hg18 build without PBS mode.

python ../ParallelCompile.py 36
Traceback (most recent call last):
File "../ParallelCompile.py", line 2, in ?
import sys, glob, re, subprocess, os, nws.sleigh, time
ImportError: No module named nws.sleigh

python version is 2.4.3

[dawe]

Anyone has used Peakseq tool for chip-seq experiments?
How to run the parallel version of compile.py? I get to see the following error when I try to run as below inside the directory where all *.fa files are saved for hg18 build without PBS mode.



python version is 2.4.3

Never used Peakseq but it seems you are missing a standard python distribution. Also, check the minimum python version as subprocess module should be part of python 2.5+

d

[seq_GA]

Hi D, Thx for your response.


Is there any HOWTO document for this tool? I have downloaded Mappability Map code and chip-seq scoring (perl code) from http://www.gersteinlab.org/proj/PeakSeq/.


Regards

[maubp]

Also, check the minimum python version as subprocess module should be part of python 2.5+

The subprocess module was included in Python 2.4 onwards, but I agree, check the version of Python they expect you to have.

The nws.sleigh import error means you are lacking a 3rd party Python library, in this case NetWorkSpaces for Python:

http://nws-py.sourceforge.net/

Double check you have installed all the documented dependencies...

[Bioinfo]

Hi all,
Has anyone used Peakseq for chipseq data analysis. Can anyone suggest which eland alignment file (export/sorted) Peakseq accepts.
Does anyone knows how to convert eland alignment files (export/sorted) to s*_eland_extended.txt or s*_eland_results.txt format.
Any help would be appreciated.

[ldong]

Hi, Bioinfo,
I use bowtie for alignment. You can use samtools to convert different formats of alignments. Here is what I did. Best, ldong

https://sites.google.com/a/brown.edu/bioinformatics-in-biomed/peakseq-for-chip-seq1

[dnusol]

Hi ldong,

I am also having problems running PeakSeq. I checked your website but still have some questions.
I downloaded the mappability.txt file for mouse. I also tried to create the mappability file but the "compile.py" script run in the directory containing the .fa files just throws the following error:

usage compile.py <merlen>

what is this merlen parameter about?

in the website you mention, you use a config.txt file in step_3, did you create it yourself??
Also, is the create_signal_map_new1.pl script included in the PeakSeq installation? or is it part of the modifications you comment at the beginning of the webpage?

Thanks for your help

[ldong]

Hi, dnusol,
merlen is the length of your reads. compile.py uses the number to get a piece of sequence of that length and map back to genome to see if there are same sequence elsewhere. So if your read length is 36bp, you should call the command:
compile.py 36

Yes, I made up the config.txt file to make it easier to run the software. All perl scipts with new.pl were modified based on the original ones. Hope his helps. Ldong

[dnusol]

Thanks Ldong for your quick answer. Just two more questions (I hope!)
Do you know what the SGR files needed by score_hits_PolII.pl are?
and is it possible to use Bowtie-aligned reads without modifying the script?

[ldong]

Ok. The sgr files created by creat_signial.pl will be used by score_hits.pl to look for potential peaks. The original creat_signal.pl read eland output, you need modify it to read bowtie output. It is very easy, just change:
while (<IN>) {
chomp;

my ($t1, $seq, $map, $t3, $t4, $t5, $chrt, $pos, $str, @rest) = split /\t/, $_;
my $read_length = length $seq;

if ($str eq "F") {
$data{$pos} += 1;
$data{$pos+$L} += -1;
}
elsif ($str eq "R") {
my $start = $pos + $read_length - $L;
$start = 1 if ($start < 1);
my $stop = $pos + $read_length;
$data{$start} += 1;
$data{$stop} += -1;
}
else {
print "PROBLEM\n";
}
}

close IN;

to:

while (<IN>) {
chomp;
my ($t1, $str, $t3, $pos, $seq, @rest) = split /\t/, $_;
my $read_length = length $seq;

if ($str eq "+") {
$data{$pos} += 1;
$data{$pos+$L} += -1;
}
elsif ($str eq "-") {
my $start = $pos + $read_length - $L;
$start = 1 if ($start < 1);
my $stop = $pos + $read_length;
$data{$start} += 1;
$data{$stop} += -1;
}
else {
print "PROBLEM\n";
}
}

close IN;
Let me if it is not clear.

[AL_B]

PeakSeq - need some help

Hi,
I am not from the area of bio-informatics, yet I need to learn how to run PeakSeq.
so I have some basic questions:
I went to this website

http://info.gersteinlab.org/PeakSeq

and I am trying to use PeakSeq evrsion 1.1
I don't understand what is the relation between the .fa files and the mappability map text file.
How do I get the .fa files and what do I do with it? according to the README file, the program should
get as an argument a mappability map text file, and there is no mentioning of .fa files.
So what should I do?

Thanks (I hope)

[ldong]

Hi, AL_B,
My understand is that the script, compile.py reads all the and .fa files, and extract many short sequences with length of 'merlen", then map the short sequences back to the fa files, so that the software know if a certain sequence shows up in other places. The results are saved in file mappablility.txt. So you need generator new mappablity file for different read length. Hope this helps.

[AL_B]

PeakSeq - need some help

Thank you so much for your fast reply.
I have another question.
Do you know some message "for reading.n chr_id_list.txt" message?
This is some output that I am getting when I am running the second part of PeakSeq
./PeakSeq -peak_select config.dat

[priya]

Hi, AL_B,
My understand is that the script, compile.py reads all the and .fa files, and extract many short sequences with length of 'merlen", then map the short sequences back to the fa files, so that the software know if a certain sequence shows up in other places. The results are saved in file mappablility.txt. So you need generator new mappablity file for different read length. Hope this helps.

Hi,
I have short sequence data of various read lengths for eg: 43bp,51bp and 52 bp. Do I need to generate the new mappibilty files for individual read lengths or can I use 50 bp as read length and use it for everything .