Entering edit mode
4.6 years ago
Morris_Chair
▴
350
Hello everyone, I am new to ChIP-seq and I want to reproduce an analysis from a paper . In SRA they loaded only one file, the one used for the ChIP experiment without any control or MOCK files, then I was told that it is possible to analyze only one file so I want to ask you if it's ok to use this command line without input as first step:
macs2 callpeak -t ChIP.bam
-f BAM -g 1.3e+8
-n Nanog-rep1
--outdir macs2
Thank you
Yes, this command is possible. If you want to know how the tool estimates the background levels check the original MACS paper (keyword is local lambda). Still, technically possible does not mean that this is a good experiment. Depending on antibody quality you might get lot of unspecific binding that a proper IgG control would've eliminated but if this is what they uploaded you can go ahead with it.
Hi ATpoint,
I run it and I got different files including an excel file for the peaks location. Next step will be to visualize/normalize this peaks, can you suggest tool for that? preferably that run with Studio
thanks
If you have only one sample, what do you want to normalize against? What is
Studio
,RStudio
?I read somewhere that when there is only one file MACS is able to apply a normalization in order to distinguish noise and peaks so it's not for this case. I will need to normalize against the input others files.
yes it was a typo, I meant Rstudio
tnx
But you said you have no controls? If you do, include them into the
macs
command with-c
. Beyond that, I do not see what one could do with only a single sample. For visualization, checkbamCoverage
fromdeeptools
. Also check online for ChIP-seq tutorials, as this is all part of standard workflows.I am doing things in parallel sorry for the confusion:D ok, I have only one file and I can still visualize significant peaks which is already good,
I thought deeptools could run only with python,
tnx
Yes. Can be run from the command line like most tools.