SEQanswers

Go Back   SEQanswers > Bioinformatics > Bioinformatics



Similar Threads
Thread Thread Starter Forum Replies Last Post
Merging MACS Peak analysis +/- duplicates? Combo Epigenetics 8 12-21-2011 08:07 AM
ChIP-seq peak calling from replicates ttnguyen Bioinformatics 4 08-10-2011 02:21 AM
Peak finding with MACS: Questions on # of reads and "strand bias" jjw14 Bioinformatics 0 07-29-2010 07:17 AM
Effective genome sizes for peak calling idonaldson Bioinformatics 1 02-26-2010 06:41 AM
User friendly Peak Calling Giles Bioinformatics 0 02-02-2010 06:58 PM

Reply
 
Thread Tools
Old 08-01-2011, 10:27 AM   #1
ywlim
Junior Member
 
Location: California

Join Date: Jul 2011
Posts: 6
Question p-value for peak-calling using MACS

Hi I am trying to do peak-calling using MACS for some of my H3K4me3 ChIP-seq data. Using the default p-value of 1e-5 I only obtained 500+ peaks. But if I change the p-value to 1e-4 I got 1500+ peaks and by increasing it more to 1e-3 I got 7000+ peaks! My question is how do I know which is the best p-value to use? I would like to get more peaks but I also don't want any false positive peaks. Thanks for your help!
ywlim is offline   Reply With Quote
Old 08-16-2011, 07:10 AM   #2
emilyjia2000
Member
 
Location: usa

Join Date: May 2011
Posts: 59
Default

I am also very interested in knowing this because I am using MACS to do the peak calling too
emilyjia2000 is offline   Reply With Quote
Old 08-17-2011, 06:57 AM   #3
mehc
Member
 
Location: Montreal

Join Date: Aug 2011
Posts: 10
Default

Depends how many false positives you're willing to handle. The default is 1E-5 and if your enrichment is good enough you should have many thousand peaks at that level.
mehc is offline   Reply With Quote
Old 08-17-2011, 07:12 AM   #4
ETHANol
Senior Member
 
Location: Western Australia

Join Date: Feb 2010
Posts: 308
Default

There is no best number and it completely depends on your data set. Load the data onto a genome visualizer and look at the peaks it is calling. Use your best judgement and validate by qPCR. Fold enrichment is another important and easy to understand parameter to consider.
__________________
--------------
Ethan
ETHANol is offline   Reply With Quote
Old 03-28-2013, 10:17 AM   #5
dgrayson
Junior Member
 
Location: Chicago

Join Date: Oct 2012
Posts: 6
Default

How do you mean? Are you talking about the BAM file from Bowtie or the BED file from MACS? I find that the output from MACS is just a bar. On the other hand the BAM files are quite big and time out on me when I try to visualize them.
dgrayson is offline   Reply With Quote
Old 03-28-2013, 10:29 AM   #6
ywlim
Junior Member
 
Location: California

Join Date: Jul 2011
Posts: 6
Default

dgrayson, if you use the -w option when running MACS you will generate a wig file which can be uploaded and visualized on a genome browser. I like to use the -w option with the -S option so that I only generate a single wig file instead of one for each chromosome.
ywlim is offline   Reply With Quote
Old 03-28-2013, 10:32 AM   #7
dgrayson
Junior Member
 
Location: Chicago

Join Date: Oct 2012
Posts: 6
Default

Thanks- I've tried that. I'm running MACS in a local Galaxy server which does not produce wig output files (MACS 1.4 in Galaxy). In addition this version of Galaxy does not allow us to view our output in UCSC browser.
dgrayson is offline   Reply With Quote
Old 03-28-2013, 10:39 AM   #8
ywlim
Junior Member
 
Location: California

Join Date: Jul 2011
Posts: 6
Default

Maybe you can try using the web-based Galaxy? It allows you to generate a wig file, which is, in my opinion, really important for visualization of your chip datasets.
ywlim is offline   Reply With Quote
Old 08-08-2013, 03:27 AM   #9
chipmonk
Member
 
Location: pune,india

Join Date: Jul 2010
Posts: 18
Default igv bam bai bed files

you can use IGV viewer. view bam ( samtools index to get bai file) of experiment control and the bed file to see an overall view.. bed will be bars of the peaks but bam with index will show tags and enrichment.
chipmonk is offline   Reply With Quote
Reply

Tags
macs peaks chip-seq

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 04:32 AM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2022, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO