SEQanswers

Go Back   SEQanswers > Bioinformatics > Bioinformatics



Similar Threads
Thread Thread Starter Forum Replies Last Post
Twin peaks, double peaks, double humps of RNA Seq, itís all very frustrating...HELP! AndyG Sample Prep / Library Generation 18 10-24-2015 03:43 PM
Count number of reads in peaks? albireo Bioinformatics 4 04-13-2013 06:51 PM
Comparing input peaks to IP peaks biznatch Epigenetics 2 09-24-2011 11:38 AM
ChIP-Seq: ZINBA integrates local covariates with DNA-seq data to identify broad and n Newsbot! Literature Watch 0 07-27-2011 03:30 AM

Reply
 
Thread Tools
Old 11-26-2013, 07:32 AM   #1
spacup
Member
 
Location: Montpellier

Join Date: Apr 2013
Posts: 17
Default Number of peaks with Zinba

Hi everybody,

I ran a ZINBA analysis for FAIRE-seq study and found around 4,000 peaks... which seems to be very few as we should find around 100,000 according to this paper : http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3202292/

So I was wondering if somebody could has an idea why I found so few peaks?
(See below the options)
Thanks for your help!

run.zinba(
align='.../athresh1_ext134',
numProc=8,
seq=.../file.bed',
basecountfile='.../file.basecount',
filetype='bed',
outfile='.../output',
twoBit='.../GRCh37.2bit',
extension=134,
printFullOut=1,
refinepeaks=1,
input='none',
threshold=0.05,
FDR=T,
interaction=T,
selectchr='22',
selectcovs=c("gcPerc", "align_perc", "exp_cnvwin_log"),
selectmodel=T
)
spacup is offline   Reply With Quote
Old 11-26-2013, 06:12 PM   #2
aaronh
Member
 
Location: California

Join Date: Sep 2008
Posts: 45
Default

it look like you only ran it on chr22
aaronh is offline   Reply With Quote
Old 11-26-2013, 11:04 PM   #3
spacup
Member
 
Location: Montpellier

Join Date: Apr 2013
Posts: 17
Default

No, I checked the peaks in IGV and there are in all chromosomes...

The option selectchr='22' is just use for model selection

Last edited by spacup; 11-26-2013 at 11:06 PM.
spacup is offline   Reply With Quote
Old 12-04-2013, 01:36 AM   #4
spacup
Member
 
Location: Montpellier

Join Date: Apr 2013
Posts: 17
Default

Nobody has an idea?
Because I downloaded ENCODE FAIRE-seq data and I used ZINBA for peak calling and althought ENCODE find around 200,000 peaks (with Fseq) I only found around 60,000...
I know they used p-value whereas I used FDR in ZINBA which is more stringent and that could be an explanation, but they fit their data to gamma distribution and I don't really understand the meaning of this...

Can anyone enlighten me?
Has anyone tried ZINBA on ENCODE data?
thanks
spacup is offline   Reply With Quote
Reply

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 07:01 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO