[BioC] do we need remove PCR duplicate before we look for DE genes

[Jonathan Cairns]

Hi Lucia,

I'm not Simon, but my preferred strategy regarding duplicate removal for TF ChIP-seq is to remove duplicates when peak-calling, but not when doing read-based analysis (i.e. using a differential count algorithm on small, defined genomic bins) for similar reasons to the ones Simon described in his response; though I find my dynamic range is more like 1:1,000 than 1:100,000, and that's on a really good day.

See the "Effect of redundant reads on narrow peak calling" section of "Systematic evaluation of factors influencing ChIP-seq fidelity" for a brief discussion on this topic. http://www.nature.com/doifinder/10.1038/nmeth.1985

Jonathan
________________________________________
From: bioconductor-bounces at r-project.org [bioconductor-bounces at r-project.org] on behalf of Lucia [luciap at iscb.org]
Sent: 13 June 2013 19:20
To: Simon Anders
Cc: bioconductor at r-project.org
Subject: Re: [BioC] do we need remove PCR duplicate before we look for DE       genes

Hi Simon,
I wonder what your take is on ChIPseq experiments and duplicate removal
I tend to cover about 10% of the genome with both RNAseq and different types of ChIPseq, but I am not sure what the dynamic range is for the second and what the effect will be
Thanks
Lucia

Sent from my iPhone

On Jun 13, 2013, at 1:09 PM, Simon Anders <anders at embl.de> wrote:

> On 13/06/13 17:06, Wang Peter wrote:
>> in RNA-seq data analysis,
>> do we need remove PCR duplicate before we look for DE (differential
>> expressed)genes
>> some people say yes, some say no
>
> Here's my reply from last time this was asked on this list:
>
> https://stat.ethz.ch/pipermail/bioconductor/2011-February/037810.html
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

_______________________________________________
Bioconductor mailing list
Bioconductor at r-project.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor