[BioC] How to specify Q75 for locfunc in DESeq2 estimateSizeFactors ? (Michael Love) (Simon Anders)

Schnell, Dan schnell.dj at pg.com
Thu May 8 14:43:38 CEST 2014


Thanks very much to both of you for the 2 simpler solutions--I really appreciate it.
Dan Schnell

------------------------------

Message: 7
Date: Wed, 07 May 2014 15:45:06 +0200
From: Simon Anders < >
To: bioconductor at r-project.org
Subject: Re: [BioC] How to specify Q75 for locfunc in DESeq2
	estimateSizeFactors ? (Michael Love)
Message-ID: <536A38E2.8010407 at embl.de>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Hi Dan

On 07.05.2014 15:00, Schnell, Dan wrote:
> Calling estimateSizeFactorsForMatrix with locfunc=quantile does
> return sizeFactors based on min, Q25, median, Q75 and max, and one
> can then go through the normalizationFactors [...]

This is because the 'locfunc' argument should be a function with one 
argument, but quantile takes to.

You should pass an anonymous function that calls 'quantile' with the 
second argument set to 0.75, as follows:

      dds <- estimateSizeFunctions( dds,
         locfunc = function(x) quantile( x, 0.75 ) )


   Simon



More information about the Bioconductor mailing list