[BioC] limma - normalizeBetweenArrays function

[Gordon K Smyth]

Dear Mike,

Loess normalization (cyclic or otherwise) should only be applied 
after log transforming the intensities.

normalizeBetweenArrays() will accept an EListRaw object containing 
unlogged intensities, but it will take logs before doing cyclic loess 
normalization.

If you pass a numeric matrix to normalizeBetweenArrays(), then you are 
expected to have logged yourself.

Best wishes
Gordon

> Date: Fri, 10 Jan 2014 17:10:25 +0000
> From: MIKE STARKEY <Michael.Starkey at aht.org.uk>
> To: "'bioconductor at r-project.org'" <bioconductor at r-project.org>
> Subject: [BioC] limma - normalizeBetweenArrays function
>
> Hi,

>    I've been using the limma normalizeBetweenArrays function to perform 
> cyclic loess normalisation on single channel microarray data (non-log 
> transformed signal intensities). Amongst the normalised signal 
> intensities returned is one negative value. There is nothing 'odd' about 
> the corresponding original value in the input signal intensity file. I 
> presume that normalizeBetweenArrays is not expected to return negative 
> signal intensities, and so any advice/insights would be gratefully 
> received. I have already referred to the limma user guide, and to the 
> 'normalizeBetweenArrays function help document' accessible online via R.
>
> Many thanks.
>
> Mike

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}