[BioC] AveExp" column has dissappeared from limma topTable

Gordon K Smyth smyth at wehi.EDU.AU
Fri Apr 6 01:05:35 CEST 2007


On Thu, April 5, 2007 11:31 pm, Mikhail, Amy wrote:
>  -----Original Message-----
> From: Mikhail, Amy
> Sent: Thursday, April 05, 2007 2:30 PM
> To: 'Seth Falcon'
> Cc: bioconductor at stat.math.ethz.ch
> Subject: RE: [BioC] AveExp" column has dissappeared from limma topTable
>
> Hi Seth and Gordon,
>
>>> Also, what is the difference between an exprSet and an ExpressionSet?
>
>>exprSet and ExpressionSet are both classes used to represent microarray
> data.  exprSet is now deprecated in favor of ExpressionSet.
>
>>+ seth
>
> In a previous post, Gordon has said:
> "If the first argument to lmFit() is an MAList or exprSet or
> ExpressionSet then Amean is computed. It is actually not clear that this
> is right behaviour in the case of an ExpressionSet, but that's how it is
> for now."
>
> So, that's what made me wonder what is the specific difference between
> exprSet and ExpressionSets that would make Amean a sensible thing to
> extract in the first case but not necessarily in the second?
>
> Many thanks,
> Amy

Amy, I'm not an ExpressionSet developer, and you can read the documentation as easily as I can
(have you done that?), but ExpressionSet seems to have more general intent than exprSet.
ExpressionSet seems intended to contain any sort of expression data at all, and that could
conceivably mean log-ratios instead of log-intensities, and hence how would you compute average
intensities?  I suspect that in practice ExpressionSet, like exprSet, will be most commonly be
used for Affy and Illumina data, hence the expression values will virtually always be
log-intensities.  As you are using Affymetrix data, so you will never have to worry about any of
this.

Gordon



More information about the Bioconductor mailing list