[BioC] LIMMA: plotMDS

Gordon K Smyth smyth at wehi.EDU.AU
Mon May 7 12:44:03 CEST 2012

Dear Kripa,

Just to be clear on terminology, MDS computes principle coordinates, which 
are not quite the same as principle components.

Regardless of the name however, there actually is no set of genes that 
corresponds to each dimension.  The "pairwise" method in plotMDS() uses a 
potentially different set of genes for every pair of samples, and all 
these pairwise sets of genes feed to some extent into all the dimensions.

Best wishes

> Date: Sun, 6 May 2012 23:21:03 +0000
> From: Kripa R <kripa777 at hotmail.com>
> To: <bioconductor at r-project.org>
> Subject: [BioC] LIMMA: plotMDS
> Hi I was wondering how I can figure out which genes correspond to each 
> of the principle components ($cmdscale.out). Gene.selection is set to 
> pairwise for the top 500 but I'd like to know exactly which genes are 
> being considered for cmdscale.out[,3], for example.
> Thanks for the help!
> .kripa

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

More information about the Bioconductor mailing list