[BioC] PLIER affinities

[James W. MacDonald]

Jeremy Gollub wrote:
> Hi, All -
> 
> I'm confused by the affinities returned by justPlier (from the plier
> package).  My understanding is that the PLIER algorithm is supposed to
> produce a single affinity per PM/MM probe pair, across the entire set of
> arrays analyzed.  However, justPlier() returns the values in a matrix of
> dimensions <# of probe pairs> X <# of chips> - e.g, 175,477 X 18, for
> 18 RAE230A arrays.  The column names of this matrix are the names of the
> samples/arrays; the row names are the probe set names, repeated
> (apparently) once for each probe pair in the set (i.e., there are 11 rows
> with the name 1367452_at).
> 
> Upon closer inspection, the matrix contains only 175,407 non-zero values,
> which would be consistent with a single value per probe pair.  All columns
> contain some non-zero values, approximately 15,923 (one per probe set) in
> the first 11 columns, and 36 in the other 7 (perhaps due to control probe
> sets with more probes).  All rows below 15,924 contain only zeros.
> 
> Do I misunderstand?  Or should the affinities in fact be a one-dimensional
> vector with one value per probe pair, making this an error in the way
> justPlier(), or the C++ code, return the values?

You are correct; justPlier should return only one value per probeset. 
You might try upgrading to R-2.2.0 and BioC 1.7 - justPlier() works as 
expected for me with the updated versions.

Best,

Jim


> 
> SessionInfo() output below.
> 
> Thanks,
> 
> - Jeremy
> 
> 
>>sessionInfo()
> 
> R version 2.1.1, 2005-06-20, i386-pc-mingw32 
> 
> attached base packages:
> [1]
> "tools"     "methods"   "stats"     "graphics"  "grDevices" "utils"     "datasets" 
> [8] "base"     
> 
> other attached packages:
> rae230acdf      plier       affy reposTools    Biobase 
>    "1.5.1"    "1.0.0"    "1.6.7"   "1.5.19"   "1.5.12" 
> 
> --
> Jeremy Gollub, Ph.D.
> jgollub at genome.stanford.edu
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor


-- 
James W. MacDonald
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623