[BioC] PLIER affinities redux

[James W. MacDonald]

Jeremy Gollub wrote:
> Hi, James -
> 
> Thanks for clarifying that.
> 
> Looking at the source code for justPlier, it seems there's a misapprehension
> about the probe affinity values - the justPlier code expects to get back
> a value for each probe (note, not probe set) in each sample, whereas the
> algorithm/C++ code actually returns a single value for each probe pair,
> which is applied to all samples.  I think the padding with many extra
> zeros is done by justPlier, not by the C++ code...?

No, there is nothing in the code of justPlier that would do that 
(anyway, I checked it out). The C++ code returns a vector of length 
<number of probes> x <number of chips> that has affinity values for the 
first <number of *probesets*> x <number of chips>, followed by the zeros.

The R code in justPlier simply puts this vector into a matrix and then 
transposes (note that the size of the matrix will be determined by the 
length of r$affinity, since no ncol argument is given).

if (get.affinities) {
         a <- t(matrix(r$affinity, nrow = num_exp))
         colnames(a) <- sampleNames(eset)
         rownames(a) <- probeNames(eset)
         res at description@preprocessing$affinity = a
     }


Best,

Jim


> 
> Crispin, unless I'm missing something, this should probably be considered
> a bug.  In the meantime, I should be able to deconvolute the matrix back
> into the expected vector, but I'm not sure which probe pair to associate
> with each value.
> 
> - Jeremy

-- 
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