[BioC] geneplotter and base position

[david neil hayes]

A question in 2 parts:

1.  Are the "probeset base positions" in the meta-data packages (for the 
affy chips) derived explicitly from the affymetrix annotation provided by 
the company in their quarterly annotation files?

2.  When there are multiple "base positions" mapped to a single probeset, is 
that due to the fact that there may be an overlapping transcript?  
Specifically I am referring to the case where one probeset seems to be 
annotated to multiple positions in the same general area of a single 
chromosome.

See below for details.
Thanks in advance,
Neil Hayes



using R - 1.9.0
geneplotter 1.4.0
hgu95av2 1.5.1 (downloaded today 5/24/04)

I used the geneplotter package to generate an object of class 
"chromLocation" for the 95av2 chip.  Next I extracted the chromosome 
locations as below:

chrObj <- buildChromLocation("hgu95av2")
ChromLoc <- chromLocs(chrObj)

As described in the vignette for the package there are a limited number of 
probesets which are annotated to multiple chromosomes.  What I did not see 
in the vignette was the phenomonon of many probesets (my analysis shows 908 
probesets) being annotated to more than one locus on a chromosome.

For example: probeset "1003_s_at" is mapped to 2 base positions on Chr 11 
(118292218) and (118301752)

The most current affy annotation (the csv file from the Affy web site) shows 
that "1003_s_at" is mapped to chr11: 118302397-118302992 with an overlapping 
transcript chr11:118301752-118303772.  I suspect the disagreement may be 
that the R annotation is from March 2004 while the current Affy annotation 
was updated in April.