[BioC] what's really in hgu133plus2.db?

[James W. MacDonald]

Hi David,

On 2/18/2011 11:41 AM, David Iles wrote:
> Dear All,
>
> Can anyone point me to a URL where I can obtain an overview of the
> sources of the data incorporated in the current version of
> hgu133plus2.db? I saw to my horror that the actual probesets are
> based on a really obsolete human genome assembly (2003), which has
> changed significantly over the years. As have also genes, gene
> locations, genomic intervals, RefSeq/UniGene entries etcetcetc......

So what exactly is the question? As you note, the chip was designed in 
the early 2000's, so was necessarily based on a (now) old version of the 
UniGene database. That is the downfall of the expression arrays; they 
are stale almost from the instant they hit the market.

Since the probesets are based on things that may now be different, it is 
to a certain extent irrelevant how current the hgu133plus2.db data are, 
because the probeset --> gene mappings may be suspect. You can update 
the gene info all you want, but if the probeset doesn't actually measure 
a given transcript, then what is the point?

We base the annotation on the probeset --> entrez gene mappings supplied 
by Affymetrix, which are supposed to be updated regularly. Not having 
checked that (and given the fact that we take no stance on the veracity 
of these mappings), they are what they are. Any significant results will 
require close inspection of the probesets to determine if you believe 
that they measure what they purport to measure.

As an alternative, you can try the MBNI re-mapped probesets, which both 
update the mappings and remove replicate probesets (by creating single 
probesets per gene/transcript/etc). They can be obtained via biocLite, 
or individually here:

http://brainarray.mbni.med.umich.edu/Brainarray/Database/CustomCDF/CDF_download.asp

Best,

Jim


>
> Thanks
>
> Dave Dr David Iles Institute for Integrative and Comparative Biology
> University of Leeds Leeds LS2 9JT
>
> d.e.iles at leeds.ac.uk
>
> _______________________________________________ Bioconductor mailing
> list Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor Search the
> archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor

-- 
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues