[BioC] Custom CDF Version 12 is released

[Yair Benita]

Dear Manhong,
I would just like to note here my own experience installing v12 on R 2.9.1
on MacosX.
Using the R package installer I selected "Other Repository" and used the
URL: http://brainarray.mbni.med.umich.edu/bioc/
Next I installed several custom CDF files in binary format and got the
following error for every cdf I tried:

Error in function (package, help, pos = 2, lib.loc = NULL, character.only =
FALSE,  : 'hgu95av2hsentrezgcdf' is not a valid installed package

For a while I could not figure this problem, but then I re-installed the
custom CDF from source (by simply un-checking "Binary Format Package") and
it started working fine. I am not sure what is wrong with the binary format
but hope it will save others some time. I am running on a Quad MacPro.

Thanks,
Yair

on 7/30/09 2:38 PM, Manhong Dai at daimh at umich.edu wrote:

> Hi,
> 
> 
> We just release Custom CDF Version 12 officially.
> 
> 
> All files are available at
> http://brainarray.mbni.med.umich.edu/Brainarray/Database/CustomCDF/genomic_cur
> ated_CDF.asp. All R packages are also in our own R repository. You should be
> able to use biocLite() to install them, just like other bioconductor packages.
> 
> 
> As a reminder, almost all Affymetrix chips are supported in Custom CDF. You
> can even use traditional method (ReadAffy and rma) to analyze latest gene
> chips, such as HuGene10stv1, HTHGU133PlusPM, etc.
> 
> 
> We also updated probeset web application that has all probe's genome location,
> user can input whatever probeset name to get sequence, chromosome and
> location. It is at http://arrayanalysis.mbni.med.umich.edu/probeset/.
> 
> 
> In addition, We released a R library 'CustomCDF' at
> http://arrayanalysis.mbni.med.umich.edu/MBNIUM.html#CustomCDF. The library has
> a few useful tools, such as retrieving probes that have a hit with snp, and
> removing them from cdf library on-the-fly, etc.
> 
> 
> By the way, our public WGAS is still using Custom CDF version 11, it should be
> updated in a few weeks.
> 
> 
> We want to thank Dr. De Bondt, An at Johnson & Johnson for finding a bug
> during preview.
>  
> 
> Best,
> Manhong Dai
> The Molecular & Behavioral Neuroscience Institute (MBNI)
> University of Michigan
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor