[BioC] Matchprobe package

[Wolfgang Huber]

Hi Wenqiong,

the error occurs in the .lgExtraParanoia() function, where the "lg" 
stands for Laurent and the function checks the consistency between the 
information in the CDF package and the probe package. The CDF packages 
have been there before, and many people are content with them, but the 
information in the probe package is a superset of the information in the 
CDF, hence the redundancy is necessary, and needs to be checked.

I cannot diagnose exactly what is going wrong from the information you 
give, and since you are using this on a home-made chip design and 
-files, please do yourself have a look at that function. Do

   debug(.lgExtraParanoia)

before calling makeProbePackage, and step through the function to see 
what is going wrong e.g., are the variables pm1, mm1, pm2, mm2 assigned 
NA values?

Best wishes
  Wolfgang


Wenqiong Chen wrote:
> Hi,
> I am trying to make probe package with the functions in the "matchprobe"
> package. However, we have modified the cdf file and probe sequence file
> to accommodate the recent update of the in house genome sequence. I have
> successfully created the cdf package with the modified cdf file, and
> everything worked as far as use rma. However, to get the GCRMA work I
> would need to create the probe package where I am having trouble. I have
> checked that all the probe sequences within the cdf file were included
> in the probe sequence file, although the probe sequence file contain
> more sequences than the ones in the cdf file. 
> Here the error message I got:
> 
> 
>>library(matchprobes)
>>makeProbePackage("disdiv712a", getProbeDataAffy("disdiv712a",
> 
> "diS_div712a_probeSequenceFile.txt", pkgname = NULL, comparewithcdf =
> TRUE), maintainer = "XXXX<XXXX at diversa.com>", version    = "0.0.1",
> force=TRUE)
> Error in rep.default(NA, max(pm1, mm1, pm2, mm2)) :
> invalid number of copies in "rep"
> In addition: Warning messages:
> 1: NAs introduced by coercion
> 2: NAs introduced by coercion
> 3: NAs introduced by coercion
> 
> Can anyone please point out what is wrong?
> Thanks in advance!
> 
> wenqiong
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor


-- 
Best wishes
   Wolfgang

-------------------------------------
Wolfgang Huber
European Bioinformatics Institute
European Molecular Biology Laboratory
Wellcome Trust Genome Campus
Cambridge CB10 1SD
England
Phone: +44 1223 494642
Http:  www.dkfz.de/abt0840/whuber