[BioC] Problem reading CEL files

[Xia Han]

Yes it was the problem in the line of DatHeader=......HG-U133A.1sq

the 46th cel file had it in lower case: hg-u133a.1sq, but others are in
upper case. I corrected it and everything works fine now.

Thanks for the help, Ben.

Xia



Ben Bolstad wrote:

> Could you look in the header of your cel files?
>
> In particular look for the line that says
>
> DatHeader=
>
> look for something ending in .1sq
>
> Do they match between the 46th file and any of the other files. Matching
> means exactly the same including capitialization.
>
> On Thu, 2004-05-06 at 13:07, Xia Han wrote:
> > Hi:
> >
> > I am trying to use affy package to normalize  79 hgu133a chips. I
> > have R1.8.1 and affy 1.3 on windows 2000.
> > I am trying to use justRMA( ) and keep getting error messages
> > like this:
> > "Cel file xxxxxx.cel does not seem to be of HG-U133A type".
> >
> > I stored 79 cel files in a vector called all.cel.files. Here is
> > what I am getting:
> > ================================================================
> > > library(affy)
> > > data <- justRMA(filenames=all.cel.files)
> > Error in read.probematrix(filenames = filenames) :
> >         Cel file XH2003031149A.cel does not seem to be of
> > HG-U133A type
> > ## This is the #46th chip in all my cel files.
> > > data <-justRMA(filenames=all.cel.files[1:45])
> > Background correcting
> > Normalizing
> > Calculating Expression
> > ## Works fine.
> > > data1 <- justRMA(filenames=all.cel.files[46:50])
> > Error in read.probematrix(filenames = filenames) :
> >         Cel file XH2003030551A.cel does not seem to be of
> > hg-u133a type
> > # This is the # 47th chip.
> > > data1 <- justRMA(filenames=all.cel.files[47:50])
> > Background correcting
> > Normalizing
> > Calculating Expression
> > # Works again.
> > ================================================================
> >
> > If I put the problem chip in the position #1 it will be fine, but
> > there're always other chips ran into the same problem. It's very
> > strange to me. Even I could do several times justRMA( ) and merge
> > them together later, I imagine it would not make sense in terms
> > of  normalization.
> >
> > Help, please!
> >
> > Xia
> >
> > _______________________________________________
> > Bioconductor mailing list
> > Bioconductor at stat.math.ethz.ch
> > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor