[BioC] converting probe names to gene list

Seth Falcon sfalcon at fhcrc.org
Tue Mar 21 18:39:54 CET 2006


Hi Mark,

"mark salsburg" <mark.salsburg at gmail.com> writes:
> I have a plain text file with 12,000 probe id names from an
> affymetrix chip.

If it is a chip for which a Bioconductor annotation package exists,
then you can do this easily with Bioconductor.  

For example, if you have probes from the hgu95av2 chip, then the
following is a *sketch* of what you would do:

  library(hgu95av2)
  probeIds <- read.table(someFileWith12kIds)
  syms <- mget(probeIds, hgu95av2SYMBOL, ifnotfound=NA)
  if (any(is.na(syms))) cat("a probeId didn't map to a symbol!\n")
  write.table(unlist(syms), row.names=FALSE)
  
> I am using AffyNet from Affymetrix.net and going to Batch Query in the
> Expression section where it says
> "Retrieve annotation for a probe list", where I search for "probe set id"
> and view the list with "gene list". Is this
> correct?
>
> Some problems encountered so far:
>
> 1) affynet tells me it can only export 3000 genes
> 2) afffynet changes the order of the outputted genes by putting probe ids
> that start with AFFY last. (they are in fact first)
> 2) affynet exports out more information than I need, I just need the gene
> symbol.
>
> Can someone please lead me through the procedure to output out a simple text
> file of gene symbol?

If you are looking for help using affynet, I think you haven't read
the posting guide :-)  That is, this is a place to ask about
Bioconductor packages.

+ seth



More information about the Bioconductor mailing list