[BioC] retrieving SNP names for MM probes using oligo

[Benilton Carvalho]

Hi Stacey,

I'm going to assume you're using the latest R and latest oligo...

I didn't see any reason why probeNames would give different results  
for MM probes, so I didn't implement it... I'll add that in the  
future....

Anyways, we can easily check the result with some SQL:

library(pd.mapping250k.nsp)
sql <- "SELECT man_fsetid FROM featureSet, mmfeature WHERE  
featureSet.fsetid = mmfeature.fsetid ORDER BY fid"
mm.results <- dbGetQuery(db(pd.mapping250k.nsp), sql)[[1]]
pm.results <- probeNames(pd.mapping250k.nsp)
identical(mm.results, pm.results)

Your impression about 40 probes per SNP is correct only for XBA and  
HIND arrays. On the 250K arrays, that doesn't happen necessarily. By  
the way, on the 250K platform, don't expect all SNPs to have probes  
on both strands, it's not a requirement for NSP/STY either.

Please, let me know if this makes sense,

Best,

b


On Oct 26, 2007, at 6:50 PM, Stacey Burrows wrote:

> Hi everyone,
>
> I am trying to retrieve SNP names for MM probes on the Affymetrix  
> 500K SNP array (specifically, some of the HapMap samples). In  
> oligo, there is a function called probeNames which gives back the  
> names, however, it only does so for PM probes. Is there an  
> analogous function for MM probes in oligo or any other package?
>
> Also, I was under the impression that each SNP has a corresponding  
> set of 40 probes (20 MM, 20 PM). However, when I view the data in  
> R, it seems like most SNPs only have 12 PM probes and a much  
> smaller proportion have 20. Is this correct or an indication that  
> the data are not imported correctly? In total, I see 6,443,200  
> probes for the Nsp chip, which doesn't equal 40 x 262,264 SNPs.
>
> Due to these problems, I am so far finding it, in general,  
> difficult to explore the data using R.
>
> Thanks for any help guys!
> Stacey