[BioC] X:Map and 1.0 ST Gene Arrays, Introns

[Thomas Hampton]

As I understand it, X:Map is designed primarily to handle the 1.0  
Exon arrays from Affymetrix. The 1.0 ST Human Gene arrays mostly
lack the extended probe coverage that allows one to sort out  
differences between alternative splicing and so forth. That being
said, the Human Gene 1.0 ST array has on it roughly 2,000 probe sets  
that target normal gene  intronic regions. I am a little unclear why  
Affy
left them in there, but I think it is reminiscent of the old  
"mismatch" probes on earlier generation arrays. I think the idea was, in
both cases, that really nothing was meant to hybridize to these  
regions, so they could offer a handy measure of non
specific binding.

Or not.

In our case, we see quite a nice signal from our intronic probes. In  
fact, differences in these intronic regions seems to correlate
well with treatment, so a high proportion of our differentially  
expressed probes turn out to be intronic.

This strikes me as odd, because I thought our isolation procedure  
would result in pretty pure mature mRNA.

I'd also like to know what is on either side if the probes that are  
coming up. If this is unspliced pre-mRNA, than
what gene does it belong to? Affy's annotations are mute on this  
subject, so I was wondering whether X:Map might be able
to help out, even though, as I understand it, it really is not  
designed for the array I am using.

Thanks awfully much.

Tom