Affymetrix Gene 1.0 ST redundant probeset sequences

[Nathan Salomonis]

I thought I would let people know (since I didn't) that there are distinct
probeset IDs on the Affymetrix Gene 1.0 ST array (probably other arrays too)
that reference the exact same set of probes. Hence, the expression values
for different probesets on different chromosomes will be identical since
they reference the same underlying probes.

For example:
7927090
8102531
7911332
7945348
7998116
8031998
7924464

All reference the same probe/atom IDs. The result will be something like
this:

7927090 8.58586 8.69401 8.37415
8102531 8.58586 8.69401 8.37415
7911332 8.58586 8.69401 8.37415

This will manifest itself as different genes having the same expression
pattern in AltAnalyze. While this is not technically wrong (100% duplicated
sequences in the genome) this means that one signal can incorrectly give
rise to multiple observations. This is something we have tried to prevent in
the design of our databases but can only observed at the pgf file-level.

Best,
Nathan