Hello, I am not sure if this is a problem, but here we go. When I analyse my Affy mouse4302 data using custom CDF file from the Dai lab that collapses probe/sets onto RefSeq IDs: > affy.batch.object at cdfName<-"Mouse4302_Mm_REFSEQ" I only get 9-digit RefSeq IDs, and none of the more common 6-digit IDs. Maybe I am missing something, but it seems a bit strange to get this bias and I wonder why could it be so? There is no apparent difference between the 6- and 9-digit IDs on the GenBank side, as far as I can tell. I would be very grateful for any clues, cheers jarek -- Jarek Bryk Max Planck Institute for Evolutionary Biology August Thienemann Str. 2 | 24306 Pl?n, Germany tel. +49 4522 763 287 | bryk at evolbio.mpg.de