Re: [aroma.affymetrix] FIRMAGene with masked CEL files

2009-11-16 Thread Mark Robinson
Hi Tim. The FIRMAGene code is not designed to play nice with NAs in the probe level data. I'm somewhat hesitant to include this because it would require making assumptions about how people have introduced NAs into their CEL files. If you want to exclude a subset of the standard collection

Re: [aroma.affymetrix] FIRMAGene with masked CEL files

2009-11-16 Thread Henrik Bengtsson
Hi. Update aroma.affymetrix to v1.3.0 and report back. If/when you get the error, please also report what traceback() outputs. /Henrik PS. I'm even surprised that you got this far because the version you have now is not compatible with the R.filesets version you have. On Mon, Nov 16, 2009 at 5:

[aroma.affymetrix] FIRMAGene with masked CEL files

2009-11-16 Thread Tim Rayner
Hi, I've been experimenting with using FIRMAGene to detect differential splicing, and while the package seems to be working well we're seeing quite a lot of splicing events which we think are due to poorly- detected probes on the array. I've been trying to filter these probes out by identifying th