Thanks Mark - your suggestions have solved my problems and everything is running quickly and smoothly now!
Mark Robinson wrote: > Hi Sebastien. > > Comments below. > > > On 04/03/2009, at 11:34 AM, Sebastien Gerega wrote: > > >> Hi Mark, >> I understand there may be some interesting genes with > 100 probes - I >> just wanted to test it out to see how it impacted on the run >> time...... >> After changing the skip threshold to 500 probes the analysis completed >> in a similar time to what is required for mogene10st and hugene10st >> arrays. >> >> I then examined which probes I eliminated by changing the threshold >> to 500: >> >>> getUnitNames(cdf)[which((cpu1 > 500))] >>> >> [1] "10700001" "10700003" "10700004" "10700027" "10700052" "10700055" >> "10700058" "10700059" "10700063" "10700064" "10700065" "10700066" >> "10700067" >> >> It turns out that all 13 of these probes are listed in the Affymetrix >> annotation file as being "control->affx". Therefore I assume I can >> safely remove them from the analysis. >> >> I also noticed in the annotation file that there are additional >> control >> probes listed as "control->bgp->antigenomic". Any idea what this >> means? >> Would it be safe to remove them? >> > > Indeed, you can safely NOT fit a probe level model for these probesets > and not be missing anything. I'm guessing that "bgp" means background > probes and "antigenomic" meaning the 25mer sequence does not map > anywhere to the (in this case, rat) genome. > > > >> And finally, how would I go about removing probes from the analysis? >> > > You probably just want to remove these probesets at the fitting > stage. My suggestion from before to do this is: > > [...] > plmTr <- RmaPlm(csN, verbose=verbose) > fit(plmTr, units=u, verbose=verbose) > > ... where 'u' is the set of units you want to fit. > > Note: if you are reading expression summaries from the ChipEffectSet > later, you'll probably also want to specify 'units=u' as well. If you > don't fit a PLM for a given probeset and read ALL the chip effects > into a matrix/data.frame, there'll be NAs for those probesets that you > didn't fit on. > > Hope that helps. > Mark > > > > >> thanks, >> Sebastien >> > > > > > > > > > --~--~---------~--~----~------------~-------~--~----~ When reporting problems on aroma.affymetrix, make sure 1) to run the latest version of the package, 2) to report the output of sessionInfo() and traceback(), and 3) to post a complete code example. You received this message because you are subscribed to the Google Groups "aroma.affymetrix" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/aroma-affymetrix?hl=en -~----------~----~----~----~------~----~------~--~---
