Hi, everyone: I have some problems with affy package in bioconductor.I ran both eset<-rma(affybatch) and eset<- expresso(affybatch, normalize.method="quantiles", bg.method="rma", pmcorrect.method="rma",summary.method="medianpolish") The results are different. The former seemed to run much faster. I also rarn dchip. eset<-expresso(affybatch, normalize.method="invariantset", bg.correct=F, pmcorrect.method="pmonly", summary.method="liwong") I have 2 groups and each group has 19 samples. The result came out finally, but with some warnings like the following: 1: No convergence in inner loop after 50 in outerler tieration 1 in: fit.li.wong(probes, ...) 2: No convergence achieved in outlier loop in: fit.li.wong(probes, ...) 3: No convergence achieved in outlier loop in: fit.li.wong(probes, ...) 4: No convergence in inner loop after 50 in outerler tieration 1 in: fit.li.wong(probes, ...) 5: No convergence in inner loop after 50 in outerler tieration 1 in: fit.li.wong(probes, ...) 6: No convergence in inner loop after 50 in outerler tieration 1 in: fit.li.wong(probes, ...) 7: No convergence in inner loop after 50 in outerler tieration 1 in: fit.li.wong(probes, ...) 8: No convergence achieved in outlier loop in: fit.li.wong(probes, ...) 9: No convergence achieved in outlier loop in: fit.li.wong(probes, ...) 10: No convergence in inner loop after 50 in outerler tieration 1 in: fit.li.wong(probes, ...) Thanks a lot Wei

On Mon, Jan 20, 2003 at 09:35:29PM -0700, Wei.Hu@uchsc.edu wrote: > Hi, everyone: > I have some problems with affy package in bioconductor.I ran both > eset<-rma(affybatch) and eset<- expresso(affybatch, > normalize.method="quantiles", bg.method="rma", > pmcorrect.method="rma",summary.method="medianpolish") > The results are different. The former seemed to run much faster. The 'rma' was supposed to be a streamlined version in C of the second one. It was included as such few days before the release. It appears that the results differ because the code differs (which is I think is both a problem and should have been stated in the documentation.) Sorry for that. > > I also rarn dchip. eset<-expresso(affybatch, > normalize.method="invariantset", bg.correct=F, pmcorrect.method="pmonly", > summary.method="liwong") I have 2 groups and each group has 19 samples. > The result came out finally, but with some warnings like the following: The warnings are only stating that something caused trouble (in this case there is no convergence). You may want to take note of these genes and be extra-careful with them in your analysis. I just saw that a part of the warning/error handling stuff did not make it to the release. You may try the following if the warnings cause you trouble: options(show.error.messages = FALSE) eset<-expresso(affybatch, normalize.method="invariantset", bg.correct=F, pmcorrect.method="pmonly", summary.method="liwong", warnings=FALSE) Hopin' it helps, Laurent > > 1: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 2: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 3: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 4: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 5: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 6: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 7: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 8: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 9: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 10: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > Thanks a lot > > Wei > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > http://www.stat.math.ethz.ch/mailman/listinfo/bioconductor

hi! the method RMA is constantly being improved. the affy package function "rma" will reflect these changes, so expect to see different versions in future release. for now, expresso, as you called it, will give you the original version of RMA. there is a subsection in the vignette entitled "C implementation of RMA" that contains more information (and in the future better documentation). rafael On Mon, 20 Jan 2003 Wei.Hu@uchsc.edu wrote: > Hi, everyone: > I have some problems with affy package in bioconductor.I ran both > eset<-rma(affybatch) and eset<- expresso(affybatch, > normalize.method="quantiles", bg.method="rma", > pmcorrect.method="rma",summary.method="medianpolish") > The results are different. The former seemed to run much faster. > > I also rarn dchip. eset<-expresso(affybatch, > normalize.method="invariantset", bg.correct=F, pmcorrect.method="pmonly", > summary.method="liwong") I have 2 groups and each group has 19 samples. > The result came out finally, but with some warnings like the following: > > 1: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 2: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 3: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 4: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 5: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 6: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 7: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > 8: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 9: No convergence achieved in outlier loop > > in: fit.li.wong(probes, ...) > > 10: No convergence in inner loop after 50 in outerler tieration 1 > > in: fit.li.wong(probes, ...) > > Thanks a lot > > Wei > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > http://www.stat.math.ethz.ch/mailman/listinfo/bioconductor >