Hello, I am a very new user of R and hope you can help. I have tried the fora and online help. I am using limma to analyse microarray data. I have run into a problem with GeneSetTest which i don't understand, but think it is something to do with my input data. My whole data set has 18296 rows of t values. I have created a separate subset for the genes on an ICE whose t values I want to compare in Gene Set Test. Some of these values are negative as is expected with t values. The small dataset is seen below > sub.2 [1] -1.628088679 1.199633873 0.652152009 3.214887735 3.233349792 [6] 2.362553469 3.525487896 4.940010275 4.897654582 -0.002203589 [11] 1.845625471 1.561264994 7.507653938 17.546755720 7.080742224 [16] -0.025233341 2.410513643 1.083376106 1.908051944 1.112365831 [21] 0.564130982 1.709511634 10.066341340 12.229110920 5.529650005 [26] 1.701063357 2.229950100 2.721246420 5.390680871 3.722832322 [31] 4.796353818 3.525246899 4.822326403 -4.002596576 0.835416155 [36] 2.537629585 1.991560632 -0.041158816 0.802931410 -1.869976733 [41] 0.238770948 -1.052793736 -0.553059499 -0.023182488 The whole array data is designated sub.3 of my data table and I have checked this contains only numeric characters ( and when run with a dummy "sel" set is fine) when I run this: library(limma) > sel<-sub.2 > stat<-sub.3 > geneSetTest(sel, stat, alternative="up") Error in statistics[selected] : only 0's may be mixed with negative subscripts I expect i am doing something simple wrong. Any help would be appreciated. Thank you, Esther Robinson

On Wed, Dec 14, 2011 at 5:09 AM, Esther Robinson < esther.robinson@sjc.ox.ac.uk> wrote: > Hello, > I am a very new user of R and hope you can help. I have tried the fora and > online help. I am using limma to analyse microarray data. > I have run into a problem with GeneSetTest which i don't understand, but > think it is something to do with my input data. My whole data set has 18296 > rows of t values. > I have created a separate subset for the genes on an ICE whose t values I > want to compare in Gene Set Test. Some of these values are negative as is > expected with t values. > The small dataset is seen below > > sub.2 > [1] -1.628088679 1.199633873 0.652152009 3.214887735 3.233349792 > [6] 2.362553469 3.525487896 4.940010275 4.897654582 -0.002203589 > [11] 1.845625471 1.561264994 7.507653938 17.546755720 7.080742224 > [16] -0.025233341 2.410513643 1.083376106 1.908051944 1.112365831 > [21] 0.564130982 1.709511634 10.066341340 12.229110920 5.529650005 > [26] 1.701063357 2.229950100 2.721246420 5.390680871 3.722832322 > [31] 4.796353818 3.525246899 4.822326403 -4.002596576 0.835416155 > [36] 2.537629585 1.991560632 -0.041158816 0.802931410 -1.869976733 > [41] 0.238770948 -1.052793736 -0.553059499 -0.023182488 > > The whole array data is designated sub.3 of my data table and I have > checked this contains only numeric characters ( and when run with a dummy > "sel" set is fine) > when I run this: > library(limma) > > sel<-sub.2 > > stat<-sub.3 > > geneSetTest(sel, stat, alternative="up") > Error in statistics[selected] : > only 0's may be mixed with negative subscripts > Hi, Esther. The first argument to geneSetTest is an index vector and is used to select the values from the statistics argument. You have used a numeric vector. Take a look at the help for geneSetTest a bit more closely and then get back to us with questions. Sean > I expect i am doing something simple wrong. Any help would be appreciated. > Thank you, > Esther Robinson > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > [[alternative HTML version deleted]]