Hi Guy -- As noted, this should be sent to the Bioconductor mailing list, see http://bioconductor.org/docs/mailList.html.
Some more comments below... guy.tillingh...@rivhs.com writes: > Full_Name: Guy W. Tillinghast > Version: 2.8.0 > OS: Windows XP professional > Submission from: (NULL) (24.248.24.3) > > > I am encountering difficulty asking R to manipulate the correct columns in > Expression Set class (object 4). > > I download the ALL data with: > library(golubEsets) > data(Golub_Merge) > > Note, the data has the samples not in order. This is not R's fault (at least > not that I can tell): >> Golub_Merge$Samples > [1] 39 40 42 47 48 49 41 43 44 45 46 70 71 72 68 69 67 55 56 59 52 53 51 50 > 54 > [26] 57 58 60 61 65 66 63 64 62 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 > 16 > [51] 17 18 19 20 21 22 23 24 25 26 27 34 35 36 37 38 28 29 30 31 32 33 'Samples' is a covariate, not an index into the ExpressionSet. It is like any of the other 11 covariates in phenoData (try pData(Golub_Merge) to get a data frame of all covariates). > I want a subset: >> learning.set<-c(1,2,3,6,8,10,11,12,13,15,16,17,18,19,20,21,23,24,26,30,31,32,34,35,36,37,39,43,44,45,46,47,48,49,50,51,52,53,55,56,57,58,59,60,62,63,64,65,66,67,68,69,70,72) >> learningEset<-Golub_Merge[,learning.set] >> learningEset$Samples > [1] 39 40 42 49 43 45 46 70 71 68 69 67 55 56 59 52 51 50 57 65 66 63 62 1 > 2 > [26] 3 5 9 10 11 12 13 14 15 16 17 18 19 21 22 23 24 25 26 34 35 36 37 38 > 28 > [51] 29 30 31 33 This selects columns 1, 2, 3 etc of the ExpressionSet, and the corresponding phenoData. The column 1 corresponds to Sample 39, so you select that Sample, and so on. If you want individuals with particular values of the Samples column of phenoData, you might say something like Golub_Merge[, Golub_Merge$Samples %in% learning.set] just as you might select all the male samples with Golub_Merge[, Golub_Merge$Gender == "M"] Kind of a powerful idiom. > Note what happened: > 1) the order is difference than learning.set > 2) samples have been switched: example: sample 72 out, sample 71 in. > > Okay, I troubleshoot: maybe it matters what order I request samples: > >> learning.set<-c(39,47,48,49,43,44,45,46,70,72,68,69,67,55,56,59,52,53,51,50,57,58,60,65,66,63,64,62,1,2,3,6,8,10,11,12,13,15,16,17,18,19,20,21,23,24,26,34,35,36,37,30,31,32) >> learningEset<-Golub_Merge[,learning.set] >> learningEset$Samples > [1] 5 13 14 15 9 10 11 12 31 33 29 30 28 21 22 25 18 19 17 16 23 24 26 37 > 38 > [26] 35 36 34 39 40 42 49 43 45 46 70 71 68 69 67 55 56 59 52 51 50 57 62 1 > 2 > [51] 3 65 66 63 > Frankly, this is troubling that R did not do what it was told. All Bioconductor packages have vignettes, which is a good place to start to understand a package. An ExpressionSet is defined in Biobase, so visit (from the front page of the Bioconductor site, following the software link) http://bioconductor.org/packages/2.3/bioc/html/Biobase.html and read the ExpressionSetIntroduction.pdf as a starting point. There is also extensive training material (under the 'workshops' tab at the top of the page) and some excellent books. The bioconductor mailing list and mailing list archives are also very helpful places. Martin > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel -- Martin Morgan Computational Biology / Fred Hutchinson Cancer Research Center 1100 Fairview Ave. N. PO Box 19024 Seattle, WA 98109 Location: Arnold Building M2 B169 Phone: (206) 667-2793 ______________________________________________ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
