<kfbargad at="" ehu.es=""> writes: > Dear list, > > sorry for this very basic question but I have searched the R manual > without success. What function do I need to randomly split a > data.frame into two data.frames (for example a data.frame containing > 200 rows and 5 columns into two of say 150 and 50 rows each)? Should I > first create a vector of 150 random numbers between 1 and 200 (but > how?) and then use subset() function? This is not a BioC specific question and would be better asked on the R-help list. Yes, you can do this by creating a vector of row indices. See help("sample"). You don't need subset. Just "[". v = sample(1:nrow(df)) df[v[1:50], ] df[v[51:nrow(df)], ] + seth -- Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center http://bioconductor.org