In my version of siggenes, I don't see any variables called x and y that can be passed to the function sam. If RG.all contains all the data you wish to compare, you only need to pass this data.frame along with a vector of classlabels that identify which class each column of data belongs to. If you are doing a two-class unpaired analysis, this will be a vector of 0s and 1s. If paired, then paired integers from 1:n, and if oneclass, a vector of 1s. HTH, Jim James W. MacDonald Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 >>> "Anna Cao" <yunie@caltech.edu> 12/19/03 01:50PM >>> Hi, I'm trying to run SAM using the siggenes package in R (version 1.8.1 in Window XP). I have a 16247x204 matrix containing all the arrays in my experiment. This matrix includes several experimental conditions including replicates. I need to run sam for each experimental condition. I try the following command and got this error. sam(RG.alls, x=RG.alls$R[1:16], y=RG.alls$G[1:16]) Error in as.double.default(structure(list(structure(c(NA, NA, NA, NA, : (list) object cannot be coerced to double I assume this error may be related to all the blank (NA) points I have in the matrix. Are there any ways to remove these empty spots for each set of replicates without having to export and arrange the matrix elsewhere? Anna _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor

Hi, it depends on what version of siggenes you have. If you still have the old version (version 0.xx), then Sean is correct. If you have the latest version (version 1.xx), then Jim is right. Sorry for this confusion, but this change in usage was necessary to fit sam() more into the "usual" usage of the Bioconductor functions. Greetings, Holger > In my version of siggenes, I don't see any variables called x and y that > can be passed to the function sam. If RG.all contains all the data you > wish to compare, you only need to pass this data.frame along with a > vector of classlabels that identify which class each column of data > belongs to. If you are doing a two-class unpaired analysis, this will be > a vector of 0s and 1s. If paired, then paired integers from 1:n, and if > oneclass, a vector of 1s. > > HTH, > > Jim > > > > James W. MacDonald > Affymetrix and cDNA Microarray Core > University of Michigan Cancer Center > 1500 E. Medical Center Drive > 7410 CCGC > Ann Arbor MI 48109 > 734-647-5623 > > >>> "Anna Cao" <yunie@caltech.edu> 12/19/03 01:50PM >>> > Hi, > > I'm trying to run SAM using the siggenes package in R (version 1.8.1 > in > Window XP). I have a 16247x204 matrix containing all the arrays in my > experiment. This matrix includes several experimental conditions > including > replicates. I need to run sam for each experimental condition. I try > the > following command and got this error. > > sam(RG.alls, x=RG.alls$R[1:16], y=RG.alls$G[1:16]) > Error in as.double.default(structure(list(structure(c(NA, NA, NA, NA, > : > (list) object cannot be coerced to double > > I assume this error may be related to all the blank (NA) points I have > in > the matrix. Are there any ways to remove these empty spots for each set > of > replicates without having to export and arrange the matrix elsewhere? > > Anna > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor > -- --