Hi Bioconductor, The library(sma) is no longer supported on R-Console. I was told to download library(limma) instead. However, I am encountering problem when I need to work on the exercises following the examples given in this website link below. Could you please let me know what I should do instead of using the code given in the examples in the link below for question 3 (Q3). Thank you. http://bioinf.wehi.edu.au/marray/ibc2004/lab4/lab4.html#5.%20R%20and%2 0Bioconductor%20WWW%20resources Q3. You might also like to create a cluster image using the function heatmap in the package stats. We will create a clustering image that is commonly seen in many microarray literature. By default, this function performs hierarchical clustering on both genes and samples and the function will slow down considerably if the number of genes are too large. For illustration purposes, we have selected 100 variable genes. library(sma) golubvar <- apply(golub, 1, var, na.rm = TRUE) top100 <- stat.gnames(golubvar, 1:length(golubvar), crit = 100)$gnames heatmap(golub[top100, ]) Regards, Sook Wah

Hi Sook Wah, On 11/1/2010 10:39 PM, Yee, Sook Wah wrote: > Hi Bioconductor, > > The library(sma) is no longer supported on R-Console. I was told to download library(limma) instead. However, I am encountering problem when I need to work on the exercises following the examples given in this website link below. Could you please let me know what I should do instead of using the code given in the examples in the link below for question 3 (Q3). Thank you. > > http://bioinf.wehi.edu.au/marray/ibc2004/lab4/lab4.html#5.%20R%20and %20Bioconductor%20WWW%20resources > > Q3. You might also like to create a cluster image using the function heatmap in the package stats. We will create a clustering image that is commonly seen in many microarray literature. By default, this function performs hierarchical clustering on both genes and samples and the function will slow down considerably if the number of genes are too large. For illustration purposes, we have selected 100 variable genes. > > library(sma) > golubvar<- apply(golub, 1, var, na.rm = TRUE) > top100<- stat.gnames(golubvar, 1:length(golubvar), crit = 100)$gnames > heatmap(golub[top100, ]) If I am not mistaken, these data can be found in the golubEsets experimental data package. In addition, the only thing that sma is used for here is to get the top 100 probesets, based on variance. We don't need sma for that. > biocLite("golubEsets") > library(golubEsets) > data(package="golubEsets") Data sets in package 'golubEsets': Golub_Merge Combined Test and Training Sets from the Golub Paper Golub_Test Test Set Data from the Golub Paper Golub_Train Training Set from the Golub Paper golubMerge Combined Test and Training Sets from the Golub Paper golubTest Test Set Data from the Golub Paper golubTrain Training Set from the Golub Paper > data(Golub_Merge) > golubvar <- esApply(Golub_Merge, 1, var, na.rm = TRUE) > ord <- order(golubvar, decreasing = TRUE) > top100 <- exprs(Golub_Merge)[ord,][1:100,] > heatmap(top100) Best, Jim > > > Regards, > > Sook Wah > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- James W. MacDonald, M.S. Biostatistician Douglas Lab University of Michigan Department of Human Genetics 5912 Buhl 1241 E. Catherine St. Ann Arbor MI 48109-5618 734-615-7826 ********************************************************** Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues