Hello, Good day and nice to meet you. After considering and reconsidering the content of this posting, I come to the conclusion that the solutions to the questons in this posting could be helpful not only for those who are just dabble into the GAGE application, but also for those who are entry users of R packages. I came across the GAGE package and found that it might be of great use for my current data analysis. After went through the manual, I understood the context and arguments written for the demo file. However, I doubted how to implement new arguments to suit my dataset. It is also noteworthy that through the web-base forum, I found no single answer that is applicable to my situation from the previous queries. Here are the questions, 1. I realized that most, if not all or the R-users wrote their own code for particular R function with slight differences. I also tried to create some for my analysis. Is the following script correct for importing the expression dataset (.csv) to be read by GAGE? Or is it better to be formatted in text-delimited format (.txt)? data <- read.table("C:/??????/Array.csv", sep=",", header=TRUE) 2. How to introduce customized gene set file from C:/ and adapt the following scripts shown in the manual to execute corresponding functions with this ???foreign??? get set file? datakegg.gs) datago.gs) lapplykegg.gs[1:3],head) 3. To start analyzing the data, would you mind to show me an example of script for the imported dataset, corresponding to the first example in the manual? What is meant by .kegg.p and .go.p? gse16873.kegg.p <- gage(gse16873, gsets = kegg.gs, ref = hn, samp = dcis) gse16873.go.p <- gage(gse16873, gsets = go.gs, ref = hn, samp = dcis) Let???s assume that the control is hn, positive sample is dcis, whilst the ???gse16873??? file is not pre-loaded into the GAGE library, but being introduced from elsewhere in C:/. 4. What is the least number of sample for control and treatment to be included for GAGE analysis, respectively? Is there any limit of the total number of expression datasets to be analyzed in each session of GAGE? Many literatures shown that other methods have statistical limitations for particular size of sample dataset, which influence the outcome of the whole analysis. Your reply is utmost appreciated. Thanks in advance! Sincerely, Aurelle -- output of sessionInfo(): The dataset contains control and treatment (2 groups). The replicate ranged from 2 to 10 for each group. -- Sent via the guest posting facility at bioconductor.org.