The original paper is not very clear about the data they provided. I am contacting with them about it. If logratio is treat vs. control, why there are two files with similar names which only differ at the dyes' names? I mean both of two files have logratio column but with different number for the same probe. I heard about JMP can handle data preprocessing , normalization , gene selection, multidimensional scaling , PCA , clustering ananysis , and annotation analysis. I am wondering if bioconductor builds a similar pipeline to perform that? This is some work left from others, really making me headache... On 6/14/07, Sean Davis <sdavis2 at="" mail.nih.gov=""> wrote: > Weiwei Shi wrote: > > Dear Listers: > > > > This is my first time to look at Agilent's data for one of my research > > on cross-platform issue; So this is very newbie's question: > > > > one example is like this: > > two files with cy3_50mg_6h_Rat_3125.txt and cy5_50my_6h_Rat_3125.txt. > > I believe one of them must be control but not sure which one (is there > > a tradition to use cy5 as control?). > > You will almost certainly need to communicate with the biologist that > did the arrays to determine the experimental design. > > > One of the dataset's format looks > > like this: > > > > ProbeName GeneName LogRatio PValueLogRatio > > (+)Pro25G-03 Pro25G -4.92E-01 9.04E-16 > > (-)3xSLv1 NegativeControl 0.00E+00 1.00E+00 > > A_43_P21252 CB546590 1.64E-02 7.96E-01 > > A_42_P534203 272585_Rn -8.69E-03 9.16E-01 > > A_43_P22195 CB547437 4.77E-02 4.90E-01 > > A_43_P16421 AA964066 -9.25E-04 9.87E-01 > > (+)Pro25G-02 Pro25G -1.58E+00 1.10E-33 > > A_43_P13118 NM_130424 4.12E-02 4.68E-01 > > A_43_P19445 CB605581 -1.74E-03 9.93E-01 > > A_43_P11302 BQ206007 -9.17E-02 1.56E-01 > > A_43_P22361 AA964019 -1.31E-01 5.83E-01 > > A_43_P10152 BF420136 6.54E-02 4.11E-01 > > A_42_P573643 234509_Rn 5.87E-02 2.43E-01 > > > > I assume the LogRatio is the signal against background? But what is > > (+)Pro25G-03 or (-)3xSLv1? > > I think the LogRatio is probably the ratio between Red and Green, but > there is not a way to tell here. Have these files been manipulated in > excel or something? Agilent files typically have a much larger number > of columns and have a 9-line header. The (+)Pro25G-03 and (-)3xSLv1 are > controls. > > > btw, are there some packages to read this type of data format? > > If you have files like the one above, read.table will read them just > fine. If you have actual Agilent files (which these are not, I don't > think), then the limma package will read them. > > Sean > -- Weiwei Shi, Ph.D Research Scientist GeneGO, Inc. "Did you always know?" "No, I did not. But I believed..." ---Matrix III