The usual choices for non-parametric location tests are the Mann?Whitney U test for independent samples, and the binomial test or the Wilcoxon signed-rank test for paired samples. The test depends on your sample see http://www.graphpad.com/www/Book/Choose.htm Have a look at the multtest package on Bioconductor and the mt.teststat function. It might help you. Nolwenn David martin wrote: > Sorry to ask but what are the tests ?? > > > Nolwenn Le Meur wrote: >> David martin wrote: >>> >>> I have tested different cell surface markers on a sub population of >>> monocytes. The matrix below shows a small snapshot of the different >>> cell surface markers tested (cdA,cdB,cdC..) in different monocytes >>> of patients that are either normal patients or treated patients. >>> The values are the geomtrical mean obtained from the flow. They are >>> log10 values. >>> The question here is to identify markers differentually expressed in >>> the monocytes subpopulation between normal patients and control >>> patients. >>> >>> marker normal normal treated treated >>> cdA -5.27 1.48 -1.28 -1.01 >>> cdB -5.31 -1.89 9.31 1.01 >>> cdC 4.12 8.1 8.16 3.6 >>> cdE 30.44 11.59 3.39 14.64 >>> CD11c 5.36 -1.48 -5.7 -4.44 >>> >>> >>> Do i hve to normalize the data first ? I though this was already >>> done by the instrument ? i might be wrong. Any idea ? >> I do not think you should normalize your data at that point. >> If any normalization would be required I think that it should be done >> on the raw data to correct for some bias identify by quality >> assessment analysis. >> >> Here you should have a look at the distribution of your data to >> choose between a parametric or non-parametric test (many you can find >> in R or BioC). >> >> Nolwenn >> >>> >>> >>> >>> Nolwenn Le Meur wrote: >>>> Hi David, >>>> >>>> I am not such I see what are the data you are manipulating. What is >>>> the experimental design and biological question(s)? What represent >>>> the fold change you want to compute? What are the rows and columns >>>> in your log10 data matrix? >>>> >>>> For data analysis of cell-based assay, you might also want to have >>>> a look at the cellHTS2 package. >>>> >>>> Best, >>>> Nolwenn >>>> >>>> David martin wrote: >>>>> Hi, >>>>> I've recently got some data from the lab coming from flow cytometry. >>>>> I have the log10 values corresponding to the geometrical mean (not >>>>> the flow cytometry files). >>>>> >>>>> Basically i would like to start from that matrix and compute the >>>>> fold changes. I'm not sure which test is most suitable as not sure >>>>> which sitribution the data follows , gaussian ??? could anybody >>>>> suggest how to move on from the log10 data matrix ? Any paper or >>>>> tutorial . >>>>> I have already looked at the flow packages within R but most of >>>>> them deal with gating and scaling the raw data, but not how to >>>>> compute fold changes >>>>> >>>>> thanks for any help, >>>>> david >>>>> >>>>> _______________________________________________ >>>>> 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 >>>> >>>> >>> >>> _______________________________________________ >>> 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 >> >> > > _______________________________________________ > 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 -- Nolwenn Le Meur, PhD IRSET - Institut de Recherche en Sante-Environnement-Travail. INSERM/IRISA - Equipes INTEREST/Symbiose Universite de Rennes I Campus de Beaulieu 35042 Rennes cedex - France Phone: +33 2 99 84 71 17 Fax: +33 2 99 84 71 71 E-mail: nlemeur at irisa.fr Website: http://www.irisa.fr/symbiose/nolwenn_le_meur