Hi Jim, Thank you for your response. It's Ok right know. thanks a lot Best, Cherif 2014-04-07 15:01 GMT+01:00 James W. MacDonald <jmacdon@uw.edu>: > Hi Cherif, > > It would be a good idea to look at the data first. I simply googled your > GSE number and clicked on one of the samples, and it says this: > > *ID_REF* > *VALUE* Signal intensity value, computed by MAS5. Scaled to 500 > *ABS_CALL* Detection call > *DETECTION P-VALUE* Detection P-value > > > And if you look at your data you will notice that everything is strictly > positive, and the maximum is somewhere near 1e14 or so. These are both > indications that your data have not been logged, which is necessary for the > analysis of these data. > > Best, > > Jim > > > > On 4/6/2014 12:48 PM, BEN HAMDA Cherif wrote: > >> Hello >> Hope that everything is good for you ! >> >> i have a strange results in my work ( series_matrix.txt microarray >> analysis ) >> and i don't know if it's normal or not ! >> >> i get high value (logFC) >> >> logFC AveExpr t P.Value adj.P.Val >> B >> 14990 *184.8393 *279.5879 7.053297 6.558774e-08 0.001457032 -3.553707 >> 11016 *-113.2735* 132.3939 -6.777861 1.404067e-07 0.001559568 -3.588662 >> 583 *565.7128 * 942.2667 -6.602527 2.288115e-07 0.001694349 >> -3.611925 >> 21147 -*103.8393* 411.5121 -6.436832 3.639169e-07 0.002021103 -3.634652 >> 2067 -*173.5342* 357.1273 -6.261612 5.958922e-07 0.002647549 -3.659485 >> 5404 -*427.6940 * 464.8182 -6.131110 8.616234e-07 0.002968063 -3.678522 >> >> >> i show my method : >> >> gse <- getGEO(filename='GSE1849-GPL96_series_matrix.txt.gz') >> >> i make the design and after that i make analysis with limma : >> >> library(limma) >> fit <- lmFit(exprs(gse), design) >> contrast.matrix <- makeContrasts(Normal_ACS = Normal - ACS , >> levels=design) >> contrast.matrix >> Ctrl_fits <- contrasts.fit(fit, contrast.matrix) >> Ctrl_ebFit <- eBayes(Ctrl_fits) >> nrow(topTable(Ctrl_ebFit, adjust = "BH", coef = 1, number = Inf) >> >> So it's normal or not ? i made mistake in my R code ? someone can help me >> ? >> >> Thank you so much >> >> Best, >> cherif >> >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor@r-project.org >> 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 > University of Washington > Environmental and Occupational Health Sciences > 4225 Roosevelt Way NE, # 100 > Seattle WA 98105-6099 > > [[alternative HTML version deleted]]

Hi Gordon, Thank you for your response. It's Ok right know Thank you very much Best, Cherif [[alternative HTML version deleted]]