Dear Sir/Madam. I seek your advice about some microarray data. I have a time serie of 6, from 0, 45m, 90m, 3h 6h and 12h. Each point has 4 replicates. I want to find diffeentially expressed genes and I am using the package timecourse. But I am not sure about the results. I run function mb.long, and get a result with "MB" and "HotellingT2" statistic data. I read the package vignette and also try read "Multivariate empirical Bayes models for replicated microarray time course data." www.stat.berkeley.edu/tech-reports/667.pdf) but I am afraid I do not completely understand the statistics. Or how to use them. All my sample sizes are the same, so in ?mb.long I read that I should use "HotellingT2". But these values go from 0 to 357.4 in my data and in the example in ?mb.long: > data(fruitfly) > gnames <- rownames(fruitfly) > assay <- rep(c("A", "B", "C"), each = 12) > time.grp <- rep(c(1:12), 3) > size <- rep(3, nrow(fruitfly)) > out1 <- mb.long(fruitfly, times=12, reps=size, rep.grp = assay, time.grp = time.grp) > summary(out1$HotellingT2)      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 3.679e-01 5.559e+01 3.584e+02 1.022e+03 1.063e+03 2.863e+04 In the article the authors write values are ranked by "HotellingT2" but how to know if they are significant? How does "HotellingT2" fit with p-values?? Instead, should I use "MB" which I think is similar to B-values from package limma? >  out2 <- mb.long(fruitfly, times=12, reps=size, rep.grp = assay, time.grp = time.grp, HotellingT2.only=FALSE) > summary(out2$MB)     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. -17.6200 -12.8800  -8.2200  -8.7910  -5.4430   0.1485 Please advice. Kindest regards Helen [[alternative HTML version deleted]]