Entering edit mode
karen
▴
30
@karen-957
Last seen 10.2 years ago
Hello,
I have an Affymetrix time series experiment, and am following the
estrogen
example of the limma User's Guide to get started. I am feeling rather
unconfident,
and would like to know whether I am performing the best analysis.
If someone has time to take a look and offer advice, I would be very
appreciative.
The study has 7 time points, 2 genotypes, 2 replicates of each (28
arrays).
Of interest are genes that are differentially expressed
between genotypes across all time states but the first (time0).
I have run the data through rma (worked beautifully!)
and set up the design with 14 treatments:
genotype1_time1, genotype1_time2....
genotype2_time1, genotype2_time2....
And set up 6 contrasts:
genotype1_time2 - genotype2_time2
genotype1_time3 - gentotype2_time3 ...
My questions follow the code below.
The code:
treatment.vector<-c(rep(1,2),rep(2,2),rep(3,2), rep(4,2), rep(5,2),
rep(6,2),
rep(7,2),
rep(8,2),rep(9,2),rep(10,2),rep(11,2),rep(12,2),rep(13,2),rep(14,2))
treatments<-factor(treatment.vector,labels=exp.labels)
design<-model.matrix(~-1+treatments)
fit <- lmFit(Mrma, design)
contrast.matrix<-makeContrasts(KO.15min-WT.15min,
KO.30min-WT.30min,
KO.90min-WT.90min,
KO.3hr-WT.3hr,
KO.6hr-WT.6hr,
KO.24hr-WT.24hr,levels=design)
fit2 <- contrasts.fit(fit, contrast.matrix) fit3 <- eBayes(fit2)
clas <- classifyTestsF(fit3,fstat.only=FALSE) FStats<-FStat(fit3)
G1<-topTable(fit3,coef=1,number=1000,adjust="fdr",sort.by="P") ##15min
G1<-topTable(fit3,coef=2,number=1000,adjust="fdr",sort.by="P") ##30min
The Questions:
1) Is this an ok model to use, or is it better to include the
intercept?
2) How would I report either linear model via an equation?
3) Should one perform a multiple testing correction on the FStat
p-values?
4) Should the genes supplied by the signficant FStats(fit3) have all
non-zero
values in the 6 contrasts? (they do not)
5) If I would like to look at all genes that are differentially
expressed
between time points 2 and 3 across genotypes, is it best to simply
extract these from the significant FStats group that have nonzero
values
in contrasts 1 and 2, or choose the intersection between G1 and G2?
argh - I have so much to learn.
Thanks, folks, for helping.
Karen Schlauch