Entering edit mode
Sindre
▴
110
@sindre-6193
Last seen 3.7 years ago
Hi!
I have two groups at two time points. And the samples are the same in
both time points. I have run this in DESeq2:
sampleFiles <-
list.files(path="/Volumes/timemachine/HTseq_DEseq2",pattern="*.txt");
status <- factor(c(rep("Healthy",26), rep("Diabetic",22)))
timepoints =
as.factor(c(1,1,1,1,1,1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2,2,2,2,2,1,1,1,
1,1,1,1,1,1,1,1,2,2,2,2,2,2,2,2,2,2,2));
sampleTable <- data.frame(sampleName = sampleFiles, fileName =
sampleFiles, status=status, timepoints=timepoints);
directory <- c("/Volumes/timemachine/HTseq_DEseq2/");
des <- formula(~timepoints+status);
ddsHTSeq <- DESeqDataSetFromHTSeqCount(sampleTable = sampleTable,
directory = directory, design= des);
ddsHTSeq
This is however not correct.
When looking at the Limma manual page 49:
http://www.bioconductor.org/packages/2.12/bioc/vignettes/limma/inst/do
c/usersguide.pdf
This example is perfect for my experiment, but "Tissue = A or B"
should
be "Timepoint = 1 or 2".
So timepoint = 1 or 2 = Paired data, and Disease vs Normal = unpaired
data.
I want to compare both within and between samples, so how can I do
this
in DESeq2?
> sampleTable
sampleName fileName status timepoints
1 D104.txt D104.txt Healthy 1
2 D121.txt D121.txt Healthy 1
3 D153.txt D153.txt Healthy 1
4 D155.txt D155.txt Healthy 1
5 D161.txt D161.txt Healthy 1
6 D162.txt D162.txt Healthy 1
7 D167.txt D167.txt Healthy 1
8 D173.txt D173.txt Healthy 1
9 D176.txt D176.txt Healthy 1
10 D177.txt D177.txt Healthy 1
11 D179.txt D179.txt Healthy 1
12 D204.txt D204.txt Healthy 1
13 D221.txt D221.txt Healthy 1
14 D253.txt D253.txt Healthy 2
15 D255.txt D255.txt Healthy 2
16 D261.txt D261.txt Healthy 2
17 D262.txt D262.txt Healthy 2
18 D267.txt D267.txt Healthy 2
19 D273.txt D273.txt Healthy 2
20 D276.txt D276.txt Healthy 2
21 D277.txt D277.txt Healthy 2
22 D279.txt D279.txt Healthy 2
23 N101.txt N101.txt Healthy 2
24 N108.txt N108.txt Healthy 2
25 N113.txt N113.txt Healthy 2
26 N170.txt N170.txt Healthy 2
27 N171.txt N171.txt Diabetic 1
28 N172.txt N172.txt Diabetic 1
29 N175.txt N175.txt Diabetic 1
30 N181.txt N181.txt Diabetic 1
31 N182.txt N182.txt Diabetic 1
32 N183.txt N183.txt Diabetic 1
33 N186.txt N186.txt Diabetic 1
34 N187.txt N187.txt Diabetic 1
35 N188.txt N188.txt Diabetic 1
36 N201.txt N201.txt Diabetic 1
37 N208.txt N208.txt Diabetic 1
38 N213.txt N213.txt Diabetic 2
39 N270.txt N270.txt Diabetic 2
40 N271.txt N271.txt Diabetic 2
41 N272.txt N272.txt Diabetic 2
42 N275.txt N275.txt Diabetic 2
43 N281.txt N281.txt Diabetic 2
44 N282.txt N282.txt Diabetic 2
45 N283.txt N283.txt Diabetic 2
46 N286.txt N286.txt Diabetic 2
47 N287.txt N287.txt Diabetic 2
48 N288.txt N288.txt Diabetic 2
The commands used in Limma (still at page 49):
targets <-
readTargets("/Volumes/timemachine/HTseq_DEseq2/Targets.rtf");
Treat <- factor(paste(targets$status,targets$timepoints,sep="."));
design <- model.matrix(~0+Treat);
colnames(design) <- levels(Treat)
So how can I create the "Targets.rtf" file? And is these commands the
same when using DESeq2?
Thank you so much!
