Question: Technical replicates using Limma (handling within-subject and between-subject factors)
1
gravatar for sh.kazempour94
8 weeks ago by
sh.kazempour9410 wrote:

Hello all, I've read Limma user's guide completely but I'm still unsure about the way I am making my design matrix. I have two groups (pre and post) for each subject. Two treatments (treatment and placebo) and sex. I want to find out if there are any differentially expressed genes between post-treatment and pre-treatment. The targets looks like this:

Sample_Name Sample_Group Sample_Source Treatment Sex
        003         post           A01   Placebo   M
       21v7          pre           B01 Rapamycin   M
       42v1          pre           C01 Rapamycin   M
      030v7          pre           F01 Rapamycin   F
       46v1          pre           G01 Rapamycin   M
       14v7          pre           H01   Placebo   M
        003          pre           A01   Placebo   M
       42v4         post           C01 Rapamycin   M
      030v9         post           F01 Rapamycin   F
       46v4         post           G01 Rapamycin   M
       14v9         post           H01   Placebo   M

I know that Sample_Group is my within-individual factor and Treatment is between-individual factor, but still do not get answers. Here is my code. I would really appreciate if anyone can help me. I am so confused.

groups <- factor($targets$Sample_Group) ## pre & post
treatment  <- factor(targets$Treatment) ## Rapamycin & placebo
individual <- factor(targets$Sample_Source)
gender <- factor(targets$Sex)
design1 <- model.matrix(~0+groups*treatment, data=targets)
dupfit <- duplicateCorrelation(eset, design1,
                               block=targets$Sample_Source)
fit <- lmFit(eset, design1,correlation=dupfit$consensus, block=individual)
fit2 <- eBayes(fit)

Thanks a lot.

Shiva.

limma technical replicate • 109 views
ADD COMMENTlink modified 8 weeks ago by Gordon Smyth39k • written 8 weeks ago by sh.kazempour9410

You have a pre entry for subject B01 but no post for the same subject. Please confirm that is correct.

What you mean when you say you don't "get answers"? How did you try to extract results from the analysis and how did they differ from what you were expecting?

ADD REPLYlink modified 8 weeks ago • written 8 weeks ago by Gordon Smyth39k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 332 users visited in the last hour