0
$\begingroup$

I am new to R and biocondunctor. I have the normalized expression values for 20 samples for a disease and for 10 controls. I wanted to get the differential expressed values for each sample with all 10 controls.

How could I do that with limma.

Any help will be appreciated.

Edit: What I want to achieve is single-sample t-test between the control population and the single sample values to get DE values

Improve this question
$\endgroup$
10
  • $\begingroup$ You need raw count data, and let limma normalize it (with e.g., voom). Read the manual, it is pretty well written. $\endgroup$
    – benn
    Commented Sep 11, 2019 at 9:16
  • $\begingroup$ @benn But I have the normalized data already, I want to get single expression values $\endgroup$
    – Vinay Bharadhwaj
    Commented Sep 11, 2019 at 9:18
  • $\begingroup$ Try to get raw counts. $\endgroup$
    – benn
    Commented Sep 11, 2019 at 9:33
  • $\begingroup$ @benn Ok , and then, I should normalize with voom, after which how should I get DE values for each sample separately? $\endgroup$
    – Vinay Bharadhwaj
    Commented Sep 11, 2019 at 9:34
  • $\begingroup$ Please go over the limma manual (bioconductor.org/packages/release/bioc/vignettes/limma/inst/doc/…) and edit your question with the potential problems you will encounter. Be careful to use the right section given the technology used (microarray or RNA-seq). $\endgroup$
    – haci
    Commented Sep 11, 2019 at 9:41

1 Answer 1

Reset to default
1
$\begingroup$

Treat your disease samples as individual groups and then follow the normal routine in limma to use contrasts to compare two group (the control group versus individual disease samples). Note that you need to think long and hard about what these results then mean. Normally we compare groups because then the results should generalize to other samples. That will not be the case when you look at individual disease samples, so I suggest you do some clustering of the samples to see if you perhaps have any coherent clustering of the disease samples into subsets that could then be used as more useful groups in DE testing.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thanks for the help! $\endgroup$
    – Vinay Bharadhwaj
    Commented Sep 12, 2019 at 10:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.