Can the DESeq2 object for paired data be set with design= ~ batch + condition?
1
0
Entering edit mode
10 months ago
bioinfo ▴ 150

Hello,

I was given the output of DESeq2 and the DESeq2 script for some paired data and I noticed that in the script the design was set as design= ~ batch + condition instead of design= ~ subject+ condition. Is that considered the same thing or does DESeq2 perform a different correction when batch is set? I was wondering if I should rerun my data. In my data S1-S3 are replicates of control and S4-S6 are replicates of treated. S1 is paired with S4, S2 with S5 and S3 with S6. Everything was prepared at the same time so I do not have batches. I was wondering if changing the batch to subject will change the results. The table would still be the same except for that. Should i rerun it and change the batch to subject or is it the same thing?

The table was set like this:

enter image description here

rna-seq deseq2 • 1.0k views
ADD COMMENT
0
Entering edit mode

Batch, subject, gender, drug, these are all just titles. DESeq2 does not know about it's meaning. If batch and subject have the same factorial values then the correction is the same, else it is different. Please show colData for a precise answer.

ADD REPLY
0
Entering edit mode

Thank you for your reply. I updated my question to make things a bit more clear.

ADD REPLY
0
Entering edit mode
10 months ago
bk11 ★ 2.8k

Do you see any batch effect in your samples? If so, design= ~ batch + condition seems logical to me. Why would you have design= ~ subject+ condition? But if you think each sample is the batch then you could run `DESeq2 with second design.

ADD COMMENT
0
Entering edit mode

I have updated my question to make things a bit more clear. I do not have batches in my samples. They were all run the same time but I have paired samples.

ADD REPLY
0
Entering edit mode

Then design= ~ batch + condition will be a good design for your experiment.

ADD REPLY
0
Entering edit mode

Thank you. I was wondering if it matters if the design is set up lie this: design= ~ batch + condition if on the sample table the first column is the condition and the second is the batch.

ADD REPLY
0
Entering edit mode

It should not matter with the column switch as far as I know.

ADD REPLY
0
Entering edit mode

Thank you so much!

ADD REPLY

Login before adding your answer.

Traffic: 1744 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6