Question

What to do when two groups with untreated baseline show differentially expressed genes/taxa

0

Entering edit mode

Puks ▴ 10

@puks-12113

Last seen 4.5 years ago

Estonia

Hi, I am working on 16s rRNA data with two groups, one control and one treated measured over 8 timepoints. We collected a sample from both groups before giving any treatment to just check if the two groups were similar at start. Unfortunately when I look at deseq2 results for this timepoint, I see some taxa being differentially expressed with reasonable differences in their mean abundances.

Is there a way to correct for this difference ?

Thanks! Hena

deseq2 • 499 views

ADD COMMENT • link updated 4.5 years ago by Michael Love 43k • written 4.5 years ago by Puks ▴ 10

score 0 · Answer 1 · 2020-06-16

0

Entering edit mode

Michael Love 43k

@mikelove

Last seen 4 days ago

United States

As I've said previously on the support site, I'm not convinced DESeq2 is appropriate for microbiome or metagenomics data.

That said, you are describing an example from the workflow (time course, last example). Take a look how the design is set up.

ADD COMMENT • link 4.5 years ago Michael Love 43k

0

Entering edit mode

Thank you Michael for you reply! I agree to your comment about microbiome data and deseq2. I have been using corncob along with deseq2. I think we all end up using it because of the waste not want not paper (https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003531). Sorry I didn't search for microbiome data and deseq2

Do you have a suggestion in general on how best tackle this issue of difference in baseline? Is there some kind of normalization that could be done to the data at the beginning of the analysis like we do for batch correction etc.

Thanks Hena

ADD REPLY • link 4.5 years ago Puks ▴ 10

0

Entering edit mode

Yes, see the workflow. You add a term to the design.

ADD REPLY • link 4.5 years ago Michael Love 43k