Search
Question: DESeq2 - how to build design for matched case-control study with repeated measurement
0
gravatar for 183227851
3 months ago by
1832278510
1832278510 wrote:

Please help me build a design that accounts for both the matched case-control design (there were 1 case [disease=1] and 2 controls [disease=0] in each pair [having the same matchid], matched for age and gender) and the correlation between multiple measurements at different timepoints (time 1, 2, 3) in a given subject. My objective is to compare the microbial differential abundance between cases and controls over all timepoints (not treating each timepoint separately). Thanks a lot!

My dataset is as following:

personid matchid time disease
1 1 1 1
1 1 2 1
1 1 3 1
2 1 1 0
2 1 2 0
2 1 3 0
3 1 1 0
3 1 2 0
3 1 3 0
4 2 1 1
4 2 2 1
4 2 3 1
5 2 1 0
5 2 2 0
5 2 3 0
6 2 1 0
6 2 2 0
6 2 3 0
7 3 1 1
7 3 2 1
7 3 3 1
8 3 1 0
8 3 2 0
8 3 3 0
9 3 1 0
9 3 2 0
9 3 3 0

 

ADD COMMENTlink modified 3 months ago by Michael Love19k • written 3 months ago by 1832278510
0
gravatar for Michael Love
3 months ago by
Michael Love19k
United States
Michael Love19k wrote:

With fixed effects, we can't include the matching variable and also a variable to distinguish the two individuals within the control, because these are confounded variables. In general, you can instead use limma with the duplicateCorrelation() correlation function when fixed effects cannot be used because of this kind of nesting.

ADD COMMENTlink modified 3 months ago • written 3 months ago by Michael Love19k

Many thanks for your professional and quick reply. If I want to stick to DESeq2 and I only want to account for correlation between multiple measurements at different timepoints (time 1, 2, 3) in a given subject(ignoring the matching for the moment),how should I build the design.My objective is to compare the microbial differential abundance between cases and controls over all timepoints (not treating each timepoint separately).Thanks a lot! 

ADD REPLYlink modified 3 months ago • written 3 months ago by 1832278510

You would use the strategy in the vignette for testing for condition effects within and across groups while controlling for individuals nested within groups.

ADD REPLYlink modified 3 months ago • written 3 months ago by Michael Love19k

Thanks Michael! Compared to limma, I am more familiar with edgeR. EdgeR can also compare the microbial differential abundance between cases and controls. Additionally, there are a lot of examples in the edgeRUsersGuide. If I use edgeR, is the design correct?

design <- model.matrix(~matchid+personid+disease+time+disease:time)

(There are paired examples and repeated measurement examples in the edgeRUsersGuide. However,in repeated measurement examples,repeated measurement at different timepoints were obtained from independent subjects. By contrast,in my study,repeated measurement at different timepoints were obtained from the same subjects.)   

ADD REPLYlink written 12 weeks ago by 1832278510

Please make a new post and tag the new packages you are asking about. I’m only aware of duplicateCorrelation in the limma package.

ADD REPLYlink written 12 weeks ago by Michael Love19k

OK! Thanks!

ADD REPLYlink written 12 weeks ago by 1832278510
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 2.2.0
Traffic: 127 users visited in the last hour