Longitudinal analysis in Limma: Block=ID as random effect
1
0
Entering edit mode
@caf637f5
Last seen 11 days ago
United States

Hello, Its my first time to deal with Longitudinal analysis in Limma and I would like to make sure that my design reflects my research question. Research question: which gene-expressions are significantly differentially expressed over time with the change of FEV1 (continuous variable) overtime. I have 2 gene expression, FEV1, and other covariants readings for 2 timepoints.

RID<-factor(rownames(combined_pheno)) #reading ID is different for each reading 
ID<-factor(combined_pheno$SubjectID) # each subject has 1 ID and 2 RID

design<- model.matrix(~FEV1+Time+FEV1:Time+ Age+Gender+Race)
fit<-voomLmFit(exp , design, block=ID, plot=TRUE)
fit<- eBayes (fit)
result<- topTable(fit,coef="FEV1:Time",n=Inf,adjust="BH",p.value = 0.05 , sort.by = "logFC")
sessionInfo( )

My question would be is it right to assume that "ID" is counted as random effect by adding (block=ID) Thank you

limma • 193 views
ADD COMMENT
0
Entering edit mode

Please don't post two questions with exactly the same title. If you wish to ask a new question, then choose a new title that gives readers some idea what the question is about.

ADD REPLY
2
Entering edit mode
@gordon-smyth
Last seen 44 minutes ago
WEHI, Melbourne, Australia

Yes. The blocking variable in limma means a random effect.

I notice however that you are trying to adjust for patient differences by including patient characteristics (Age, Gender and Race) in the design matrix. It is a bit unusual to include patient as a random effect and adjust for patient differences at the same time. I guess it still might work ok if you have a lot of samples. It depends on your experimental design.

ADD COMMENT
0
Entering edit mode

Thank you for your suggestion

ADD REPLY

Login before adding your answer.

Traffic: 576 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