[BioC] paired t-test

[James W. MacDonald]

Hi Yogi,

Yogi Sundaravadanam wrote:
> Hi all,
> 
> I have six samples with the following design
> ID    Sample    Type    C1    A    Control    C2    A    Treated
> C3    B    Control    C4    B    Treated
> C5    C    Control
> C6    C    Treated
> 
> A, B, and C are three patients
> 
> 
> Would a paired t-test work better than one-way ANOVA here? Do I need 
> more replicates?

You definitely want to account for the intra-patient dependence 
structure somehow. One way would be to do a paired t-test. The other way 
would be to do a linear mixed model-ish thing using 
duplicateCorrelation() to capture the intra-patient correlation.

There have been numerous posts about the latter, and if I am not 
mistaken there is an example in the Limma User's Guide as well.

Best,

Jim


> 
> thanks for your help
> Yogi
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: 
> http://news.gmane.org/gmane.science.biology.informatics.conductor

-- 
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826