Dear Bioconductors,

I have a NanoString nCounter dataset consisting of paired data (conditions A and B) with some additional measurements for samples in condition B.

My current setting consists of 5 patients and 2 conditions:

df<-data.frame(pair=c("p1","p1","p2","p2","p2","p3","p3","p4","p4","p4","p5","p5"), condition=c("A","B","A","B","B","A","B","A","B","B","A","B"))

df

pair condition

p1 A

p1 B

p2 A

p2 B

p2 B

p3 A

p3 B

p4 A

p4 B

p4 B

p5 A

p5 B

I'd like to assess differential expression between conditions A and B by using the Bioconductor package <NanoStringDiff>.

So my definition of a design matrix would look like this

design<-model.matrix(~0+factor(condition)+factor(pair),data=df) colnames(design)<-c("A","B","p2","p3","p4","p5") design A B p2 p3 p4 p5 1 0 0 0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 1 1 0 0 0 0 1 1 0 0 0 1 0 0 1 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 1 0 0 1 0 0 1 0 0 1 0 1 0 0 0 0 1 0 1 0 0 0 1 attr(,"assign") [1] 1 1 2 2 2 2 attr(,"contrasts") attr(,"contrasts")$`factor(condition)` [1] "contr.treatment"

attr(,"contrasts")$`factor(pair)`

[1] "contr.treatment"

In addition, I defined the contrast to be

contrast<-c(1,-1,0,0,0,0) # should encode for A-B

Are the definitions of the design matrix and the contrast correct?

I have this workflow in my mind (following the <NanoStringDiff> vignette) :

# [1] create NanoStringSet NanoStringData=createNanoStringSetFromCsv(pathToMyData,header=TRUE,design) # [2] estimate normalisation parameters NanoStringData1=estNormalizationFactors(NanoStringData1) # [3] run the Generalize linear model likelihood ratio test result=glm.LRT(NanoStringData1,design,contrast=contrast)

So does this workflow indeed answer my question which features differ significantly between the two conditions A and B (considering the paired nature of the data)?

I feel like I made a mistake, but I don't know for sure. Hence, I'd like to check with you, whether it is correct or not.

As the Lab people will do their experiments based on my resulting gene list, I want to be sure whether my definitions of the design matrix and the contrast are correct.

Thank you very much for any help / suggestions

Christian