Regression Analysis between RNA-seq data and expression measured from ELISA

Question

[halo22]

Hello All,

I am trying to run a regression analysis on RNA-seq expected counts generated using RSEM. To perform the analysis I performed TMM normalization and extracted CPM values. I then used voomWithQualityWeights, lmFit and eBayes to perform association analysis. Since RNA has a different data distribution than the ELISA values, I would like to know if the steps that I followed are correct. I appreciate all your help and suggestions.

Following is my code:

 d <- DGEList(counts=data, group=groups)
 TMM <- calcNormFactors(d, method="TMM")
  logCPM <- cpm(TMM, log=TRUE)

design <- model.matrix(~LN_AB, data=Pheno)
voom <- voomWithQualityWeights(data, design=design, lib.size=NULL, normalize.method="none")
fit1 <- lmFit(normalizeddata,design)
fit2 <- eBayes(fit1)
newFrame <- topTableF(fit2)
print(newFrame)

[Kevin Blighe]

Cross-posted: https://support.bioconductor.org/p/133153/