I am a bit frustrated right now. I wanted to double check my DESEQ2 Differential Expression (RNA SEQ) results in Limma and now I am getting different results. I dont know what I did differently that the results are so different. I am here interested in the difference in Gender: Male vs. Female. Please find below the relevant part of the two different approaches, if you can see any obvious difference I would be very happy:
dds0 <- DESeqDataSetFromMatrix(countData = cts, colData = colData, design = ~ batch + AGE.GROUP + Sample.Site + BMI.Tertile + Diagnosis + Gender)
# Second approach for filtering ### Set thresholds padj.cutoff <- 0.05 lfc.cutoff <- 1.2
design = model.matrix( ~ batch + AGE.GROUP + Sample.Site + BMI.Tertile + Diagnosis + Gender, metadata) fitDupCor <- lmFit(geneExpr, design) # Fit Empirical Bayes for moderated t-statistics fitDupCor <- eBayes( fitDupCor )
result <- topTable(fitDupCor, number = 50, adjust = "BH", p.value = 0.05, lfc = 1.2, coef = "GenderM")
I used raw counts for both analysis, maybe that was a mistake?
Thank you all!!!