This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Deseq2 Result

I use Deseq2 to perform differential gene expression analysis. I have successively changed the LFC, i.e.LFC > 1.5 and LFC > 2 as shown in the following code:

dds_ALL <- DESeq(dds_ALL)
GNF_D6_vs_GNF_WT =results(dds_ALL,contrast = c("genotype","TreatedD6","TreatedWT"),lfcThreshold = 1,alpha = 0.05)
summary(GNF_D6_vs_GNF_WT,rm.na = TRUE)
head(GNF_D6_vs_GNF_WT)

 treshlod <- GNF_D6_vs_GNF_WT$padj < 0.05 & abs(GNF_D6_vs_GNF_WT$log2FoldChange) >1

 GNF_D6_vs_GNF_WT$treshlod<-treshlod
sig_D6WT <- data.frame(subset(GNF_D6_vs_GNF_WT,treshlod==TRUE))
write.csv(sig_D6WT , 
          file="TreatedD6_vs_TreatedWT.csv")

 write.csv(rownames(sig_D6WT) , 
          file="TreatedD6_vs_TreatedWT_ID.csv",row.names = FALSE)

When I keep the LFC > 1.5, all the groups of genes that have for example a LFC = 3, I expect to see the same genes when I put the LFC to 2 and with the same value of LFC, But however I notice a slight variation of the LFC of these genes instead of 3 they come back to 2.8. So I would like to know if this variation is due to the Deseq2 algorithm or due to my code?

Thank you

transcriptomics rna-seq deseq2
• 1,659 views • link
updated by Ram Ram • written by Diango • Diango

Take a look at the answer of Michael Love respect to this point in Bioconductor. Also, I suggest you to read the manuscript mentioned there in order to understand what DESeq2 is doing when estimates the lfc.

• 0 views • link
written by rodolfo.peacewalker • rodolfo.peacewalker

Thank you rodolfo.peacewalker.

• 0 views • link
written by Diango • Diango

0 answers

No answers yet.

Log in to answer this question.

More posts like this