Comparing gene expression of specific genes between samples in DESeq2 / R

Question

[masicioglu]

Hi folks,

Using DESeq2, I've obtained log2 fold change (used to compare between two conditions), plot, heatmap ext. In addition to, I have two questions:

First, I've learnt TPM normalisation is suitable for compare gene expression level in the same sample but DESeq2 doesn't provide TPM normalisation code. How can I do TPM normalization in one sample for all interested genes from un-normalizate counts in R?

Secondly, I am wondering how can I compare one gene (for example p53) between a couple of samples(or two sample for example liver vs spleen)? Actually, heatmap gives this info but I want to graph comparing samples for one gene. Is it possible?

Could you help me, please?

Thanks

[Kevin Blighe]

I've learnt TPM normalisation is suitable for compare gene expression level in the same sample

Can you quote your source for that?

Secondly, I am wondering how can I compare one gene (for example p53) between a couple of samples(or two sample for example liver vs spleen)? Actually, heatmap gives this info but I want to graph comparing samples for one gene. Is it possible?

You could take the normalised counts ( counts(dds, normalized=TRUE) ) and plot them as barplots (as in A: Network/Pathway Analysis from Mass Spec data ):

You could also take the regularised log counts and plot these as boxplots with scatter-plot overlays (as in A: Boxplot in ggplot2 ):