Comparing gene expression between two groups, using boxplots

0

Entering edit mode

2.2 years ago

JACKY ▴ 160

I have several cancer datasets (RNA-seq), with rows as genes and samples as columns. Each sample in those data sets is labeled either Response (people who responded to the therapy), and NoResponse (people who did not respond).

For each dataset, I conducted a differential expression analysis with limma. In order to assess the results, I picked some genes and compared their expression between the Response and NoResponse groups in each dataset, using BoxPlots.

My question is, should I use TPM normalized data for the comparison or the original counts data? I figured that the normalization won't make much difference in this situation, because the comparison is between samples regarding the same gene. So.. should I take the expression of the genes that is in the counts data, or the TPM data?

This is the boxplots using `facet_wrap`, to compare the expression of CD274 in each dataset, between Response and NoResponse. Here I used counts data.

enter image description here

r DEG TPM boxplot • 665 views

ADD COMMENT • link 2.2 years ago by JACKY ▴ 160

0

Entering edit mode

Please, any help would be appreciated. TPM or counts data for comparison between samples?

ADD REPLY • link 2.2 years ago by JACKY ▴ 160

Login before adding your answer.

This is the boxplots using facet_wrap, to compare the expression of CD274 in each dataset, between Response and NoResponse. Here I used counts data.

This is the boxplots using `facet_wrap`, to compare the expression of CD274 in each dataset, between Response and NoResponse. Here I used counts data.