Help doing differential expression analysis -experimental design and gProfiler TF interpretation-

Question

[Camilo Andres]

Hi! I'm trying to do a differential expression analysis using breast cancer TCGA data.

Firstly, I split the breast cancer cohort into two groups based on the expression level in z-score of a particular gene I'm interested in, so the groups were the samples with z-score < -2 and all the others, giving two groups (n = 35 and n = 1000). Then, I used the rsem data to perform a differential expression analysis using DEseq2, but when do a PCA to compare the two groups using the 500 most differentially expressed genes, I don't get two clearly split groups.

Then, I used gprofiler2, and one of the significative results was "TF", and when I see the term name, it appears "Factor: LUMAN; motif: CYCAGCYYCY; match class: 1".

So, my questions are:

1. Should I better compare a group with z-score < -2 versus another group with z-score > 2?
2. What does it mean the result of gprofiler related to TF? I have tried to look for information but I think I don't understand it properly... I think that it says that genes regulated by this transcription factor are enriched, but I would like to have some confirmation.

Thanks in advance!

Answer 1

[Zhenyu Zhang]

Why z-score? It sounds like a weird setup to split cohort.

[rajabolacuan]

Good Information Sir