goseq for Top over−represented & under-represented GO terms plot

, ,
1

Hi there! I’ve performed goseq for Top over−represented categories. And It’s fine according to “Ranked category list - Wallenius method” column 8 - “p_adjust_over_represented”. But I has signifficantly changed categories in column 9 (“p_adjust_under_represented”) which are not plotted (no option).
My questions:

  1. how to get plot of Top under-represented GO terms?
  2. “over_represented” means upregulated and “under-represented” - downregulated?
2

Hi @olha_strilbytska

We have several tutorials with examples for goseq. Find these on the tool form – scroll down to the Help section to find the links. I am including the GOEnrichment tool, too.

Start with the last tutorial in the list, and you will find the answers to your questions with more nuance than I can reproduce here. In short: goseq is just one tool, and there are others you can try.

Up and Down regulation isn’t straightforward because these data can have biases. These tools are helping to normalize the expression results in the presence of those biases, and each does that a bit differently.

Example: goseq considers transcript length. That’s why you are supplying the length data at runtime – the primary up/down input file does not include that data point yet.

Hope this helps :slight_smile:

3

Hi @jennaj

Thank you for these recommendations! I would hope in goseq there will appear an option to visualise underrepresented terms in the feature