Entering edit mode
3.4 years ago
biosjm
•
0
Hi,
I`ve run topGo for go enrichment analysis (fisher,weight) and obtained a list with go terms and assigned genes. e.g. GO:0006793 genes: a,b,c,d I had a look to my functional annotation list to one of the genes (e.g. gene b), and recognized, that the go term GO:0006793 was not annotated to this gene b.
Can this be right?
If you have doubts, I would recommend using additional tools to verify the results.
Have a look at this tool, Gene Set Clustering based on Functional annotation (GeneSCF)
Ok. I have a GO annotation dataframe (gene id with GO numbers) from trinotate, so it`s not a model organism. If I´m right, GeneSCF works mainly with known species right?
Here is how you can use your custom annotation file with GeneSCF v1.0,
A: GO enrichment analysis using a Text file with all the genes and GO ids associat
Okay, thanks. I will take a look at it and check it again. But just to clarify my question: I realized that genes were assigned to the Go-Terms (results from the Fischer test), but on the other hand some were not assigned to these genes in my annotation table (trinotate pipeline). When applying the test/algorithm, is it possible that genes were "reassigned" to other GO-terms due to the hierarchical structure?
You can give a look at this post in which I encountered a similar issue: topGO is performing enrichment test on GO terms that are not present in my dataset
However, the short answer is: yes, it can be right. topGO can add terms due to hierarchical structure. I think it is mostly ok, but maybe is not ideal, and that's why I sometimes use a fisher test on my own to perform enrichment tests. You could also look for additional packages, I am sure there are several out, but I didn't investigated the issue recently.
Thank you, Fabio. This is helpful! There are some packages, but it`s hard to find packages that work with custom annotation table.