how remove large gene ontology terms from gprofiler2 output

Question

[BISEP]

Hello I have used 400 genes to conduct The GO analysis using gprofiler2 output I am getting some 150 Go terms as a output how can I reduce this GO terms. I know I can use REVIGO is there any other tool i can use because have another gene set it gave me 500 GO terms, REVIGO can take only handle 350 GO so please help, any alternative solution also helpful Thank you

Answer 1

[jared.andrews07]

You could limit based on the term_size, filtering terms from the results that have a ton of associated genes and tend to be less interesting. I've found limiting to terms with <1000 members to be helpful at times.

[BISEP]

Yes, I have tried using term_size I am little bit confused because I have 5 different gene set with the different gene number if I put same term size in some cases i am not able get any significant GO term, what is the minimum term size I can choose because some logic should be there right to use this term size. so what I did I took all the significant GO terms and used REVIGO to remove the redundant GO terms. Actually I did not understand what is this term size and how it works. Can you please help me to figure it out.

[jared.andrews07]

There are aspects of this question you're going to have to answer yourself. Your gene set doesn't affect term size. Term size refers to the number of genes associated with a given GO term (or pathway, etc). More broad terms (like "cytoplasm") have lots of genes associated, sometimes to the point where they aren't really interesting because there are 5000 genes associated with them. As for what the limit should be, that's really up to you. Like I said, I usually find 1-2k a decent limit, but it really depends on your specific study design and goals. Removing or collapsing redundant terms is also perfectly valid.