Importing eggnog annotations into topGO or similar?

Question

[rmash]

I'm working with a non-model organism. I took some genes of interest and mapped it to eggnog. I got an annotation file that has a column with sometimes multiple GO terms per row.

What is the best way to look at GO term enrichment?

I can import the relevant columns (ID) and column with GO terms into R.

But do I still need a reference database now that these are mapped?

What's the best way to look like GO term enrichment?

This is what happens when I try to import my custom annation into topGO, where one column are some IDs and the second column is of GOterms

geneID2GO <- readMappings(file = "fc_test.csv", sep="\t") 
Error in `[.data.frame`(a, , 2) : undefined columns selected

[Jean-Karim Heriche]

Without seeing the file content, one can only guess at what the problem is although most likely it is a data format issue. Check that your file conforms to what's expected by the readMappings function, i.e. columns are tab-separated and the first one contains gene IDs, the second one a comma-separated list of GO term IDs, e.g.:

068724  GO:0005488, GO:0003774, GO:0001539, GO:0006935, GO:0009288
119608  GO:0005634, GO:0030528, GO:0006355, GO:0045449, GO:0003677, GO:0007275

[rmash]

Thanks, you were right!

This is more of an issue with EggNOG but do you find that it doesn't assign GO terms to most of the orthologs?