IanDMedeiros

`funannotate test -t predict --cpus 10` The test completes, but still only finds 175 BUSCO loci > ######################################################### > Running `funannotate predict` unit testing > Downloading: https://osf.io/te2pf/download?version=1 Bytes: 1489808 >...

I am working to set up an alternative version of Augustus now. The error I am seeing with my existing Augustus 3.3.3 compiled without bioconda seems to be in how...

I still have no idea what is the problem with my Augustus 3.3.3, but I have solved the immediate issue by modifying `funannotate-BUSCO2.py`. I inserted two lines into the function...

Wondering if `--numbering` defaults to nothing rather than 1 if it isn't explicitly called, and so all predicted proteins got an identical locus tag and overwrote one another? Testing that...

Ok. I have verified that `funannotate predict` behaves correctly if I leave off `--name` and `--numbering` entirely, but if I use either` --name $locus_tag` alone or `--name $locus_tag --numbering 1`,...

Weird. I am working around it now but I will revisit at some point and try to figure out what is going wrong on my end.

This turned out to be an error in how I was providing the locus tag prefix to funannotate; because of a poorly constructed table file, there was a hidden newline...

@pbuttigieg @cmungall I think I have figured out how to do this myself with Protege and git fork-pull... could I get a range of IDs to work with?

Although now I am also confused, because while there is no "lichen material" in https://www.ebi.ac.uk/ols/ontologies/envo, I can see "lichen material" as a subclass of organic material when I explore envo-edit.owl...

Another alternative would be to not have any "lichen material" class but to to use the existing class "algal material" and a new class "fungal material" together in any context...