Unable to combine f2a results with funannotate2 annotate

[Elep798]

Hello!

Thank you for the great tool!

Sorry if it's the wrong place, I didn't know whether to post here or in funannotate2-addons.

I am trying to combine results from the funannotate addons (eggnog and antiSMASH) with funannotate2 main results, however I get an error with the parse_annotations function.

I simply ran funannotate2 annotate -i Sc_res/ -s "Saccharomyces cerevisiae" and got the following traceback:

Traceback (most recent call last):
File ".local/bin/funannotate2", line 8, in
sys.exit(main())
^^^^^^
File "/.local/lib/python3.12/site-packages/funannotate2/main.py", line 30, in main
annotate(args)
File "/.local/lib/python3.12/site-packages/funannotate2/annotate.py", line 283, in annotate
a = parse_annotations(annotfile)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/.local/lib/python3.12/site-packages/funannotate2/search.py", line 599, in parse_annotations
gene, db, value = line.split("\t")
^^^^^^^^^^^^^^^
ValueError: not enough values to unpack (expected 3, got 1)

I checked the logs for emapper and antismash and they both finished with no problems, generating 3-column tables with file name ending in .annotations.txt.

On the other hand, when I tried running it with the all.annotations.txt file generated with funannotate1 it worked well. As I'd prefer to run the whole thing with the same version, how should I solve this issue in a better way?

Thanks!

[nextgenusfs]

What files are in the Sc_res/annotate_misc folder? check if there is an empty file or something, one of them must be mis-formatted. The f2a antismash is probably to blame, it's not working yet I don't think.

[Elep798]

Apparently, f2a antismash ran and the antismash file is not empty. The beginning formatting looks ok:

(Funannotate2) $ head -n 5 Sc_results/annotate_misc/antismash.annotations.txt
#gene_id        annotation_type annotation_value
Sc_002479   note    antiSMASH: cluster_1 [NRPS-like]
Sc_002480   note    antiSMASH: cluster_1 [NRPS-like]
Sc_002481   note    antiSMASH: cluster_1 [NRPS-like]
Sc_002482   note    antiSMASH: cluster_1 [NRPS-like]

but the formating at the end seems a bit different, probably due to the extra info.

(Funannotate2) $ tail Sc_results/annotate_misc/antismash.annotations.txt
Sc_002581   SMCOG   SMCOG1002: AMP-dependent synthetase and ligase (Score:
304.6; E-value: 1.7e-92)
Sc_002581   sec_met_domain  AMP-binding (E-value: 9e-88, bitscore:
289.0, seeds: 400, tool: rule-based-clusters)
Sc_002581   NRPS_PKS_domain Domain: AMP-binding (18-439). E-value: 9.4e-77.
Score: 250.1. Matches aSDomain:
nrpspksdomains_Sc3830_002581_AMP-binding.1
Sc_003887   biosynthetic_domain     rule-based-clusters: phytoene_synt
Sc_003887   sec_met_domain  phytoene_synt (E-value: 3.2e-10, bitscore:
33.7, seeds: 8, tool: rule-based-clusters)

For the emapper file, I don't see formating differences between head and tail, but I have the same error regardless.

And weirdly enough, when testing funannotate2 anotate with the all.annotations.txt from funannotate1 on different strains, there were still 2 out of 6 strains for which it returned the same not enough values to unpack (expected 3, got 2) error. I ran the different annotations steps for the 6 strains with SLURM arrays, so they should have been generated the same way.

I tried looking at the file formatting too, but I couldn't see any obvious difference and when running the funannotate annotate function from f1, it "worked well" for all strains until the tbl2asn step.

[nextgenusfs]

yeah, sorry I just haven't had time to spend on the f2a tools -- they should be "use at your own risk" until a release is tagged.

[Elep798]

No worries at all, I totally understand. Thank you for the quick reply and your work on the tool!

[nextgenusfs]

I just pushed some changes to the f2a repo, if you install the latest and then try again to run f2a hopefully it will integrate the functional annotation properly.

[Elep798]

Hello, sorry for the delayed answer. Thank you for the update, it indeed resolved the problem of combining all the f2a results.