Clean EVM gff is not written when no gene filtered out

[abretaud]

Are you using the latest release?
Yes

Describe the bug
When no gene gets filtered out after EVM, the clean EVM gff is not written, generating an error later.
The bug comes from this line I guess: https://github.com/nextgenusfs/funannotate/blob/master/funannotate/library.py#L6782
= the code should just write everything to the clean output when numTotal == 0

What command did you issue?
funannotate predict --input '/tmp/tmps7cf2xyi/files/000/dataset_1.dat' --out output --database '/funannotate_db' --species 'fvd' --isolate 'vfd' --strain 'bgf' --organism 'other' --ploidy 1 --SeqCenter 'CFMR' --SeqAccession '12345' --name 'FUN_' --numbering 1 --p2g_pident 80 --p2g_prefilter diamond --augustus_species 'fly' --min_training_models 200 --busco_seed_species 'fly' --busco_db 'insecta' --evm-partition-interval 1500 --min_intronlen 10 --max_intronlen 3000 --min_protlen 10 --keep_no_stops --repeat_filter overlap blast

Logfiles

[Jul 20 05:06 PM]: 18 total gene models from EVM
[Jul 20 05:06 PM]: Generating protein fasta files from 18 EVM models
[Jul 20 05:06 PM]: now filtering out bad gene models (< 10 aa in length, transposable elements, etc).
[Jul 20 05:06 PM]: 0 gene models remaining
[Jul 20 05:06 PM]: Predicting tRNAs
[Jul 20 05:06 PM]: CMD ERROR: tRNAscan-SE -o output/predict_misc/tRNAscan.out --thread 1 /tmp/tmps7cf2xyi/job_working_directory/000/4/working/output/predict_misc/genome.softmasked.fa
[Jul 20 05:06 PM]: tRNAscan-SE seems to have failed, check logfile for error. You can pass precalculated results to --trnascan
[Jul 20 05:06 PM]: Generating GenBank tbl annotation file
Traceback (most recent call last):
  File "/home/abretaud/miniconda3/envs/__funannotate@1.8.7/bin/funannotate", line 10, in <module>
    sys.exit(main())
  File "/home/abretaud/miniconda3/envs/__funannotate@1.8.7/lib/python3.7/site-packages/funannotate/funannotate.py", line 705, in main
    mod.main(arguments)
  File "/home/abretaud/miniconda3/envs/__funannotate@1.8.7/lib/python3.7/site-packages/funannotate/predict.py", line 1860, in main
    args.numbering, args.SeqCenter, args.SeqAccession, tbl_file)
  File "/home/abretaud/miniconda3/envs/__funannotate@1.8.7/lib/python3.7/site-packages/funannotate/library.py", line 3246, in GFF2tbl
    Genes = gff2dict(evm, fasta, Genes)
  File "/home/abretaud/miniconda3/envs/__funannotate@1.8.7/lib/python3.7/site-packages/funannotate/library.py", line 4230, in gff2dict
    with open(file, 'r') as input:
FileNotFoundError: [Errno 2] No such file or directory: 'output/predict_misc/evm.cleaned.gff3'

OS/Install Information
The latest package from conda...

[nextgenusfs]

Okay, I guess it should die gracefully if there are 0 gene models remaining..... as there is nothing to write in the GFF output.

[abretaud]

Hum, no, the problem is when all gene models remain (=0 removal)

[nextgenusfs]

[Jul 20 05:06 PM]: 18 total gene models from EVM
[Jul 20 05:06 PM]: Generating protein fasta files from 18 EVM models
[Jul 20 05:06 PM]: now filtering out bad gene models (< 10 aa in length, transposable elements, etc).
[Jul 20 05:06 PM]: 0 gene models remaining

Sorry I was on my phone and just read the above..... you are right, missing the else clause writing them all.

[abretaud]

Thanks for the quick fix! Tested locally, it works great