create.py unable to parse prodigal GFF

[christinehe]

Hi,

I'm using Phava to look for invertons in a de novo metagenomic assembly. I'm feeding Phava a GFF output by prodigal v2.6.3, but get a parsing error in create.py, which expects locus_tag and transl_table attributes in column 9. I believe locus_tag is required for NCBI genomes, but is not output by prodigal. Below are the first few lines of my GFF file:

##gff-version  3
# Sequence Data: seqnum=1;seqlen=13410;seqhdr="ctg5672064 length=13410 coverage=3.88 circular=no"
# Model Data: version=Prodigal.v2.6.3;run_type=Metagenomic;model="36|Ralstonia_solanacearum_PSI07|B|66.1|11|1";gc_cont=66.10;transl_table=11;uses_sd=1
ctg5672064      Prodigal_v2.6.3 CDS     2       1351    222.9   -       0       ID=1_1;partial=10;start_type=ATG;rbs_motif=None;rbs_spacer=None;gc_cont=0.665;conf=99.99;score=223.98;cscore=223.94;sscore=0.04;rscore=-3.31;uscore=-2.31;tscore=4.54;

And here is my phava command:

phava variation_wf -d phava_out/ -t 12 -i input.fasta -r reads.fastq --genes input.fasta.genes.gff --genesFormat gff

Inverton search without gene input works well, so for now I'll manually parse the GFF outside of Phava. Could a prodigal GFF input option be added, and/or the documentation changed to clarify that the current GFF requirement is an NCBI GFF?

Thanks,
Christine

[patrickwest]

Hi Christine,

Thanks for the bug report, I have changed the documentation to indicate it expects an NCBI format gff rather than prodigal. I'll work on adding parsing for prodigal gff format as well to support de novo annotations. Sorry for the inconvenience

Patrick