Augustus CMD ERROR during f2 predict

[edwhisnant]

Hi again, with another Augustus issue.

On another assembly, I run into an issue with augustus during the prediction step, where I get back a CMD ERROR. The error occurs on every contig in the assembly. Please let me know what information might be most helpful for this.

CMD ERROR: augustus --species=e121cdd7-f5e6-411d-85c8-7497a41dd37b --gff3=on --UTR=off --stopCodonExcludedFromCDS=False --extrinsicCfgFile=/hpc/group/bio1/ewhisnant/miniconda3/envs/funannotate2/lib/python3.12/site-packages/funannotate2/resources/extrinsic.E.XNT.RM.cfg --AUGUSTUS_CONFIG_PATH=/work/edw36/peltigerales/funannotate2/v26.03.30/peltigerales/Leptogium_sp._P12641/train_misc/augustus --hintsfile=/work/edw36/peltigerales/funannotate2/v26.03.30/peltigerales/f2-intermediate-files/tmp/Leptogium_sp._P12641_predict/contig_6.fasta.hintsfile /work/edw36/peltigerales/funannotate2/v26.03.30/peltigerales/f2-intermediate-files/tmp/Leptogium_sp._P12641_predict/contig_6.fasta

[nextgenusfs]

A small test that I can replicate locally is probably the easiest. Thanks.

[edwhisnant]

I have seen this error on two other assemblies today. Following up on the first one I reported previously:

I still do not know what is causing this, but the error is always the same: CMD ERROR: augustus [...]

Deleting the previous train and predict directories and reloading my species into the trained database seems to have made the augustus error go away and f2 predict run successfully. Currently testing if this will also work on my other two assemblies with this error.

[edwhisnant]

Okay, the scorched earth method seemed to fix this. This seemed to arise from re-starting the pipeline midway (e.g., being bumped to a different node on an HPC) without completely starting from scratch. This is the only commonality between the jobs that I noticed this error.