No prothint_augustus.gff output

[peritob]

Hi and thanks for providing this software.

I am running some distant evolutionary proteins against a fasta to input into braker2. I get the prothint.gff output but no prothint_augustus.gff.

head prothint.gff
Chr06:237871926-237874493+      ProtHint        Intron  628     721     1       +       .       al_score=0.236774; splice_sites=gt_ag;
Chr17:109173795-109214495+      ProtHint        Intron  19687   19944   1       +       .       al_score=0.152497; splice_sites=gt_ag;
Chr17:70316079-70386418-        ProtHint        Intron  20280   20342   1       +       .       al_score=0.129907; splice_sites=gt_ag;

The stop.gff is empty. Is there a flag I am missing?

Also, I am not getting an output from running wholly within braker2, hence trying this way.

thanks in advance for your help

[peritob]

Found the script to convert. Probably should have looked first. All sorted but is there a flag I should use when running the initial command?

[tomasbruna]

Hi @peritob,

ProtHint should call the conversion script automatically, I don't know why it does not in your case. Can you share the ProtHint log (it is printed to stderr)? I'd like to see whether there were any errors reported.

Best,
Tomas

[peritob]

Thanks for following up Tomas,

See attached. I should add that I am testing something here - by not masking the genome.fasta. I want to find genes that contain repetitive regions that are sometimes missed in annotation. This might be the issue?

prothint_eror.txt

[tomasbruna]

Hi @peritob,

I see what the error is, thanks for sharing the log.

There are no stop codons in the ProtHint output. This can be caused by some issues in the input data of just because your input is too small. How large is your input genome? And how many proteins do you have in the database?

Also, I noticed that you are using an older ProtHint version (2.4.0). Please try getting the latest release (ProtHint 2.6.0), some of the errors might be fixed in the latest version.

Best,
Tomas

[peritob]

Thanks Tomas,

I ran with v.2.6.0 and it completed correctly. Yes, I have a small input data but it still seems to work which is great. Many thanks for following up on this.

I will now test within the baker2 and report back.

[tomasbruna]

I'm going to close this issue, please open a new one if you encounter errors in your test.