Hi, I want to use TSEBRA for my BRAKER3 outputs to improve gene prediction. the thing is in the galaxy webserver I can put in only one GTF file as input and I thought TSEBRA needs 2 GTF inputs and then combine the 2. Can someone clarify this for me?
Thanks in advance!
Welcome, @rachelihadjez
I see two places for GTF types of information on the form. Do those match up with what you expected? If not, can you explain more about what you did expect? Maybe with screenshots or links to the guide or notes you are using? We can maybe explain what is going on, and even report problems to the developers that way if needed.
Screenshot of the tool form
Notice that I used the “accepted formats” toggle under input areas to review the datatype format of the expected inputs. I see two GTF inputs here.
Now, the BRAKER3 output can include alternative predictions in the output. But you wouldn’t be running this tool with both the primary and predicted at the same time, right? But you could run both as separate jobs (different datasets, or put those in a collection, label with # tags, then run as a batch). Am I misunderstanding how that is usually used?
Hope this helps, and we can follow up more, or let us know if this explained things. 
Hi,
Thank you so much for your reply!
I’m trying to annotate a genome that hasn’t been annotated before, so I ran BRAKER3 twice:
- Once with RNA reads mapped to the genome.
- Another time with RNA reads and the proteins from a related species.
Each BRAKER3 run gave me a GTF file (the “main” annotation). My understanding was that combining these two GTFs with TSEBRA would give me a better final annotation, but maybe I misunderstood how TSEBRA works.
About the input: I thought there would be two places for GTF inputs (like the blue section that says “GTF input”), but instead, I see only one GTF input and a GFF file for hints.
Could you clarify this for me please?
Sorry if I wasn’t clear before—I’m very new to this! Thank you so much for your help!
Best,
Racheli
There are some licensing issue with these tools, which is probably part of the confusion.
BRAKER3 runs TSEBRA internally. The result in Galaxy is the single primary output: braker.gtf. That is the first top output listed here →
Then this is in the Galaxy repository, and may explain why some of the other files are not output? If you search the tool panel with the keywork “braker” you’ll find the tools and versions of those tool that can be hosted on a public server. If you can’t find one of the tools mentioned in a publication or in Github, the licensing issue is probably why but you can ask.
Then, to complicated it 
TSEBRA can also be run by itself, and that can use another file that is created by one of the tools that Galaxy cannot host on public sites, but you could upload it (the “hints” files).
- GitHub - Gaius-Augustus/TSEBRA: TSEBRA: Transcript Selector for BRAKER
- tools-iuc/tools/tsebra at main · galaxyproject/tools-iuc · GitHub
Does this help?
Yes this helps! Thank you so much!!!
1 Like