StringTie output to train Augustus

Question

[Ric]

Hi, I ran StringTie and transdecoder. How is it possible to use these outputs as input for Augustus to predict genes with UTR?

Thank you in advance

[Juke34]

have a look here https://github.com/ikmb-denbi/genome-annotation they produce a lot of different evidence to feed Augustus. They have a lot of utilities to produce hints from gtf/gff

[Ric]

Thank you for your link. I also found Incorporating RNA-Seq into AUGUSTUS but they used BLAT. However, I have done all my alignment with HIsat2. Does anyone know how to change the commands to BAM files?

[Juke34]

From Hisat bam file you can run bam2hints

[Ric]

Thank you, is there a way or necessary to add strand information to the output of bam2hints?

UPDATE:

Bam2hints resutls:

> wc -l .intron-hints.gff 
438503 .intron-hints.gff

.

> head .intron-hints.gff 
NbV1Ch01        b2h     intron  212732  212828  0       .       .       mult=150;pri=4;src=E
NbV1Ch01        b2h     intron  212969  215836  0       .       .       mult=4;pri=4;src=E
NbV1Ch01        b2h     intron  212969  218079  0       .       .       mult=141;pri=4;src=E
NbV1Ch01        b2h     intron  212969  219060  0       .       .       pri=4;src=E
NbV1Ch01        b2h     intron  218387  218908  0       .       .       mult=4;pri=4;src=E
NbV1Ch01        b2h     intron  218387  219060  0       .       .       mult=3;pri=4;src=E
NbV1Ch01        b2h     intron  218671  222563  0       .       .       pri=4;src=E
NbV1Ch01        b2h     intron  218970  219060  0       .       .       mult=4;pri=4;src=E
NbV1Ch01        b2h     intron  219193  219270  0       .       .       pri=4;src=E
NbV1Ch01        b2h     intron  219199  219270  0       .       .       mult=2;pri=4;src=E

In section 3.1 Set hint parameters of this tutorial they asking to adjust the file extrinsic.cfg that holds the hint parameters. However, where is the file located and how to change the settings?