Artemis Or Apollo

Question

[Daniel Fernández]

Hi, I have some very simple questions,

• Is possible to load the Exonerate's output (GFF2) in Artemis ?

  ... I trying with some scripts and exe to convert but without success (processexonerategff3fast.pl; gffread)

• If the first question is not. Can Apollo load GFF2 and/or GFF3?

  ... I trying to install apollo but I have some errors in the installation

• Or ... Can you tell me another program to predict genes by homology, and which format is the output, GFF2 or GFF3, please please please?

  ... I've desperate, I posted here before but still I can not visualize my annotation format.

Please, any help.

=======================================================================

Hi Zev, I use the script that you tell me, but it show a little error, "gff_version|gf|v|:s",

$ exonerate -q Scaffold.fa -Q dna -t krt.fa -T dna -m e2g --refine region --showtargetgff yes --ryo ">%qi length=%ql alnlen=%qal\n>%ti length=%tl alnlen=%tal\n" > out.exonerate
$ ./process_exonerate_gff3.pl -t EST out.exonerate > out.gff
Error in option spec: "gff_version|gf|v|:s"

I add "-gf 3" and in othe order I add "-gff_version 3" and they both display the same error. can you help me

[Zev.Kronenberg]

See edits below

[Daniel Fernández]

Sorry, I'm not good in that, I tried with "fix_exhon -gf 2" and it show me the same error.

[Zev.Kronenberg]

Send me the new exonerate output file. Same email.

Answer 1

[Zev.Kronenberg]

Apollo is designed to look at gene annotations and can take gff3. I could help you bring your annotation file up to gff3 spec. That is if you are willing to share it.

Okay, after looking at your output I think you may need to do a little tweaking to get your data in a format that can be viewed.

You want to use this script: Gmod GFF3 exonerate.

The guys over at Gmod have their act together and I think this is the easiest solution.

However exonerate must be run with the settings below:

    −−showtargetgff yes
    −−ryo ">%qi length=%ql alnlen=%qal\n>%ti length=%tl alnlen=%tal\n"

---

  perl fix_exhon -gf 2 ~/Downloads/exonerate.gff > t.gff3

[Daniel Fernández]

yes, thanks, do I give you my output gff2? or how to

[Zev.Kronenberg]

If it is small you can email it to zev.kronenberg [at] gmail.com. If it is larger you can create a drop box. Also just for piece of mind: I have no intention to distribute or use these data.

[Daniel Fernández]

thanks I will send you by mail my gff2.

[Daniel Fernández]

Hi, I sent my gff2 by mail on Saturday, I hope you have received.

[Zev.Kronenberg]

Yes, I am playing with it now.

[Zev.Kronenberg]

see edits above

[Daniel Fernández]

Hi, thanks for the script, it show me a little error, see above

Answer 2

[JC]

Is possible to load the Exonerate's output (GFF2) in Artemis ?

You can try to convert the file with biopython: http://biopython.org/wiki/GFF_Parsing

If the first question is not. Can Apollo load GFF2 and/or GFF3?

Apollo can read both -> http://apollo.berkeleybop.org/current/userguide.html#ReadingGFF

Or ... Can you tell me another program to predict genes by homology, and which format is the output, GFF2 or GFF3, please please please?

Augustus http://bioinf.uni-greifswald.de/augustus/ can produce GFF3 and it's a really good gene predictor (you can use the homolog genes as "hints").

[Daniel Fernández]

Now I was trying with augustus, but the result doesn't convince me, because the homology.out.gff is very similar to the abinitio.out.gff, mmm ... this biopython, how I can install the modules? because there said that is not yet integrated to biopython, so I guess I only need this modules.

from BCBio import GFF

from Bio import SeqIO

Answer 3

[Christian]

Have you looked at genBlastG? It can produce GFF3 output.

genBlastG: using BLAST searches to build homologous gene models.

Bioinformatics (2011) 27 (15): 2141-2143.