[maker-devel] QI codes insufficient - how to get frac exons with EST only?

[Matt Simenc]

Very good, thank you!

Matt

On Wed, Oct 11, 2017 at 8:22 AM, Carson Holt <carsonhh at gmail.com> wrote:

> Also look at GAL for building GFF3 feature queries —>
> https://github.com/The-Sequence-Ontology/GAL
>
> —Carson
>
>
>
>
> On Oct 11, 2017, at 9:18 AM, Michael Campbell <
> michael.s.campbell1 at gmail.com> wrote:
>
> Hi Matt,
>
> I have a hacky way that I’ve done it. It requires running MAKER two more
> times but they are quicker runs.
>
> To identify the genes that have protein support I pass all of the
> annotation back to MAKER using the model_gff option in the maker_opts.ctl
> file. Then I pull out all of the protein2genome features from the big MAKER
> GFF3 file and pass them in using the protein_gff option. I turn off all
> repeat masking and run MAKER. It runs fast because it doesn’t have to run
> any gene finders, align evidence, or repeatmask. In the output any gene
> with an AED less than 1 has protein support. Then I do the same thing with
> est2genome lines from the big GFF3 file and put them in as est_gff. The
> output of that one gives you genes with EST support. Then the genes with an
> AED of less than one in both sets have support from protein and EST.
>
> Hope this helps,
> Mike
>
> On Oct 11, 2017, at 10:53 AM, Matt Simenc <mcsimenc at gmail.com> wrote:
>
> Hey MAKER people,
>
> I would like to make a Venn diagram showing the kinds of evidence
> supporting gene models in my MAKER annotation where the left side shows
> number of genes with EST support only, the right side shows number of genes
> with protein support only, and the intersection shows number of genes with
> EST and protein support.
>
> QI summary has:
>
> Fraction of exons that overlap an EST alignment
> Fraction of exons that overlap EST or Protein alignments
>
> Please correct me if I'm wrong, because I am interpreting the first to be
> fraction of exons that overlap an EST alignment and possibly also a protein
> alignment. If that is the case then we can't calculate the number of genes
> that overlap only EST or (EST and protein) from the QI information.
>
> Anyone have a way to do this or have a script to parse the MAKER GFF3 to
> get this?
>
> Thanks!!!
> Matt Simenc
> _______________________________________________
> maker-devel mailing list
> maker-devel at box290.bluehost.com
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
>
>
>
> _______________________________________________
> maker-devel mailing list
> maker-devel at box290.bluehost.com
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20171011/b854fffe/attachment-0003.html>