[maker-devel] Running maker with suboptimal evidence

[Carson Holt]

Run it both ways then look at it in a browser like Apollo.  You can then visually see how evidence alignments compare to the models and if they are spurious. While not practical in most situations, manual review is still the gold standard for genome annotation.

—Carson



> On Aug 8, 2019, at 6:41 PM, Martin, John <jmartin at wustl.edu> wrote:
> 
> Greetings,
> 
>     I would like to annotate a worm genome of ~90Mb but the evidence I
> have is not all of good quality.  I have 2 high quality protein sets
> from previously finished & curated, closely related worms.  I also have
> a small amount of RNAseq and some old EST data neither of which give
> good coverage of the transcriptome.  And I have a previously run Maker
> geneset for this worm that I believe was generated using probably all
> nematodes from genbank and that poor set of RNAseq.  I also have a
> 'fair' set of predictions from running Braker2 using only the high
> quality protein data I mentioned.   My opinion that the previous Maker
> geneset is of poor quality comes from comparing that geneset to my
> recent Braker2 annotations and that RNAseq.
> 
>     I would like to try and build an improved annotation for this
> assembly but I'm unsure of whether I should use all this evidence, or if
> I would be better off not using the low coverage RNAseq, EST data and
> previous (questionable looking) Maker geneset.   I'm looking for
> opinions on whether I should throw all the evidence I have into Maker or
> should I use only evidence that I consider of good quality?
> 
> 
> Thanks,
> 
> John Martin
> 
> 
> 
> ________________________________
> The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.
> _______________________________________________
> maker-devel mailing list
> maker-devel at box290.bluehost.com
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org