[maker-devel] maker pipeline 2nd round updating augustus

[Carson Holt]

Don’t use BUSCO to train for the second round, there is a bias in the models it produces for conserved genes that tend to be short and intron poor., You will want to avoid this bias in the second round. You want to use a broad selection of gene models instead.

Use the maker2zff script to select gene models for training (examples on doing this can be found on the maker tutorial wiki). Then use this script to convert ZFF to GenBank format to train Augustus —> https://github.com/hyphaltip/genome-scripts/blob/master/gene_prediction/zff2augustus_gbk.pl <https://github.com/hyphaltip/genome-scripts/blob/master/gene_prediction/zff2augustus_gbk.pl>

This is a nice guide to train Augustus using GenBank format input—> https://vcru.wisc.edu/simonlab/bioinformatics/programs/augustus/docs/tutorial2015/training.html <https://vcru.wisc.edu/simonlab/bioinformatics/programs/augustus/docs/tutorial2015/training.html>

—Carson



> On Jan 23, 2018, at 1:35 PM, Giroux, Emily (CFIA/ACIA) <Emily.Giroux at inspection.gc.ca> wrote:
> 
> Hi,
>  
> I completed a first round of Maker, followed by snap and BUSCO to train augustus. I then placed the newly-trained species-specific augustus files in the augustus species directory and used this for my second round of maker.
>  
> What I’m wondering now is whether I should repeat this process after completeing round 2 of maker, and follow this with using BUSCO to retrain the augustus files again and replace the previous species-specific libraries from round 1 with those from round 2 and use these as input for my third round of maker.
>  
> Thank-you very much,
>  
> Emily
>  
> _______________________________________________
> maker-devel mailing list
> maker-devel at box290.bluehost.com <mailto:maker-devel at box290.bluehost.com>
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org <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/20180125/0f39c27a/attachment-0002.html>