[maker-devel] Difference between AED=1 and keep_preds=1

[Carson Holt]

keep_preds=1 will take all the raw predictions remove any that overlap the current supported gene set, then filter for redundancy (i.e. if you ran both SNAP and Augustus), then add what’s left to the final gene models. It’s basically the equivalent of taking all the models in the non-overlapping.abiinit file in a normal MAKER run and adding them to the final models.

—Carson


> On Dec 1, 2017, at 1:27 PM, Patrick Tran Van <patrick.tranvan at unil.ch> wrote:
> 
> Hi,
> 
> From what I understand, an AED of 1 means that the gene predicted by gene finder is not supported by external evidence (transcriptome and protein).
> 
> So how is it different with keep_preds=1, which is describe as "add unsupported gene prediction" ?
> Because I have way more gene with keep_pred=1. 
> 
> Maybe I missed something.
> 
> Thanks.
> 
> Patrick 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20171201/738e56de/attachment-0003.html>