[maker-devel] keep_preds parameter

[dhivya arasappan]

Thank you Daniel. But I thought an AED score of zero indicates complete agreement of annotation to evidence and that 1 would mean no agreement?

Dhivya

On Apr 7, 2014, at 11:06 AM, Daniel Ence <dence at genetics.utah.edu> wrote:

> Hi Dhivya, 
> 
> That's a correct understanding of keep_preds, and it is a binary parameter; you either tell MAKER to keep the unsupported predictions or not to keep the unsupported predictions. In the output, you can tell which genes are supported by the _AED attribute in the gff3 file. Genes with and AED equal to zero have no support from the evidence sets (protein and EST and alt_EST). 
> 
> ~Daniel
> On Apr 7, 2014, at 9:57 AM, dhivya arasappan <darasappan at gmail.com>
> wrote:
> 
>> Hello,
>> 
>> I’m looking for a little more explanation about keep_preds parameter.  The documentation says that it is a threshold to add unsupported gene predictions. Along with some other changes, I set keep_preds=1 and saw a huge jump in the number of genes I was getting. Is setting this parameter to 1 equivalent to saying, include all predicted genes in my output, even if they are not supported by my set or protein data?  Is there a way to tell from my output which genes are unsupported and which are not?  Also, are the only two options for this parameter 0 and 1?
>> 
>> Thanks
>> dhivya
>> _______________________________________________
>> maker-devel mailing list
>> maker-devel at box290.bluehost.com
>> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
>