[maker-devel] AED score

[Carson Holt]

Use the AED_threshold option in the maker_opts.ctl file if you just want
to restrict final gene models to close matches directly within maker.  On
the other hand, if you are trying to build a dataset for training gene
predictors, use the maker2zff script for generating a filtered dataset for
SNAP training.  There are a number of filters available. Just call the
script once without parameters to see the options.

Thanks,
Carson




On 12-11-27 5:55 PM, "Daniel Ence" <dence at genetics.utah.edu> wrote:

>Hi Parul, 
>
>I think the way you described (with the maker_opts.ctl file) is how you
>want to proceed. You still need to give the genome too.
>
>Daniel 
>
>
>Daniel Ence
>Graduate Student
>Eccles Institute of Human Genetics
>University of Utah
>15 North 2030 East, Room 2100
>Salt Lake City, UT 84112-5330
>________________________________________
>From: maker-devel-bounces at yandell-lab.org
>[maker-devel-bounces at yandell-lab.org] on behalf of Parul Kudtarkar
>[parulk at caltech.edu]
>Sent: Tuesday, November 27, 2012 3:41 PM
>To: Parul Kudtarkar
>Cc: maker-devel at yandell-lab.org
>Subject: Re: [maker-devel] AED score
>
>Also, are there any other parameters that are required when filtering
>based on AED score?
>
>> Hello Carson,
>>
>> Just to confirm, Is there a script that would filter gene models at
>> specific AED score.
>> Alternatively if I were to do this within maker with regards to
>>parameters
>> in maker_opts.ctl file I would have to provide my predicted genes gff3
>> file to model_gff and  set AED_threshold at desired threshold?
>>
>> Thanks and regards,
>> Parul Kudtarkar
>>
>>> AED score with 1 are the ones you don't want.  0 is best and 1 is worst
>>> as
>>> it is a distance metric.  You can use the AED_threshold parameter to
>>> require better matching to the evidence by setting it closer to 0. You
>>> can
>>> also try to increase protein homology evidence as some of your calls
>>>may
>>> be split genes due to lack of evidence linking them.
>>>
>>> --Carson
>>>
>>>
>>> On 12-11-26 4:35 PM, "Parul Kudtarkar" <parulk at caltech.edu> wrote:
>>>
>>>>Dear Maker community,
>>>>
>>>>For gene-prediction I get training data-set from evidence based
>>>>prediction, I use this data-set to train SNAP as well as Augustus
>>>>predictions, followed by boot-strapping. I would typically expect
>>>>20-30K
>>>>genes however I am getting 8 times the expected gene count indicating
>>>> too
>>>>many false positives. Is there a way to further refine these
>>>>predication/script to retain predictions with AED score 1 and if yes
>>>>how
>>>>to go about this?
>>>>
>>>>Thanks and regards,
>>>>Parul Kudtarkar
>>>>
>>>>--
>>>>Scientific Programmer
>>>>Center for Computational Regulatory Genomics
>>>>Beckman Institute,
>>>>California Institute of Technology
>>>>http://www.spbase.org
>>>>
>>>>
>>>>_______________________________________________
>>>>maker-devel mailing list
>>>>maker-devel at box290.bluehost.com
>>>>http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
>>>
>>>
>>>
>>
>>
>> --
>> Scientific Programmer
>> Center for Computational Regulatory Genomics
>> Beckman Institute,
>> California Institute of Technology
>> http://www.spbase.org
>>
>
>
>--
>Scientific Programmer
>Center for Computational Regulatory Genomics
>Beckman Institute,
>California Institute of Technology
>http://www.spbase.org
>
>
>_______________________________________________
>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