GeneMark-ES problem

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

GeneMark-ES problem

朱英杰
Dear Sir,
 
I am a researcher from IMPLAD. Recently, I use GeneMark-ES and Maker for annotate a fungi genome. I trained a model using gm_es.pl.  Then I take this model to gmhmme3. Unfortunately,  I can't find any output results except for gm.stats, but gm.stats include only a title. No error information is reported. If I have some error in maniplation, please tell me correct method.
 
Hope your reply as soon as possible!
 
 
Best regards,
 
Yingjie
 
Institute of medicinal plant development, Beijing
 

_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: GeneMark-ES problem

Jason Stajich-5
<base href="x-msg://26/">
Yingjie -


Did you get a specific error when running the program for the initial training/annotation run?  There should be a folder called 'mod' in the directory where you ran it.

Sometimes it fails on short contigs or with weird characters -- i think newest version of GM does better but you may still want to eliminate short contigs when doing the initial training.


On Jun 22, 2011, at 12:36 AM, 朱英杰 wrote:

Dear Sir,
 
I am a researcher from IMPLAD. Recently, I use GeneMark-ES and Maker for annotate a fungi genome. I trained a model using gm_es.pl.  Then I take this model to gmhmme3. Unfortunately,  I can't find any output results except for gm.stats, but gm.stats include only a title. No error information is reported. If I have some error in maniplation, please tell me correct method.
 
Hope your reply as soon as possible!
 
 
Best regards,
 
Yingjie
 
Institute of medicinal plant development, Beijing
 
_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org



-- 
Jason E Stajich, PhD
Assistant Professor
Dept of Plant Pathology & Microbiology
University of California, Riverside
951.827.2363
twitter @stajichlab @hyphaltip @fungalgenomes @fungidb
http://planpathology.ucr.edu http://genomics.ucr.edu 


_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: 答复: GeneMark-ES problem

Jason Stajich-5
<base href="x-msg://26/">Hi Yingjie -

You probably should email the genemark developers to get feedback on your error, though it may be problems with some short contigs -- you may want to remove all contigs  <5kb for your training procedure and see if this lets it complete the training.

For SNAP you could just run Maker with a starting model of something close  -- I provide a bunch trained from published genome sets here - http://fungalgenomes.org/data/Gene_prediction/Parameters/SNAP/ and then take the best results and then follow the retraining guidelines -- Maker makes this really easy:

Jason

On Jun 22, 2011, at 6:38 PM, 朱英杰 wrote:

Hi Jason,
 
I will try again follow your instruction. If you have some time, please look this log file. You will find the reasons of error. In addtion, I am using SNAP, I can't find a good method to train a mode. Can you teach me the common method? I will appreciate your any help you can give me.
 
Best,
 
Yingjie Zhu
 
 

发件人: Jason Stajich [[hidden email]]
发送时间: 2011年6月23日 0:50
收件人: 朱英杰
Cc: [hidden email]
主题: Re: [maker-devel] GeneMark-ES problem

Yingjie -


Did you get a specific error when running the program for the initial training/annotation run?  There should be a folder called 'mod' in the directory where you ran it.

Sometimes it fails on short contigs or with weird characters -- i think newest version of GM does better but you may still want to eliminate short contigs when doing the initial training.


On Jun 22, 2011, at 12:36 AM, 朱英杰 wrote:

Dear Sir,
 
I am a researcher from IMPLAD. Recently, I use GeneMark-ES and Maker for annotate a fungi genome. I trained a model using gm_es.pl.  Then I take this model to gmhmme3. Unfortunately,  I can't find any output results except for gm.stats, but gm.stats include only a title. No error information is reported. If I have some error in maniplation, please tell me correct method.
 
Hope your reply as soon as possible!
 
 
Best regards,
 
Yingjie
 
Institute of medicinal plant development, Beijing
 
_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org



-- 
Jason E Stajich, PhD
Assistant Professor
Dept of Plant Pathology & Microbiology
University of California, Riverside
951.827.2363
twitter @stajichlab @hyphaltip @fungalgenomes @fungidb
http://planpathology.ucr.edu http://genomics.ucr.edu 

<nohup.out>

-- 
Jason E Stajich, PhD
Assistant Professor
Dept of Plant Pathology & Microbiology
University of California, Riverside
951.827.2363
twitter @stajichlab @hyphaltip @fungalgenomes @fungidb
http://planpathology.ucr.edu http://genomics.ucr.edu 


_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: 答复: GeneMark-ES problem

Carson Hinton Holt
Re: [maker-devel] ��:  GeneMark-ES problem Also look for runs of NNNNNNNN in your contigs.  GeneMark auto-splits contigs when there are 40 or more N’s in a row.  The result is that a contig you thought was large is split into many smaller contigs.  If none of these smaller contigs are above GeneMark’s default threshold (20,000 bp). Then GeneMark will not produce any results.  This is a big gotcha for some assemblies.

If you are using MAKER, just supply the .mod file produced by gm_es.pl.  MAKER will run
gmhmme3 for you.

--Carson


On 6/22/11 11:39 PM, "Jason Stajich" <jason.stajich@...> wrote:

Hi Yingjie -

You probably should email the genemark developers to get feedback on your error, though it may be problems with some short contigs -- you may want to remove all contigs  <5kb for your training procedure and see if this lets it complete the training.

For SNAP you could just run Maker with a starting model of something close  -- I provide a bunch trained from published genome sets here - http://fungalgenomes.org/data/Gene_prediction/Parameters/SNAP/ and then take the best results and then follow the retraining guidelines -- Maker makes this really easy:
http://gmod.org/wiki/MAKER_Tutorial#Training_ab_initio_Gene_Predictors

 <http://gmod.org/wiki/MAKER_Tutorial#Training_ab_initio_Gene_Predictors> Jason

On Jun 22, 2011, at 6:38 PM, ��Ӣ�� wrote:

Hi Jason,
 
I will try again follow your instruction. If you have some time, please look this log file. You will find the reasons of error. In addtion, I am using SNAP, I can't find a good method to train a mode. Can you teach me the common method? I will appreciate your any help you can give me.
 
Best,
 
Yingjie Zhu
 
 

������: Jason Stajich [jason.stajich@...]
����ʱ��: 2011��6��23�� 0:50
�ռ���: ��Ӣ��
Cc: maker-devel@...
����: Re: [maker-devel] GeneMark-ES problem

Yingjie -


Did you get a specific error when running the program for the initial training/annotation run?  There should be a folder called 'mod' in the directory where you ran it.

Sometimes it fails on short contigs or with weird characters -- i think newest version of GM does better but you may still want to eliminate short contigs when doing the initial training.


On Jun 22, 2011, at 12:36 AM, ��Ӣ�� wrote:

Dear Sir,
 
I am a researcher from IMPLAD. Recently, I use GeneMark-ES and Maker for annotate a fungi genome. I trained a model using gm_es.pl.  Then I take this model to gmhmme3. Unfortunately,  I can't find any output results except for gm.stats, but gm.stats include only a title. No error information is reported. If I have some error in maniplation, please tell me correct method.
 
Hope your reply as soon as possible!
 
 
Best regards,
 
Yingjie
 
Institute of medicinal plant development, Beijing
 
_______________________________________________
maker-devel mailing list
maker-devel@...
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org



Carson Holt
Graduate Student
Yandell Lab
http:/www.yandell-lab.org/
Eccles Institute of Human Genetics
University of Utah

_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org