Two questions about maker

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

Two questions about maker

Zhan, Shuai
Hello,
 
I'm new for the maker system.
Currently, I have two minor questions.
 
1) As a test for my data, maker always reported "Segmentation fault" at the last step, but it had generated the log information of "Maker is now finished!!!". I also noticed the expected result files, such as *.gff, were also generated, and saw all scaffolds were "FINISHED" in another log file named "_master_datastore_index.log".
BUT I'm still not clear if it's normal and if I'll miss some key result when maker failed at this step.
 
2) If I have some other genesets other than "SNAP", "Augustus", and "GeneMark", my understanding is to add the generated gff3 files on "other_gff=" of maker_opts.ctl. But is there something required for maker to capture information from my gff3 files? I'm not clear if I need set some specific type or group values in my gff3 file.
 
I'd greatly appreciate any help.
 
Shuai Zhan

 

_______________________________________________
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: Two questions about maker

Carson Hinton Holt
Re: [maker-devel] Two questions about maker I’ve generated the segmentation fault before, but only for mpi_maker under the new hydra MPICH2.  When you use older MPICH2 or the MPD communicator this error should not occur.

Under hydra, the error is harmless as it occurs after perl and MAKER have exited.  It is a bug related to hydra and its incompatibility with perl signals which are not instantaneous like C signals.

The other_gff is meant for data that MAKER cannot handle but which you want in the GFF3.  It just prints out all lines as is to the output file.  You can get the same results by merging the output GFF3 with your other GFF3 file using the gff3_merge script that comes with MAKER.

If you want to pass in existing gene models or gene predictions, use the model_gff and pred_gff options.

Thanks,
Carson


On 2/24/11 12:03 PM, "Zhan, Shuai" <Shuai.Zhan@...> wrote:

Hello,

I'm new for the maker system.
Currently, I have two minor questions.
 
1) As a test for my data, maker always reported "Segmentation fault" at the last step, but it had generated the log information of "Maker is now finished!!!". I also noticed the expected result files, such as *.gff, were also generated, and saw all scaffolds were "FINISHED" in another log file named "_master_datastore_index.log".
BUT I'm still not clear if it's normal and if I'll miss some key result when maker failed at this step.
 
2) If I have some other genesets other than "SNAP", "Augustus", and "GeneMark", my understanding is to add the generated gff3 files on "other_gff=" of maker_opts.ctl. But is there something required for maker to capture information from my gff3 files? I'm not clear if I need set some specific type or group values in my gff3 file.
 
I'd greatly appreciate any help.
 
Shuai Zhan

 


_______________________________________________
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
|

答复: Two questions about maker

Zhan, Shuai
Thanks for pointing it out Carson, I've known where to set them.

Do you know which types and tags are required or recognized by maker in external GFF3 file?

Thanks,
Shuai
________________________________________
发件人: Carson Holt [[hidden email]]
发送时间: 2011年2月24日 14:14
收件人: Zhan, Shuai; [hidden email]
主题: Re: [maker-devel] Two questions about maker

I’ve generated the segmentation fault before, but only for mpi_maker under the new hydra MPICH2.  When you use older MPICH2 or the MPD communicator this error should not occur.

Under hydra, the error is harmless as it occurs after perl and MAKER have exited.  It is a bug related to hydra and its incompatibility with perl signals which are not instantaneous like C signals.

The other_gff is meant for data that MAKER cannot handle but which you want in the GFF3.  It just prints out all lines as is to the output file.  You can get the same results by merging the output GFF3 with your other GFF3 file using the gff3_merge script that comes with MAKER.

If you want to pass in existing gene models or gene predictions, use the model_gff and pred_gff options.

Thanks,
Carson


On 2/24/11 12:03 PM, "Zhan, Shuai" <[hidden email]> wrote:

Hello,

I'm new for the maker system.
Currently, I have two minor questions.

1) As a test for my data, maker always reported "Segmentation fault" at the last step, but it had generated the log information of "Maker is now finished!!!". I also noticed the expected result files, such as *.gff, were also generated, and saw all scaffolds were "FINISHED" in another log file named "_master_datastore_index.log".
BUT I'm still not clear if it's normal and if I'll miss some key result when maker failed at this step.

2) If I have some other genesets other than "SNAP", "Augustus", and "GeneMark", my understanding is to add the generated gff3 files on "other_gff=" of maker_opts.ctl. But is there something required for maker to capture information from my gff3 files? I'm not clear if I need set some specific type or group values in my gff3 file.

I'd greatly appreciate any help.

Shuai Zhan



_______________________________________________
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: Two questions about maker

Carson Hinton Holt
Re: 答复: [maker-devel] Two questions about maker By default everything in the pred_gff will be considered a prediction regardless of the type tag (column 3).  This is also true for est_gff, protein_gff, etc.  They expect everything in them to be of the type specified.  However to avoid any weird behavior predictions should be provided as either match/match_part or gene/mRNA/exon/CDS features even though if other tags are there MAKER will try and turn them into at least a two level structure.

Model_gff expects gene/mRNA/exon/CDS features.

Column 2 the “source” tag will basically be ignored, because you already identified feature types by providing the file as either pred_gff, protein_gff, etc.

Thanks,
Carson


On 2/24/11 12:31 PM, "Zhan, Shuai" <Shuai.Zhan@...> wrote:

Thanks for pointing it out Carson, I've known where to set them.

Do you know which types and tags are required or recognized by maker in external GFF3 file?

Thanks,
Shuai
________________________________________
发件人: Carson Holt [carson.holt@...]
发送时间: 2011年2月24日 14:14
收件人: Zhan, Shuai; maker-devel@...
主题: Re: [maker-devel] Two questions about maker

I’ve generated the segmentation fault before, but only for mpi_maker under the new hydra MPICH2.  When you use older MPICH2 or the MPD communicator this error should not occur.

Under hydra, the error is harmless as it occurs after perl and MAKER have exited.  It is a bug related to hydra and its incompatibility with perl signals which are not instantaneous like C signals.

The other_gff is meant for data that MAKER cannot handle but which you want in the GFF3.  It just prints out all lines as is to the output file.  You can get the same results by merging the output GFF3 with your other GFF3 file using the gff3_merge script that comes with MAKER.

If you want to pass in existing gene models or gene predictions, use the model_gff and pred_gff options.

Thanks,
Carson


On 2/24/11 12:03 PM, "Zhan, Shuai" <Shuai.Zhan@...> wrote:

Hello,

I'm new for the maker system.
Currently, I have two minor questions.

1) As a test for my data, maker always reported "Segmentation fault" at the last step, but it had generated the log information of "Maker is now finished!!!". I also noticed the expected result files, such as *.gff, were also generated, and saw all scaffolds were "FINISHED" in another log file named "_master_datastore_index.log".
BUT I'm still not clear if it's normal and if I'll miss some key result when maker failed at this step.

2) If I have some other genesets other than "SNAP", "Augustus", and "GeneMark", my understanding is to add the generated gff3 files on "other_gff=" of maker_opts.ctl. But is there something required for maker to capture information from my gff3 files? I'm not clear if I need set some specific type or group values in my gff3 file.

I'd greatly appreciate any help.

Shuai Zhan





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