[maker-devel] Failed while polishig ESTs
Jason E Stajich
jason.stajich at ucr.edu
Fri Oct 5 01:02:50 MDT 2012
I wrote up how to use it on this list msg - http://brie4.cshl.edu/pipermail/gmod-help/2012-June/001724.html
Script is in this github repo -
https://github.com/hyphaltip/genome-scripts/blob/master/gene_prediction/zff2augustus_gbk.pl
Patches and documentation updates are always welcomed, I'll get around to that eventually.
Jason
On Oct 4, 2012, at 10:34 PM, Parul Kudtarkar <parulk at caltech.edu<mailto:parulk at caltech.edu>>
wrote:
Carson, thanks for very quick response. Jason, could you please
elaborate/share document if any on using zff2augustus_gbk.pl
Thanks,
Parul Kudtarkar
Just concatenating won't work. Use the gff3_merge tool to safely merge
separate GFF3 files.
In the tutorial dataset their should only be one contig to train with, but
with real data you will have multiple contigs and will want to merge the
GFF3 files.
Augustus has a separate training procedure that is a little more
complicated than SNAP's. You will need to read their documentation on
training which is a little cryptic.
I know Jason Stajich developed a tool for training augustus from snap
results called zff2augustus_gbk.pl (I've CC'd him). Perhaps he would be
willing to share it with you ;-)
Thanks,
Carson
On 12-10-04 7:06 PM, "Parul Kudtarkar" <parulk at caltech.edu<mailto:parulk at caltech.edu>> wrote:
Hi,
Thanks, reinstalling BioPerl fixed the issue. Now we are training our
genome to generate training dataset for SNAP and Augustus, as per the
instructions provided at
http://gmod.org/wiki/MAKER_Tutorial#Training_ab_initio_Gene_Predictors
However the training dataset would have multiple gff3 prediction files,
one for each contig. Is it recommended to cat(command) all the gff3 files
to generate a single file and finally generate hmm file with the steps
mentioned in the tutorial. Would the same hmm file work as training
dataset for AUGUSTUS?
Many thanks,
Parul Kudtarkar
You probably need to reinstall Bio-perl.
Other things that can cause the same error are setting TMP in the
maker_opts.ctl file to a tmpfs type filesystem (i.e. in memory drive)
or
sometimes setting it to an NFS mount.
A full drive can cause this as well or a broken Berkley DB. Use df -h
to
see if any of the drives used are full (either current working
directory
or TMP location). You can also swap Berkley DB for a different backend
by
setting AnyDBM_ISA during setup.
Example:
cd ../maker/src/
perl Build.PL --AnyDBM_ISA SDBM_File
./Build install
--Carson
On 12-10-03 10:14 PM, "Parul Kudtarkar" <parulk at caltech.edu> wrote:
I am running maker on example data that comes along with installation
and
is cited at http://gmod.org/wiki/MAKER_Tutorial_2012#Note
Please advice on the aforementioned error.
---------------------
Maker is now finished!!!
re reading blast report.
/usr/local/maker/test/dpp_contig.maker.output/dpp_contig_datastore/05/1F
/c
ontig-dpp-500-500//theVoid.contig-dpp-500-500/contig-dpp-500-500.0.te_pr
ot
eins%2Efasta.repeatrunner
deleted:0 hits
in cluster:shadow cluster...
i_size:0 j_size:0
sorting hits in shadow cluster...
... finished.
re reading blast report.
/usr/local/maker/test/dpp_contig.maker.output/dpp_contig_datastore/05/1F
/c
ontig-dpp-500-500//theVoid.contig-dpp-500-500/contig-dpp-500-500.0.dpp_e
st
%2Efasta.blastn
deleted:-1 hits
re reading blast report.
/usr/local/maker/test/dpp_contig.maker.output/dpp_contig_datastore/05/1F
/c
ontig-dpp-500-500//theVoid.contig-dpp-500-500/contig-dpp-500-500.0.dpp_p
ro
tein%2Efasta.blastx
WARNING: Multiple MAKER processes have been started in the
same directory.
deleted:0 hits
WARNING: Cannot find> dpp-mRNA-4, trying to re-index the fasta.
stop here:dpp-mRNA-4
ERROR: Fasta index error
FATAL ERROR
Maker is now finished!!!
re reading blast report.
/usr/local/maker/test/dpp_contig.maker.output/dpp_contig_datastore/05/1F
/c
ontig-dpp-500-500//theVoid.contig-dpp-500-500/contig-dpp-500-500.0.te_pr
ot
eins%2Efasta.repeatrunner
deleted:0 hits
in cluster:shadow cluster...
i_size:0 j_size:0
sorting hits in shadow cluster...
... finished.
re reading blast report.
/usr/local/maker/test/dpp_contig.maker.output/dpp_contig_datastore/05/1F
/c
ontig-dpp-500-500//theVoid.contig-dpp-500-500/contig-dpp-500-500.0.dpp_e
st
%2Efasta.blastn
deleted:-1 hits
re reading blast report.
/usr/local/maker/test/dpp_contig.maker.output/dpp_contig_datastore/05/1F
/c
ontig-dpp-500-500//theVoid.contig-dpp-500-500/contig-dpp-500-500.0.dpp_p
ro
tein%2Efasta.blastx
WARNING: Multiple MAKER processes have been started in the
same directory.
deleted:0 hits
WARNING: Cannot find> dpp-mRNA-4, trying to re-index the fasta.
stop here:dpp-mRNA-4
ERROR: Fasta index error
FATAL ERROR
ERROR: Failed while polishig ESTs!!
ERROR: Chunk failed at level 14
!!
FAILED CONTIG:contig-dpp-500-500
ERROR: Chunk failed at level 14
!!
FAILED CONTIG:contig-dpp-500-500
------------------------------------
Many thanks,
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
--
Jason E Stajich, PhD
Assistant Professor
Plant Pathology & Microbiology
University of California, Riverside
951.827.2363
http://lab.stajich.org<http://lab.stajich.org/> http://fungalgenomes.org<http://fungalgenomes.org/> http://fungidb.org<http://fungidb.org/> http://1000.fungalgenomes.org/
twitter @stajichlab @hyphaltip @fungalgenomes @fungidb
http://plantpathology.ucr.edu<http://plantpathology.ucr.edu/> http://genomics.ucr.edu<http://genomics.ucr.edu/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://yandell-lab.org/pipermail/maker-devel_yandell-lab.org/attachments/20121005/ee912ee3/attachment-0003.html>
More information about the maker-devel
mailing list