[Bioperl-pipeline] help needed : Genewise in pipeline not getting
input
Shawn Hoon
shawnh at stanford.edu
Sun Jan 18 18:29:05 EST 2004
Hi Tania,
the way it works is as follows:
setup_genewise has two params :
<module>setup_genewise</module>
<rank>1</rank>
<argument>
<tag>slice_ioh</tag>
<value>$slice_ioh</value>
</argument>
<argument>
<tag>protein_ioh</tag>
<value>$protein_ioh</value>
</argument>
slice_ioh and protein_ioh: since they both have the $ in front, it
refers to a variable defined in the global tag: ( not relevant here,
but this might be deprecated soon)
<global rootdir="/Users/tania/work"
ens_dbname="test_pipe"
ens_dbdriver="mysql"
ens_dbhost="localhost"
ens_dbuser="root"
ens_dbpasswd=""
blast1_logicname="BlastCerevisiae"
blast1_dbname="SCerevisiae"
blast1_dbfile="$rootdir/data/S_cerevisiae_sgd.fa"
blast1_dbdir = "$rootdir/data/"
blast1_dbversion ="test version"
blast1_param = " -p blastx -e 1e-5"
slice_ioh="4"
protein_ioh="5"
/>
So it has the value 4 and 5 here, which in turn point to the
iohandler_id=4 and iohandler_id=5
for your xml, its:
<iohandler id="4">
<adaptor_id>1</adaptor_id>
<adaptor_type>DB</adaptor_type>
<iohandler_type>INPUT</iohandler_type>
<method>
<name>get_CloneAdaptor</name>
<rank>1</rank>
</method>
<method>
<name>fetch_by_dbID</name>
<argument>
<value>!INPUT!</value>
<type>SCALAR</type>
<rank>1</rank>
</argument>
<rank>2</rank>
</method>
<method>
<name>get_all_ProteinAlignFeatures</name>
<rank>3</rank>
</method>
</iohandler>
<iohandler id="5">
<adaptor_id>2</adaptor_id>
<adaptor_type>STREAM</adaptor_type>
<iohandler_type>INPUT</iohandler_type>
<method>
<name>new</name>
<rank>1</rank>
<argument>
<value>$blast1_dbdir</value>
</argument>
</method>
<method>
<name>get_Seq_by_id</name>
<argument>
<value>!INPUT!</value>
<type>SCALAR</type>
<rank>1</rank>
</argument>
</method>
</iohandler>
Thus setup_genewise will create jobs for genewise with 2 inputs that
with a clone and protein id with iohandler_id 4 and 5
respectively. ONe possible problem that I see is that setup_genewise
was written specifically for working with ensembl slices and thus
creates inputs with arguments chr_name, chr_start, chr_end. This was
done so that we can subseq a dna sequence to run genewise in the
region of the query given some padding. Thus the input made was for an
iohandler that has three arguments.
If all you want to do is provide a clone object, you may need to modify
setup_genewise so that all it creates is an input that has the clone id
so that it
can be passed to your iohandler_id=4 which has only one argument.
Hope that will clear some things up.
shawn
On Friday, January 16, 2004, at 3:07PM, Tania Oh wrote:
>
> Hi all,
>
> am having some problems with getting biopipe to run genewise.
>
> I get the error from genewise that there is no protein sequence
> entered and target dna.
>
> Setup_genewise.pm seems to run okay with the input table getting
> filled up with the relevant datat.
>
> but where is it in biopipe that the data from the input table is
> retrieved to be sent to genewise? the object that is sent to the
> genewise runnable doesn't contain the query_pep or the target_dna
> required by genewise.
>
> attached is my xml stylesheet. Pls have a look and tell me if I
> missed out something?
>
> thanks in advance,
> tania
>
>
> <genebuild.xml>_______________________________________________
> bioperl-pipeline mailing list
> bioperl-pipeline at bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-pipeline
More information about the bioperl-pipeline
mailing list