[Bioperl-pipeline] Genewise matters(start,stop modification)

[Shawn]

I'm currently working on the DataMonger for genewise. 
A solution for coordinate mapping.

Have the Genewise runnable take in a Location object in addition to the 
subseq virtual contig. Then it will do the mapping (offset) for the
predicted features. The way of passing in dynamic parameters into
runnable is now done through the dynamic_argument table (creative I
know). For a Location object, we will use the SteamAdaptor table. 

Currently, each job is now associated with static arguments(found in
argument table, i.e. know before run time) and generated arguments in
the dynamic arguments). The arguments are then merged during fetch input
based on rank with precedence of rank given to static arguments.

basically 
static s1,s3,s4
dyanmic d1,d2

final arglist
s1,d1,s3,d2,s4


prolly makes sense only for kiran and elia but in anycase, we will need
to code review this. 


New Objects:
Bio::Pipeline::Filter::feature_coverage (does the max_coverage
calculations and selects the best features)

Bio::Pipeline::InputCreate::setup_genewise (setups input,job,dynamic
argument tables etc)

Bio::Pipeline::Runnable::DataMonger (contains filter and inputcreate)


I'm currently testing and trying to get it all to work.
Question, for ciona, what is the method call for getting the contig
slice?

fetch_VirtualContig_by_chr_start_end

we need a static_golden_path for this I reckon?


the other thing to be done is implementing the converter functionality
tied with IOHandlers.



shawn








The simple way would be like Bala said (if I understand it properly)
is to pass into the genewise genewise


On Tue, 2002-09-10 at 01:53, Bala wrote:
> Hi Elia,
> 
> 
> >
> > No, meaning that when you are finished buildign the genes, and you wan tto
> > store them you can't store them as they are because you have passed ina
> > piece of the contig (start,end,strand) so you n eed to convert the
> > coordinates to the original ones before stroing it.
> 
> I have an idea, to solve the above issue...(hope its not a hack)
> 
> In the Genewise wrapper right,
> 
> In the  _run subroutine ,
> 
> my $gene = $genewiseParser->next_prediction($filehandle);
> 
> Hence once we get the $gene object,
> 
> We add the original start stop of the genomic and protein to
> the current start stop of the gene,exon and the
> supporting features(in this sub routine itself or another internal function
> eg  _modify_genes(@genes)).
> 
> genomic_new start = the current start + $self->_subject_dna_seq->start
> genomic_new end  = the current start +$self->_subject_dna_seq->end
> 
> protein_new_start= the current start + $self->_query_pep_seq->start
> protein_new end   = the current start + $self->_query_pep_seq->end
> 
> and then return the modified genes.
> 
> please comment
> 
> 
> Thank you
> 
> Bala
> 
> 
> 
> 
> _______________________________________________
> bioperl-pipeline mailing list
> bioperl-pipeline@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-pipeline
>