[Biojava-l] make feature to create embl or genbank file

[George Waldon]

Hi Hedwig,

The problem holds with StrandedFeature. The strandeness of a feature  
is the transdeness of its location. StrandedFeature should be  
eliminated from bj1. Use biojavaX instead, something like this, once  
you have created a RichLocation on the appropriate strand:

public Feature.Template getFeatureTemplate(RichSequence  
parent,RichLocation loc) {
         RichFeature.Template templ = new RichFeature.Template();
         RichAnnotation rans = new SimpleRichAnnotation();
	templ.annotation = rans;
         templ.sourceTerm = // find an appropriate term
         templ.typeTerm =  
RichObjectFactory.getDefaultOntology().getOrCreateTerm("CDS");
         templ.featureRelationshipSet = new TreeSet();
         templ.rankedCrossRefs = new TreeSet();
         templ.location = loc;

         // add notes if any you'd like

         return templ;
}

That should make it into the output file.

Regards,
George


Quoting Hedwig Kurka <kurka at mikro.biologie.tu-muenchen.de>:

> Hello all,
>
> I have a problem concerning creating EMBL or Genbank files.
> Below is a fragment of my code and an example of how the EMBL file looks
> like.
>
>        String name = "test genome";
>        String seqString = pFasta.getSequence(1, pFasta.getLength());
>        Sequence seq = DNATools.createDNASequence(seqString, name);
>        Alphabet dna =  AlphabetManager.alphabetForName("DNA");
>        RichSequence rs =
> Tools.createRichSequence(RichObjectFactory.getDefaultNamespace(), name,
> seqString, dna);
>        Set<Feature> rfeatSet = new HashSet<Feature>();
>        StrandedFeature.Template t = new StrandedFeature.Template();
>        for(int i=0; i<annotierten.size(); i++){
>                    int start = (int) Math.abs(anno.get(i).getStart());
>                    int stop = (int) Math.abs(anno.get(i).getStop());
>                    t.type = "CDS";
>                    if(start < stop){
>                        t.location = new RangeLocation(start, stop);
>                        t.strand = StrandedFeature.POSITIVE;
>                    }
>                    if(start > stop){
>                        t.location = new RangeLocation(stop, start);
>                        t.strand = StrandedFeature.NEGATIVE;
>                    }
>                    Feature f = seq.createFeature(t);
>                    RichFeature rf = RichFeature.Tools.enrich(f);
>                    rfeatSet.add(rf);
>        }
>        rs.setFeatureSet(rfeatSet);
>        rs = RichSequence.Tools.enrich(rs);
>        RichSequence.IOTools.writeEMBL(output, rs,
> RichObjectFactory.getDefaultNamespace());
>
> EMBL file:
> FT   any             1889536..1890903
> FT   any             134636..136987
> FT   any             3727110..3727625
> FT   any             2812636..2813517
> FT   any             580648..581643
> FT   any             2330962..2331921
> FT   any             1012371..1013513
> FT   any             1260854..1261720
> FT   any             1602858..1603706
> FT   any             4108079..4108999
> FT   any             346637..347731
> FT   any             4073395..4074549
>
> I wonder where the information of plus and minus strand is, why is there
> "any" in the file and not "CDS" and so on.
>
> As tutorial I found that:
> http://www.biojava.org/wiki/BioJava:Cookbook:Locations:Feature. Is there
> another one?
>
> Thank you for your help!
>
> And any help is appreciated,
>
> Hedwig
>
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-l
>