Bioperl: trying XML in biology...

[Lincoln Stein]

I've just put together an xml2boulder converter (65 lines of code).
Here's what's nice about it.

Say you have an XML file that looks like this (real life example):

<contig>
 <acc>NT_001817</acc>
 <name>Hs1_1938</name>
 <chr>1</chr>
 <labs>SC</labs>
 <partslist>
   <part>
    <cpos1>0</cpos1>
    <cpos2>41014</cpos2>
    <acc>AL049198</acc>
    <seqlen>41015</seqlen>
    <clone>590K14</clone>
    <ctype>pac</ctype>
   </part>
   <part>
    <cpos1>40915</cpos1>
    <cpos2>125326</cpos2>
    <acc>AL033533</acc>
    <seqlen>84412</seqlen>
    <clone>973M2</clone>
   </part>
</contig>

Then after converting it into boulder format, you can access the bits
and pieces as methods, like so:

    $xml = Boulder::XML->new($xmlfile);
    $chromosome = $xml->chr;
    @parts = $xml->partslist->parts;
    for my $part (@parts) {
       my $accession = $part->acc;
    }

I haven't written the corresponding boulder2xml converter, but if
people are interested, I'll send out what I've got now for beta
testing and experimentation.

Lincoln

Catherine Letondal writes:
 > 
 > Hi,
 > 
 > We have made some tries in XML here at Pasteur (thanks to an XML course 
 > by S. Bortzmeyer). 
 > 
 > The main idea is to first convince authors of biologicial software how XML
 > makes it easy to parse a program's output, and that producing
 > XML output is easy as well (well, as long as you are the author of the code).
 > 
 > That's why this page: http://www-alt.pasteur.fr/~letondal/XML/
 > shows examples of softs developped locally at Pasteur, to which we have
 > added an '-x' option in order to have them produce an XML output (CDS, satellites).
 > 
 > It shows also an XMLized fasta output with a parser, just to show
 > that a parser in XML is very easy to write, and how it could lower the 
 > difficulty of parsing this kind of texts.
 > 
 > We have also tried to see what a genbank entry would be like in XML. This is
 > naive of course, for only the authors of databanks may decide about this format, but
 > these are mainly examples. P. Bouige has also written a converter from
 > Genbank to XML, that works with the DTD we have set up.
 > 
 > We have made also simple examples of more general objects, like Phylip trees, 
 > but I guess that such kinds of objects will be discussed at the Bioperl 
 > workshop.
 > 
 > 
 > -- 
 > Catherine Letondal -- letondal@pasteur.fr -- +33 (1) 40 61 31 91
 > =========== Bioperl Project Mailing List Message Footer =======
 > Project URL: http://bio.perl.org/
 > For info about how to (un)subscribe, where messages are archived, etc:
 > http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
 > ====================================================================

-- 
========================================================================
Lincoln D. Stein                           Cold Spring Harbor Laboratory
lstein@cshl.org			                  Cold Spring Harbor, NY
========================================================================
=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org/
For info about how to (un)subscribe, where messages are archived, etc:
http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
====================================================================