This short HowTo guides you from downloading obogaf::parser to make your first parsing with obogaf::parser.


Please goto the Installation section and chose one of the shown ways to install obogaf::parser (we suggest to use the Installation via Conda option).

Load obogaf::parser library

To load obogaf::parser module, just type inside a Perl script use obogaf::parser. More precisely, the header of your Perl script should be:


use strict;
use warnings;

use obogaf::parser;

... beginning of your perl code ...

To run the Perl script you can make it executable by typing chmod +x perl-script.pl or by prefacing the script with the Perl interpreter (perl perl-script.pl).

Your first parsing

Let us use obogaf::parser to extract edges (in the form source-destination) from the Gene Ontology (GO) obo file. Firstly we must download the GO obo file from the Gene Ontoloy website. Then we use obogaf::parser to extract edges.

## perl shebang (unix)

## load the module
use obogaf::parser;

## download GO obo file
my $obofile= "gobasic.obo";
my $gobo= qx{wget --output-document=$obofile http://purl.obolibrary.org/obo/go/go-basic.obo};
print "GO obo file downloaded: done\n\n";

## extract and print GO edges
my $gores= obogaf::parser::build_edges($obofile);
print "${$gores}";

obogaf::parser can do much more than that! Go to the Tutorial section to discover what this module can do! But first get a look to Installation section …