From Code to Community: Sponsoring The Perl and Raku Conference 2025 Learn more

# $Id: FTLocationFactory.pm,v 1.21.4.2 2006/10/02 23:10:18 sendu Exp $
#
# BioPerl module for Bio::Factory::FTLocationFactory
#
# Cared for by Hilmar Lapp <hlapp at gmx.net>
#
# Copyright Hilmar Lapp
#
# You may distribute this module under the same terms as perl itself
#
# (c) Hilmar Lapp, hlapp at gnf.org, 2002.
# (c) GNF, Genomics Institute of the Novartis Research Foundation, 2002.
#
# You may distribute this module under the same terms as perl itself.
# Refer to the Perl Artistic License (see the license accompanying this
# for the terms under which you may use, modify, and redistribute this module.
#
# THIS PACKAGE IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED
# WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
# MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
#
# POD documentation - main docs before the code
=head1 NAME
Bio::Factory::FTLocationFactory - A FeatureTable Location Parser
=head1 SYNOPSIS
# parse a string into a location object
$loc = Bio::Factory::FTLocationFactory->from_string("join(100..200,
400..500");
=head1 DESCRIPTION
Implementation of string-encoded location parsing for the Genbank feature
table encoding of locations.
=head1 FEEDBACK
=head2 Mailing Lists
User feedback is an integral part of the evolution of this and other
Bioperl modules. Send your comments and suggestions preferably to
the Bioperl mailing list. Your participation is much appreciated.
bioperl-l@bioperl.org - General discussion
http://bioperl.org/wiki/Mailing_lists - About the mailing lists
=head2 Reporting Bugs
Report bugs to the Bioperl bug tracking system to help us keep track
of the bugs and their resolution. Bug reports can be submitted via the
web:
=head1 AUTHOR - Hilmar Lapp
Email hlapp at gmx.net
=head1 CONTRIBUTORS
Jason Stajich, jason-at-bioperl-dot-org
Chris Fields, cjfields-at-uiuc-dot-edu
=head1 APPENDIX
The rest of the documentation details each of the object methods.
Internal methods are usually preceded with a _
=cut
# Let the code begin...
use vars qw($LOCREG);
use strict;
# Object preamble - inherits from Bio::Root::Root
use base qw(Bio::Root::Root Bio::Factory::LocationFactoryI);
BEGIN {
# the below is an optimized regex obj. from J. Freidl's Mastering Reg Exp.
$LOCREG = qr{
(?>
[^()]+
|
\(
(??{$LOCREG})
\)
)*
}x;
}
=head2 new
Title : new
Usage : my $obj = new Bio::Factory::FTLocationFactory();
Function: Builds a new Bio::Factory::FTLocationFactory object
Returns : an instance of Bio::Factory::FTLocationFactory
Args :
=cut
=head2 from_string
Title : from_string
Usage : $loc = $locfactory->from_string("100..200");
Function: Parses the given string and returns a Bio::LocationI implementing
object representing the location encoded by the string.
This implementation parses the Genbank feature table
encoding of locations.
Example :
Returns : A Bio::LocationI implementing object.
Args : A string.
=cut
sub from_string {
my ($self,$locstr,$op) = @_;
my $loc;
#$self->debug("$locstr\n");
# $op for operator (error handling)
# run on first pass only
# Note : These location types are now deprecated in GenBank (Oct. 2006)
if (!defined($op)) {
# convert all (X.Y) to [X.Y]
$locstr =~ s{\((\d+\.\d+)\)}{\[$1\]}g;
# convert ABC123:(X..Y) to ABC123:[X..Y]
# we should never see the above
$locstr =~ s{:\((\d+\.{2}\d+)\)}{:\[$1\]}g;
}
if ($locstr =~ m{(.*?)\(($LOCREG)\)(.*)}o) { # any matching parentheses?
my ($beg, $mid, $end) = ($1, $2, $3);
my (@sublocs) = (split(q(,),$beg), $mid, split(q(,),$end));
my @loc_objs;
my $loc_obj;
SUBLOCS:
while (@sublocs) {
my $subloc = shift @sublocs;
next if !$subloc;
my $oparg = ($subloc eq 'join' || $subloc eq 'bond' ||
$subloc eq 'order' || $subloc eq 'complement') ? $subloc : undef;
# has operator, requires further work (recurse)
if ($oparg) {
my $sub = shift @sublocs;
if (($oparg eq 'join' || $oparg eq 'order' || $oparg eq 'bond' )
&& $sub !~ m{$oparg}) {
my @splitlocs = split(q(,), $sub);
$loc_obj = Bio::Location::Split->new();
while (my $splitloc = shift @splitlocs) {
next unless $splitloc;
#$loc_obj->add_sub_Location($self->from_string($splitloc, 1));
# this should work but doesn't
my $sobj;
if ($splitloc =~ m{\(($LOCREG)\)}) {
my $comploc = $1;
$sobj = $self->_parse_location($comploc);
$sobj->strand(-1);
} else {
$sobj = $self->_parse_location($splitloc);
}
$loc_obj->add_sub_Location($sobj);
}
} else {
$loc_obj = $self->from_string($sub, $oparg);
}
}
# no operator, simple or fuzzy
else {
$loc_obj = $self->from_string($subloc,1);
}
$loc_obj->strand(-1) if ($op && $op eq 'complement');
push @loc_objs, $loc_obj;
}
my $ct = @loc_objs;
if ($op && !($op eq 'join' || $op eq 'order' || $op eq 'bond')
&& $ct > 1 ) {
$self->throw("Bad operator $op: had multiple locations ".
scalar(@loc_objs).", should be SplitLocationI");
}
if ($ct > 1) {
$loc = Bio::Location::Split->new();
$loc->add_sub_Location(shift @loc_objs) while (@loc_objs);
return $loc;
} else {
$loc = shift @loc_objs;
return $loc;
}
} else { # simple location(s)
$loc = $self->_parse_location($locstr);
$loc->strand(-1) if ($op && $op eq 'complement');
}
return $loc;
}
=head2 _parse_location
Title : _parse_location
Usage : $loc = $locfactory->_parse_location( $loc_string)
Function: Parses the given location string and returns a location object
with start() and end() and strand() set appropriately.
Note that this method is private.
Returns : A Bio::LocationI implementing object or undef on failure
Args : location string
=cut
sub _parse_location {
my ($self, $locstr) = @_;
my ($loc, $seqid);
#$self->debug( "Location parse, processing $locstr\n");
# 'remote' location?
if($locstr =~ m{^(\S+):(.*)$}o) {
# yes; memorize remote ID and strip from location string
$seqid = $1;
$locstr = $2;
}
# split into start and end
my ($start, $end) = split(/\.\./, $locstr);
# remove enclosing parentheses if any; note that because of parentheses
# possibly surrounding the entire location the parentheses around start
# and/or may be asymmetrical
# Note: these are from X.Y fuzzy locations, which are deprecated!
$start =~ s/(?:^\[+|\]+$)//g if $start;
$end =~ s/(?:^\[+|\]+$)//g if $end;
# Is this a simple (exact) or a fuzzy location? Simples have exact start
# and end, or is between two adjacent bases. Everything else is fuzzy.
my $loctype = ".."; # exact with start and end as default
$loctype = '?' if ( ($locstr =~ /\?/) && ($locstr !~ /\?\d+/) );
my $locclass = "Bio::Location::Simple";
if(! defined($end)) {
if($locstr =~ /(\d+)([\.\^])(\d+)/) {
$start = $1;
$end = $3;
$loctype = $2;
$locclass = "Bio::Location::Fuzzy"
unless (abs($end-$start) <= 1) && ($loctype eq "^");
} else {
$end = $start;
}
}
# start_num and end_num are for the numeric only versions of
# start and end so they can be compared
# in a few lines
my ($start_num, $end_num) = ($start,$end);
if ( ($start =~ /[\>\<\?\.\^]/) || ($end =~ /[\>\<\?\.\^]/) ) {
$locclass = 'Bio::Location::Fuzzy';
if($start =~ /(\d+)/) {
($start_num) = $1;
} else {
$start_num = 0
}
if ($end =~ /(\d+)/) {
($end_num) = $1;
} else { $end_num = 0 }
}
my $strand = 1;
if( $start_num > $end_num && $loctype ne '?') {
($start,$end,$strand) = ($end,$start,-1);
}
# instantiate location and initialize
$loc = $locclass->new(-verbose => $self->verbose,
-start => $start,
-end => $end,
-strand => $strand,
-location_type => $loctype);
# set remote ID if remote location
if($seqid) {
$loc->is_remote(1);
$loc->seq_id($seqid);
}
# done (hopefully)
return $loc;
}
1;