view release on metacpan or  search on metacpan

ABI.pm  view on Meta::CPAN

#$Id: ABI.pm,v 1.3.2.4 2006/11/20 03:18:12 malay Exp $
# Perl module to parse ABI chromatogram file
# Malay <malay@bioinformatics.org>
# Copyright (c) 2002, 2003, 2004, 2005, 2006 Malay Kumar Basu
# You may distribute this module under the same terms as perl itself
# Thanks to David H. Klatte for all the hard work!
package ABI;
use IO::File;
use Carp;
use strict;

=head1 NAME

ABI.pm - Perl module to parse chromatogram files generated by
Applied Biosystems (ABI) automated DNA sequencing machine.

Please cite:
ABI.pm - Perl module to parse chromatogram files generated by
Applied Biosystems (ABI) automated DNA sequencing machine.
by Malay K Basu (malay@bioinformatics.org); source code available at:
http://search.cpan.org/~malay

=head1 VERSION

Version 1.0

=cut

our $VERSION = '1.0';

=head1 SYNOPSIS

  my $abi = ABI->new(-file=>"mysequence.abi");
  my $seq = $abi->get_sequence(); # To get the sequence
  my @trace_a = $abi->get_trace("A"); # Get the raw traces for "A"
  my @trace_g = $abi->get_trace("G"); # Get the raw traces for "G"
  my @base_calls = $abi->get_base_calls(); # Get the base calls

=head1 DESCRIPTION

An ABI chromatogram file is in binary format. It contain several 
information only some of which is required for normal use. This
module only gives access to the most used information stored in
ABI file. All the accesses are read only.

If you have edited the file using a trace editor, then you can use the corresponding 
method to access the edited sequence and base calls.



=head1 CONSTRUCTOR

=head2 new()

  Usage : $abi = ABI->new(-file=>"filename");
          $abi = ABI->new("filename"); # same thing

=cut

sub new {
	my $class = shift;
	my $self  = {};
	bless $self, ref($class) || $class;
	$self->_init(@_);

	#print "****", $self->{_mac_header}, "\n";
	return $self;
}

sub _init {
	my ( $self, @args ) = @_;
	my ($file) = $self->_rearrange( ["FILE"], @args );
	if ( !defined($file) ) {
		croak "Can't open the input file\n";
	} else {
		$self->set_file_handle($file);
	}
	$self->{_sequence}             = "";
	$self->{_sequence_corrected}   = "";
	$self->{_sample}               = "";
	$self->{A}                     = [];
	$self->{T}                     = [];
	$self->{G}                     = [];
	$self->{C}                     = [];
	$self->{_basecalls}            = [];
	$self->{_basecalls_corrected}  = [];
	$self->{_trace_length}         = 0;
	$self->{_seq_length}           = 0;
	$self->{_seq_length_corrected} = 0;
	$self->{_abs_index}            = 26;
	$self->{_index}                = undef;
	$self->{PLOC1}                 = undef;
	$self->{PLOC}                  = undef;
	$self->{_a_start}              = undef;
	$self->{_g_start}              = undef;
	$self->{_c_start}              = undef;
	$self->{_t_start}              = undef;
	$self->{DATA9}                 = undef;
	$self->{DATA10}                = undef;
	$self->{DATA11}                = undef;
	$self->{DATA12}                = undef;
	$self->{PBAS1}                 = undef;
	$self->{PBAS2}                 = undef;
	$self->{FWO}                   = undef;
	$self->{_mac_header}           = 0;
	$self->{_maximum_trace}        = 0;

ABI.pm  view on Meta::CPAN

}

sub _set_sample_name {
	my $self = shift;
	my $buf;
	my $fh = $self->{_fh};
	seek( $fh, $self->{SMPL}, 0 );
	read( $fh, $buf, 1 );
	my $length = unpack( "C", $buf );
	read( $fh, $buf, $length );
	$self->{_sample} = $buf;
}

=head1 METHODS

=head2 get_max_trace()

  Title    :  get_max_trace()
  Usage    :  $max = $abi->get_max_trace();
  Function :  Returns the maximum trace value of all the traces.
  Args     :  Nothing
  Returns  :  A scalar

=cut

sub get_max_trace {
	my $self = shift;
	return $self->{_maximum_trace};
}

=head2 get_trace()

  Title    :  get_trace()
  Usage    :  my @a = $abi->get_trace("A");
  Function :  Returns the raw traces as array.
  Args     :  "A" or "G" or "C" or "T"
  Returns  :  An array

=cut

sub get_trace {
	my $self   = shift;
	my $symbol = shift;
	if ( $symbol =~ /A/i ) {
		return @{ $self->{A} };
	} elsif ( $symbol =~ /G/i ) {
		return @{ $self->{G} };
	} elsif ( $symbol =~ /C/i ) {
		return @{ $self->{C} };
	} elsif ( $symbol =~ /T/i ) {
		return @{ $self->{T} };
	} else {
		croak "Illegal symbol\n";
	}
}

=head2 get_sequence()

  Title    : get_sequence()
  Usage    : my $seq = $abi->get_sequence();
  Function : Returns the original unedited sequence as string. If you want to access the edited
             sequence use "get_corrected_sequence()" instead.
  Args     : Nothing
  Returns  : A scalar

=cut

sub get_sequence {
	my $self = shift;
	return $self->{_sequence};
}

=head2 get_corrected_sequence()

  Title    : get_corrected_sequence()
  Usage    : my $seq = $abi->get_corrected_sequence();
  Function : Returns the corrected sequence as string. If you want to access the original 
             unedited sequence, use "get_sequence()" instead.
  Args     : Nothing
  Returns  : A scalar

=cut

sub get_corrected_sequence {
	my $self = shift;
	return $self->{_sequence_corrected};
}

=head2 get_sequence_length()

  Title    : get_sequence_length()
  Usage    : my $seq_length = $abi->get_sequence_length();
  Function : Returns the sequence length of the orginal unedited sequence.
  Args     : Nothing
  Returns  : A scalar

=cut

sub get_sequence_length {
	my $self = shift;
	return $self->{_seq_length};
}

=head2 get_corrected_sequence_length()

  Title    : get_corrected_sequence_length()
  Usage    : my $seq_length = $abi->get_corrected_sequence_length();
  Function : Returns the length of the edited sequence. 
  Args     : Nothing
  Returns  : A scalar

=cut

sub get_corrected_sequence_length {
	my $self = shift;

	#print STDERR "**ABI**",$self->{_seq_length_corrected},"\n";
	return $self->{_seq_length_corrected};
}

=head2 get_trace_length()

  Title    : get_trace_length()
  Usage    : my $trace_length = $abi->get_trace_length();
  Function : Returns the trace length
  Args     : Nothing
  Returns  : A scalar

=cut

sub get_trace_length {
	my $self = shift;
	return $self->{_trace_length};
}

=head2 get_base_calls()

  Title    : get_base_calls()
  Usage    : my @base_calls = $abi->get_base_calls();
  Function : Returns the called bases by the base caller. This method will return the unedited 
  	         original basecalls created by the basecaller.
  Args     : Nothing
  Returns  : An array

=cut

sub get_base_calls {
	my $self = shift;
	return @{ $self->{_basecalls} };
}

=head2 get_corrected_base_calls()

  Title    : get_corrected_base_calls()
  Usage    : my @base_calls = $abi->get_corrected_base_calls();
  Function : If you have edited the trace file you can get the corrected base call 
             with this method
  Args     : Nothing
  Returns  : An array

=cut

sub get_corrected_base_calls {
	my $self = shift;
	return @{ $self->{_basecalls_corrected} };
}

=head2 get_sample_name()

  Title    : get_sample_name()
  Usage    : my $sample = $abi->get_sample_name();
  Function : Returns hard coded sample name
  Args     : Nothing
  Returns  : A scalar

=cut

sub get_sample_name {
	my $self = shift;
	return $self->{_sample};
}

=head1 AUTHOR

Malay <malay@bioinformatics.org>

=head1 BUGS

Please report any bugs or feature requests to
C<bug-abi at rt.cpan.org>, or through the web interface at
L<http://rt.cpan.org/NoAuth/ReportBug.html?Queue=ABI>.
I will be notified, and then you'll automatically be notified of progress on
your bug as I make changes.

or 

You can directly contact me to my email address.


=head1 SUPPORT

You can find documentation for this module with the perldoc command.

    perldoc ABI

You can also look for information at:

=over 4

=item * AnnoCPAN: Annotated CPAN documentation

L<http://annocpan.org/dist/ABI>

=item * CPAN Ratings

L<http://cpanratings.perl.org/d/ABI>