The Perl and Raku Conference 2025: Greenville, South Carolina - June 27-29 Learn more

# $Id: mast.pm,v 1.21.4.2 2006/10/02 23:10:22 sendu Exp $
=head1 NAME
Bio::Matrix::PSM::IO::mast - PSM mast parser implementation
=head1 SYNOPSIS
See Bio::Matrix::PSM::IO for detailed documentation on how to
use PSM parsers
=head1 DESCRIPTION
Parser for mast. This driver unlike meme or transfac for example is
dedicated more to PSM sequence matches, than to PSM themselves.
=head1 TO DO
Section III should be parsed too, otherwise no real sequence is
available, so we supply 'NNNNN....' as a seq which is not right.
=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 one of the Bioperl mailing lists. 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
the bugs and their resolution. Bug reports can be submitted via the
web:
=head1 AUTHOR - Stefan Kirov
Email skirov@utk.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 strict;
use base qw(Bio::Matrix::PSM::PsmHeader Bio::Matrix::PSM::IO);
=head2 new
Title : new
Usage : my $psmIO = new Bio::Matrix::PSM::IO(-format=>'mast',
-file=>$file);
Function: Associates a file with the appropriate parser
Throws : Throws if the file passed is in HTML format or if
some criteria for the file
format are not met.
Example :
Returns : psm object, associated with a file with matrix file
Args : hash
return : "Bio::Matrix::PSM::$format"->new(@args);
=cut
sub new {
my($class, @args)=@_;
my $self = $class->SUPER::new(@args);
my (%instances,@header,$n);
my ($file)=$self->_rearrange(['FILE'], @args);
$self->{file} = $file;
$self->{_factor}=1;
$self->_initialize_io(@args) || warn "Did you intend to use STDIN?"; #Read only for now
$self->{_end}=0;
undef $self->{hid};
return $self if ($file=~/^>/);#Just writing
my $buf=$self->_readline;
$self->throw('Cannot parse HTML format yet') if ($buf =~/^<HTML>/);
# this should probably be moved to its own function
while ( defined($buf=$self->_readline)) {
chomp($buf);
if ($buf=~/DATABASE AND MOTIFS/) {
while ($buf=$self->_readline) {
if ($buf=~/DATABASE/) {
$buf=~s/^[\s\t]+//;
chomp $buf;
($n,$self->{_dbname},$self->{_dbtype})=split(/\s/,$buf);
$self->{_dbtype}=~s/[\(\)]//g;
}
if ($buf=~/MOTIFS/) {
$buf=~s/^[\s\t]+//;
chomp $buf;
($n,$self->{_mrsc},$self->{_msrctype})=split(/\s/,$buf);
$self->{_msrctype}=~s/[\(\)]//g;
last;
}
}
if ($self->{_msrctype} ne $self->{_dbtype}) {#Assume we have protein motifs, nuc DB (not handling opp.)
$self->{_factor}=3;
$self->{_mixquery}=1;
}
}
if ($buf=~m/MOTIF WIDTH BEST POSSIBLE MATCH/) {
$self->_readline;
while (defined($buf=$self->_readline)) {
last if ($buf!~/\w/);
$buf=~s/\t+//g;
$buf=~s/^\s+//g;
my ($id,$width,$seq)=split(/\s+/,$buf);
push @{$self->{hid}},$id;
$self->{length}->{$id}=$width;
$self->{seq}->{$id}=$seq;
}
next;
}
if ($buf=~m/section i:/i) {
$self->_readline;
$self->_readline;
$self->_readline;
%instances=_get_genes($self);
$self->{instances}=\%instances;
if (!(%instances)) {
$self->warn ("Your MAST analysis did not find any matches satisfying the current threshold.\nSee MAST documentation for more information.\n");
return $self; #The header might be useful so we return the object, not undef
}
next;
}
if ($buf=~m/section ii:/i) {
$self->_readline;
$self->_readline;
$self->_readline;
last;
}
$buf=~s/[\t+\s+]/ /g;
push @header,$buf unless (($buf=~/\*{10,}/)||($buf!~/\w/));
}
$self->throw('Could not read Section I, probably wrong format, make sure it is not HTML, giving up...') if !(%instances);
$self->warn( "This file might be an unreadable version, proceed with caution!\n") if (!grep(/\s+MAST\s+version\s+3/,@header));
$self->{unstructured} = \@header;
$self->_initialize;
return $self;
}
# Get the file header and put store it as a hash, which later we'll use to create
# the header for each Psm. See Bio::Matrix::PSM::PsmI for header function.
sub _get_genes {
my $self=shift;
my %llid;
my $ok=0;
my $i=0;
my %instances;
while (my $line=$self->_readline) {
last if ($line=~/^[\s\t*]/); # Well, ids can be nearly anything...???
chomp($line);
$i++;
next if ($line eq '');
$line=~s/\s+/,/g;
my ($id,$key,$eval,$len)=split(/,/,$line);
unless ($len) {
warn "Malformed data found: $line\n";
next;
}
$instances{$id}=new Bio::Matrix::PSM::InstanceSite (-id=>$id,
-desc=>$key,
-score=>$eval,
-width=>$len,
-seq=>'ACGT');
}
return %instances;
}
=head2 next_psm
Title : next_psm
Usage : my $psm=$psmIO->next_psm();
Function: Reads the next PSM from the input file, associated with this object
Throws : Throws if there ara format violations in the input file (checking is not
very strict with all drivers).
Example :
Returns : Bio::Matrix::PSM::Psm object
Args : none
=cut
sub next_psm {
my $self=shift;
return if ($self->{_end}==1);
my (@lmotifsm,%index,$eval,$scheme,$sid);
%index= %{$self->{length}};
my (@instances,%instances);
my $line=$self->_readline;
$line=~s/[\t\n]//;
if ($line =~ /\*{10,}/) { #Endo of Section II if we do only section II
$self->{_end}=1;
return ;
}
do {
if ($line!~/^\s/) {
($sid,$eval,$scheme)=split(/\s+/,$line,3);
}
else
{ $scheme .=$line; }
$line=$self->_readline;
$line=~s/[\t\n]//;
} until ($line!~/^\s/);
my $pos=1;
$scheme=~s/\s+//g;
$scheme=~s/\n//g;
my @motifs=split(/_/,$scheme);
while (@motifs) {
my $next=shift(@motifs);
if (!($next=~/\D/)) {
last if (!@motifs);
$pos+=$next;
next;
}
my $id=$next;
my $score= $id=~m/\[/ ? 'strong' : 'weak' ;
my $frame;
my $strand = $id =~ m/\-\d/ ? -1 : 1 ;
if ($self->{_mixquery}) {
$frame = 0 if $id =~ m/\d+a/ ;
$frame = 1 if $id =~ m/\d+b/ ;
$frame = 2 if $id =~ m/\d+c/ ;
}
$id=~s/\D+//g;
my @s;
my $width=$index{$id};
#We don't know the sequence, but we know the length
my $seq='N' x ($width*$self->{_factor}); #Future version will have to parse Section tree nad get the real seq
my $instance=new Bio::Matrix::PSM::InstanceSite
( -id=>"$id\@$sid",
-mid=>$id,
-accession_number=>$sid,
-desc=>"Motif $id occurrance in $sid",
-score=>$score,
-seq=>$seq,
-alphabet => 'dna',
-start=>$pos,
-strand=>$strand);
$instance->frame($frame) if ($self->{_mixquery});
push @instances,$instance;
$pos+=$index{$id}*$self->{_factor};
}
my $psm= new Bio::Matrix::PSM::Psm (-instances=> \@instances,
-e_val => $eval,
-id => $sid);
$self->_pushback($line);
return $psm;
}
=head2 write_psm
Title : write_psm
Usage : #Get SiteMatrix object somehow (see Bio::Matrix::PSM::SiteMatrix)
my $matrix=$psmin->next_matrix;
#Create the stream
my $psmio=new(-file=>">psms.mast",-format=>'mast');
$psmio->write_psm($matrix);
#Will warn if only PFM data is contained in $matrix, recalculate the PWM
#based on normal distribution (A=>0.25, C=>0.25, etc)
Function: writes pwm in mast format
Throws :
Example :
Args : SiteMatrix object
Returns :
=cut
sub write_psm {
my ($self,$matrix)=@_;
# my $idline=">". $matrix->id . "\n";
my $w=$matrix->width;
my $header="ALPHABET= ACGT\nlog-odds matrix: alength= 4 w= $w\n";
$self->_print($header);
unless ($matrix->get_logs_array('A')) {
warn "No log-odds data, available, using normal distribution to recalculate the PWM";
$matrix->calc_weight({A=>0.25, C=>0.25, G=>0.25,T=>0.25});
}
while (my %h=$matrix->next_pos) {
$self->_print (join("\t",$h{lA},$h{lC},$h{lG},$h{lT},"\n"));
}
}
1;