Merge pull request #138 from andrewjpage/extract_id

Extract IDs from GFF file using Bio::Perl

dist.ini

@@ -1,5 +1,5 @@
 name    = Bio-Roary
-version = 2.2.6
+version = 2.3.0
 author  = Andrew J. Page <ap13@sanger.ac.uk>
 license = GPL_3
 copyright_holder = Wellcome Trust Sanger Institute

lib/Bio/Roary/BedFromGFFRole.pm

@@ -0,0 +1,54 @@
+package Bio::Roary::BedFromGFFRole;
+
+# ABSTRACT: A role to create a bed file from a gff
+
+=head1 SYNOPSIS
+
+ A role to create a bed file from a gff
+   with 'Bio::Roary::BedFromGFFRole';
+
+=cut
+
+use Moose::Role;
+use Bio::Tools::GFF;
+
+has '_tags_to_filter'   => ( is => 'ro', isa => 'Str', default => '(CDS|ncRNA|tRNA|tmRNA|rRNA)' );
+has 'min_gene_size_in_nucleotides'   => ( is => 'ro', isa => 'Int',  default  => 120 );
+
+sub _bed_output_filename {
+    my ($self) = @_;
+    return join( '.', ( $self->output_filename, 'intermediate.bed' ) );
+}
+
+sub _create_bed_file_from_gff {
+    my ($self) = @_;
+
+    open( my $bed_fh, '>', $self->_bed_output_filename );
+    my $gffio = Bio::Tools::GFF->new( -file => $self->gff_file, -gff_version => 3 );
+    while ( my $feature = $gffio->next_feature() ) {
+
+        next unless defined($feature);
+
+        # Only interested in a few tags
+        my $tags_regex = $self->_tags_to_filter;
+        next if !( $feature->primary_tag =~ /$tags_regex/ );
+
+        # Must have an ID tag
+        next unless ( $feature->has_tag('ID') );
+
+        #filter out small genes
+        next if ( ( $feature->end - $feature->start ) < $self->min_gene_size_in_nucleotides );
+
+        my ( $gene_id, @junk ) = $feature->get_tag_values('ID');
+        $gene_id =~ s!["']!!g;
+        next if ( $gene_id eq "" );
+
+        my $strand = ($feature->strand > 0)? '+':'-' ;
+        print {$bed_fh} join( "\t", ( $feature->seq_id, $feature->start -1, $feature->end, $gene_id, 1, $strand ) ) . "\n";
+    }
+    $gffio->close();
+}
+
+
+
+1;

lib/Bio/Roary/External/Blastp.pm

@@ -33,7 +33,7 @@ has '_num_threads'      => ( is => 'ro', isa => 'Int', default  => 1 );
 has '_max_target_seqs'  => ( is => 'ro', isa => 'Int', default  => 2000 );
 has '_logging'          => ( is => 'ro', isa => 'Str', default  => '2> /dev/null' );
 has 'output_file'       => ( is => 'ro', isa => 'Str', default  => 'results.out' );
-has 'perc_identity'    => ( is => 'ro', isa => 'Num', default  => 98 );
+has 'perc_identity'     => ( is => 'ro', isa => 'Num', default  => 98 );
 
 sub _command_to_run {
     my ($self) = @_;

lib/Bio/Roary/External/Makeblastdb.pm

@@ -27,7 +27,6 @@ use Cwd;
 with 'Bio::Roary::JobRunner::Role';
 
 has 'fasta_file'         => ( is => 'ro', isa => 'Str', required => 1 );
-has 'mask_data'          => ( is => 'ro', isa => 'Str', required => 1  );
 has 'exec'               => ( is => 'ro', isa => 'Str', default  => 'makeblastdb' );
 has '_working_directory' => ( is => 'ro', isa => 'File::Temp::Dir', default  => sub { File::Temp->newdir( DIR => getcwd, CLEANUP => 1 ); } );
 has '_dbtype'            => ( is => 'ro', isa => 'Str', default  => 'prot' );
@@ -51,7 +50,6 @@ sub _command_to_run {
             '-in',      $self->fasta_file,       
             '-dbtype',  $self->_dbtype, 
             '-parse_seqids',
-            '-mask_data', $self->mask_data,
             '-out',     $self->output_database, 
             '-logfile', $self->_logfile
         )