run mcl and inflate results

dist.ini

@@ -14,6 +14,8 @@ repository.type = git
 requires = blastp
 requires = makeblastdb
 requires = cd-hit
+requires = mcl
+requires = mcxdeblast
 
 [@Basic]
 [PruneCruft]

lib/Bio/PanGenome.pm

@@ -1,8 +1,77 @@
 package Bio::PanGenome;
+
+# ABSTRACT: Create a pan genome
+
+=head1 SYNOPSIS
+
+Create a pan genome
+
+=cut
+
 use Moose;
-# ABSTRACT: Pan Genomes
+use Bio::PanGenome::ParallelAllAgainstAllBlast;
+use Bio::PanGenome::CombinedProteome;
+use Bio::PanGenome::External::Cdhit;
+use Bio::PanGenome::External::Mcl;
+use Bio::PanGenome::InflateClusters;
+
+has 'fasta_files'      => ( is => 'rw', isa => 'ArrayRef' );
+has 'output_filename'  => ( is => 'rw', isa => 'Str', default => 'clustered_proteins' );
+has 'job_runner'       => ( is => 'rw', isa => 'Str', default => 'LSF' );
+has 'makeblastdb_exec' => ( is => 'rw', isa => 'Str', default => 'makeblastdb' );
+has 'blastp_exec'      => ( is => 'rw', isa => 'Str', default => 'blastp' );
+has 'mcxdeblast_exec'  => ( is => 'ro', isa => 'Str', default => 'mcxdeblast' );
+has 'mcl_exec'         => ( is => 'ro', isa => 'Str', default => 'mcl' );
+
+sub run {
+    my ($self) = @_;
+
+    my $output_combined_filename      = 'combined_files.faa';
+    my $output_cd_hit_filename        = 'clustered.faa';
+    my $output_blast_results_filename = 'blast_results';
+    my $output_mcl_filename           = 'uninflated_mcl_groups';
+
+    my $combine_fasta_files = Bio::PanGenome::CombinedProteome->new(
+        proteome_files        => $self->fasta_files,
+        output_filename       => $output_combined_filename,
+        apply_unknowns_filter => 1
+    );
+    $combine_fasta_files->create_combined_proteome_file;
+
+    my $cdhit_obj = Bio::PanGenome::External::Cdhit->new(
+        input_file  => $output_combined_filename,
+        output_base => $output_cd_hit_filename
+    );
+    $cdhit_obj->run();
+
+    my $blast_obj = Bio::PanGenome::ParallelAllAgainstAllBlast->new(
+        fasta_file              => $output_cd_hit_filename,
+        blast_results_file_name => $output_blast_results_filename,
+        job_runner              => $self->job_runner,
+        makeblastdb_exec        => $self->makeblastdb_exec,
+        blastp_exec             => $self->blastp_exec
+    );
+    $blast_obj->run();
+
+    my $mcl = Bio::PanGenome::External::Mcl->new(
+        blast_results   => $output_blast_results_filename,
+        mcxdeblast_exec => $self->mcxdeblast_exec,
+        mcl_exec        => $self->mcl_exec,
+        output_file     => $output_mcl_filename
+    );
+    $mcl->run();
+
+    my $inflate_clusters = Bio::PanGenome::InflateClusters->new(
+      clusters_filename  => $cdhit_obj->clusters_filename,
+      mcl_filename       => $output_mcl_filename,
+      output_file        => $self->output_filename
+    );
+    $inflate_clusters->inflate();
+
+    # Cleanup files
+}
 
 no Moose;
 __PACKAGE__->meta->make_immutable;
 
-1;
+1;

lib/Bio/PanGenome/CommandLine/CreatePanGenome.pm

@@ -1,47 +1,59 @@
 package Bio::PanGenome::CommandLine::CreatePanGenome;
 
-# ABSTRACT: Create a pan genome from a set of proteins in a FASTA file
+# ABSTRACT: Take in FASTA files of proteins and cluster them
 
 =head1 SYNOPSIS
 
-Create a pan genome from a set of proteins in a FASTA file
+Take in FASTA files of proteins and cluster them
 
 =cut
 
 use Moose;
 use Getopt::Long qw(GetOptionsFromArray);
-use Bio::PanGenome::CombinedProteome;
-use Bio::PanGenome::External::Cdhit;
-use Bio::PanGenome::External::Makeblastdb;
-use Bio::PanGenome::External::Blastp;
-use Bio::PanGenome::GGFile;
+use Bio::PanGenome;
 
-has 'args'        => ( is => 'ro', isa => 'ArrayRef', required => 1 );
-has 'script_name' => ( is => 'ro', isa => 'Str',      required => 1 );
-has 'help'        => ( is => 'rw', isa => 'Bool',     default  => 0 );
 
-has 'fasta_files'     => ( is => 'rw', isa => 'ArrayRef' );
-has 'output_filename' => ( is => 'rw', isa => 'Str' );
+has 'args'              => ( is => 'ro', isa => 'ArrayRef', required => 1 );
+has 'script_name'       => ( is => 'ro', isa => 'Str',      required => 1 );
+has 'help'              => ( is => 'rw', isa => 'Bool',     default  => 0 );
 
-has '_error_message' => ( is => 'rw', isa => 'Str' );
+has 'fasta_files'       => ( is => 'rw', isa => 'ArrayRef' );
+has 'output_filename'   => ( is => 'rw', isa => 'Str', default => 'clustered_proteins' );
+has 'job_runner'        => ( is => 'rw', isa => 'Str', default => 'LSF' );
+has 'makeblastdb_exec'  => ( is => 'rw', isa => 'Str', default => 'makeblastdb' );
+has 'blastp_exec'       => ( is => 'rw', isa => 'Str', default => 'blastp' );
+has 'mcxdeblast_exec'   => ( is => 'ro', isa => 'Str', default => 'mcxdeblast' );
+has 'mcl_exec'          => ( is => 'ro', isa => 'Str', default => 'mcl' );
+
+has '_error_message'    => ( is => 'rw', isa => 'Str' );
 
 sub BUILD {
     my ($self) = @_;
 
-    my ( $fasta_files, $output_filename, $help );
+    my ( $fasta_files, $output_filename, $job_runner, $makeblastdb_exec,$mcxdeblast_exec,$mcl_exec, $blastp_exec, $help );
 
     GetOptionsFromArray(
         $self->args,
-        'o|output=s' => \$output_filename,
-        'h|help'     => \$help,
+        'o|output=s'           => \$output_filename,
+        'j|job_runner=s'       => \$job_runner,
+        'm|makeblastdb_exec=s' => \$makeblastdb_exec,
+        'b|blastp_exec=s'      => \$blastp_exec,
+        'd|mcxdeblast_exec'    => \$mcxdeblast_exec,
+        'c|mcl_exec'           => \$mcl_exec,
+        'h|help'               => \$help,
     );
-
-    $self->output_filename($output_filename) if ( defined($output_filename) );
-
+
     if ( @{ $self->args } == 0 ) {
-        $self->_error_message("Error: You need to provide at least 1 FASTA file");
+        $self->_error_message("Error: You need to provide a FASTA file");
     }
 
+    $self->output_filename($output_filename)   if ( defined($output_filename) );
+    $self->job_runner($job_runner)             if ( defined($job_runner) );
+    $self->makeblastdb_exec($makeblastdb_exec) if ( defined($makeblastdb_exec) );
+    $self->blastp_exec($blastp_exec)           if ( defined($blastp_exec) );
+    $self->mcxdeblast_exec($mcxdeblast_exec)   if ( defined($mcxdeblast_exec) );
+    $self->mcl_exec($mcl_exec)                 if ( defined($mcl_exec) );
+
     for my $filename ( @{ $self->args } ) {
         if ( !-e $filename ) {
             $self->_error_message("Error: Cant access file $filename");
@@ -60,46 +72,29 @@ sub run {
         print $self->_error_message . "\n";
         die $self->usage_text;
     }
-
-    my $combined_proteome_obj = Bio::PanGenome::CombinedProteome->new(
-          proteome_files  => $self->fasta_files,
-          output_filename => 'combined_proteome.faa'
+
+    my $pan_genome_obj = Bio::PanGenome->new(
+        fasta_files      => $self->fasta_files,
+        output_filename  => $self->output_filename,
+        job_runner       => $self->job_runner,
+        makeblastdb_exec => $self->makeblastdb_exec,
+        blastp_exec      => $self->blastp_exec
       );
-   $combined_proteome_obj->create_combined_proteome_file;
-   print "Created combined file:\n";
-   my $percentage_sequences_ignored = (($combined_proteome_obj->number_of_sequences_ignored/$combined_proteome_obj->number_of_sequences_seen)*100);
-   print $percentage_sequences_ignored."  percent of sequences ignored\n";
-
-   print "Clustering the data:\n";
-   my $cdhit_obj = Bio::PanGenome::External::Cdhit->new( input_file   => 'combined_proteome.faa', output_base  => 'clustered.faa');
-   $cdhit_obj->run();
-
-   print "Creating a blast database:\n";
-   my $blast_database= Bio::PanGenome::External::Makeblastdb->new(fasta_file => 'clustered.faa');
-   $blast_database->run();
-
-   print "Blasting all against all:\n";
-   my $blastp_obj =  Bio::PanGenome::External::Blastp->new(
-     fasta_file     => 'clustered.faa',
-     blast_database => $blast_database->output_database,
-     output_file    => 'results.out'
-   );
-   $blastp_obj->run();
-
+    $pan_genome_obj->run();
 }
 
 sub usage_text {
     my ($self) = @_;
 
     return <<USAGE;
     Usage: create_pan_geneome [options]
-    Create a pan genome from a set of proteins in a FASTA file
-
-    # Create a pan genome from some FASTA files
-    create_pan_geneome *.faa
+    Take in FASTA files of proteins and cluster them
+    
+    # Take in FASTA files of proteins and cluster them
+    create_pan_geneome example.faa
 
     # Provide an output filename
-    create_pan_geneome -o outputfile.faa *.faa
+    create_pan_geneome -o results *.faa
 
     # This help message
     create_pan_geneome -h

lib/Bio/PanGenome/External/Cdhit.pm

@@ -28,6 +28,12 @@ has '_length_difference_cutoff'    => ( is => 'ro', isa => 'Num',  default  => 0
 has '_sequence_identity_threshold' => ( is => 'ro', isa => 'Num',  default  => 0.99 );
 has '_logging'          => ( is => 'ro', isa => 'Str', default  => '2> /dev/null' );
 
+sub clusters_filename
+{
+  my ($self) = @_;
+  return join('.',($self->output_base,'clstr'));
+}
+
 sub _command_to_run {
     my ($self) = @_;
     return join(

lib/Bio/PanGenome/External/Mcl.pm

@@ -0,0 +1,54 @@
+package Bio::PanGenome::External::Mcl;
+
+# ABSTRACT: Wrapper around MCL which takes in blast results and outputs clustered results
+
+=head1 SYNOPSIS
+
+Wrapper around MCL which takes in blast results and outputs clustered results
+
+   use Bio::PanGenome::External::Mcl;
+
+   my $mcl= Bio::PanGenome::External::Mcl->new(
+     blast_results     => 'db',
+     mcxdeblast_exec   => 'mcxdeblast',
+     mcl_exec          => 'mcl',
+     output_file       => 'output.groups'
+   );
+
+   $mcl->run();
+
+=cut
+
+use Moose;
+
+has 'blast_results'   => ( is => 'ro', isa => 'Str', required => 1 );
+has 'mcxdeblast_exec' => ( is => 'ro', isa => 'Str', default  => 'mcxdeblast' );
+has 'mcl_exec'        => ( is => 'ro', isa => 'Str', default  => 'mcl' );
+has 'output_file'     => ( is => 'ro', isa => 'Str', default  => 'output_groups' );
+
+has '_inflation_value' => ( is => 'ro', isa => 'Num', default => 1.5 );
+has '_logging'          => ( is => 'ro', isa => 'Str', default  => '2> /dev/null' );
+
+sub _command_to_run {
+    my ($self) = @_;
+    return join(
+        " ",
+        (
+            $self->mcxdeblast_exec, '-m9', 
+            '--line-mode=abc', $self->blast_results, 
+            '|', $self->mcl_exec, '-', '--abc',
+            '-I', $self->_inflation_value, '-o', $self->output_file, 
+            $self->_logging
+        )
+    );
+}
+
+sub run {
+    my ($self) = @_;
+    system( $self->_command_to_run );
+    1;
+}
+
+no Moose;
+__PACKAGE__->meta->make_immutable;
+1;