Merge pull request #8 from andrewjpage/master

try to fix broken input fastas and more overall tests

get_sequence_type

@@ -22,6 +22,7 @@ path-help@sanger.ac.uk
 package SequenceType::Main;
 
 BEGIN { unshift(@INC, './modules') }
+use lib "/software/pathogen/internal/prod/lib";
 use Moose;
 use Getopt::Long;
 use Cwd;
@@ -102,5 +103,3 @@ else
   );
   $multiple_fastas->create_result_files;
 }
-
-# list all available MLST databases

modules/MLST/CompareAlleles.pm

@@ -21,8 +21,9 @@ use File::Basename;
 use Bio::SeqIO;
 use MLST::Blast::Database;
 use MLST::Blast::BlastN;
+use MLST::Types;
 
-has 'sequence_filename'      => ( is => 'ro', isa => 'Str',      required => 1 );
+has 'sequence_filename'      => ( is => 'ro', isa => 'MLST::File',      required => 1 );
 has 'allele_filenames'       => ( is => 'ro', isa => 'ArrayRef', required => 1 );
 has 'makeblastdb_exec'       => ( is => 'ro', isa => 'Str',      default  => 'makeblastdb' );
 has 'blastn_exec'            => ( is => 'ro', isa => 'Str',      default  => 'blastn' );

modules/MLST/Exceptions.pm

@@ -0,0 +1,7 @@
+package MLST::Exceptions;
+
+use Exception::Class (
+    MLST::Exceptions::FileDoestExist  => { description => 'File doesnt exist' },
+);  
+
+1;

modules/MLST/MultipleFastas.pm

@@ -22,6 +22,8 @@ use Moose;
 use Parallel::ForkManager;
 use MLST::ProcessFasta;
 use MLST::Spreadsheet::File;
+use MLST::NormaliseFasta;
+use File::Temp;
 
 has 'species'               => ( is => 'ro', isa => 'Str',      required => 1 ); 
 has 'base_directory'        => ( is => 'ro', isa => 'Str',      required => 1 ); 
@@ -41,6 +43,7 @@ has '_input_fasta_files'    => ( is => 'ro', isa => 'ArrayRef', lazy => 1, build
 
 has '_concat_names'      => ( is => 'rw', isa => 'ArrayRef', default => sub {[]} ); 
 has '_concat_sequences' => ( is => 'rw', isa => 'ArrayRef', default => sub {[]} ); 
+has '_working_directory' => ( is => 'ro', isa => 'File::Temp::Dir', default => sub { File::Temp->newdir(CLEANUP => 1); });
 
 sub _generate_spreadsheet_rows
 {
@@ -93,18 +96,18 @@ sub _generate_spreadsheet_rows
 sub _build__input_fasta_files
 {
   my($self) = @_;
-  # TODO: Validate and Reformat the fasta files if theres a pipe character
-
-  # Validate
+  my @normalised_fasta_files;
+
   for my $fastafile (@{$self->raw_input_fasta_files})
   {
-    if(!(-e $fastafile ))
-    {
-      die "Input file doesnt exist: $fastafile\n";
-    }
+    my $output_fasta_obj = MLST::NormaliseFasta->new(
+      fasta_filename     => $fastafile,
+      working_directory  => $self->_working_directory->dirname()
+    );
+    push(@normalised_fasta_files,$output_fasta_obj->processed_fasta_filename());
   }
 
-  return $self->raw_input_fasta_files;
+  return \@normalised_fasta_files;
 }
 
 sub create_result_files

modules/MLST/NormaliseFasta.pm

@@ -0,0 +1,71 @@
+=head1 NAME
+
+NormaliseFasta - Take in a Fasta file, check for invalid characters and build a corrected file if needed. This is needed for NCBI makeblastdb which doesnt like the pipe character in the sequence name
+
+=head1 SYNOPSIS
+
+use MLST::NormaliseFasta;
+
+my $output_fasta = MLST::NormaliseFasta->new(
+  fasta_filename     => 'Filename.fasta'
+
+);
+$output_fasta->processed_fasta_filename();
+
+=cut
+
+package MLST::NormaliseFasta;
+use Moose;
+use Bio::SeqIO;
+use File::Basename;
+use MLST::Types;
+
+has 'fasta_filename'      => ( is => 'ro', isa => 'MLST::File',  required => 1 ); 
+has 'working_directory'   => ( is => 'ro', isa => 'Str',         required => 1 ); 
+
+has '_normalised_fasta_filename' => ( is => 'ro', isa => 'Str',  lazy => 1, builder => '_build__normalised_fasta_filename' ); 
+
+sub _build__normalised_fasta_filename
+{
+  my($self) = @_;
+  my $fasta_obj =  Bio::SeqIO->new( -file => $self->fasta_filename , -format => 'Fasta');
+
+  while(my $seq = $fasta_obj->next_seq())
+  {
+    if($seq->id =~ m/\|/ )
+    {
+      return $self->_rename_sequences();
+    }
+  }
+
+  return $self->fasta_filename;
+}
+
+sub _rename_sequences
+{
+  my($self) = @_;
+  my $in_fasta_obj =  Bio::SeqIO->new( -file => $self->fasta_filename , -format => 'Fasta');
+  my($filename, $directories, $suffix) = fileparse($self->fasta_filename);
+  my $output_filename = $self->working_directory.'/'.$filename.$suffix ;
+  my $out_fasta_obj = Bio::SeqIO->new(-file => "+>".$output_filename , -format => 'Fasta');
+
+  my $counter = 1;
+  while(my $seq = $in_fasta_obj->next_seq())
+  {
+    $seq->id($counter."");
+    $out_fasta_obj->write_seq($seq);
+    $counter++;
+  }
+  return $output_filename;
+}
+
+sub processed_fasta_filename
+{
+  my($self) = @_;
+  return $self->_normalised_fasta_filename;
+}
+
+no Moose;
+__PACKAGE__->meta->make_immutable;
+1;
+

modules/MLST/OutputFasta.pm

@@ -22,11 +22,12 @@ use File::Basename;
 use File::Path qw(make_path);
 use Bio::PrimarySeq;
 use Bio::SeqIO;
+use MLST::Types;
 
 has 'matching_sequences'      => ( is => 'ro', isa => 'Maybe[HashRef]',      required => 1 ); 
 has 'non_matching_sequences'  => ( is => 'ro', isa => 'Maybe[HashRef]',      required => 1 ); 
 has 'output_directory'        => ( is => 'ro', isa => 'Str',          required => 1 ); 
-has 'input_fasta_file'        => ( is => 'ro', isa => 'Str',          required => 1 ); 
+has 'input_fasta_file'        => ( is => 'ro', isa => 'MLST::File',          required => 1 ); 
 
 has '_fasta_filename'         => ( is => 'ro', isa => 'Str',          lazy => 1, builder => '_build__fasta_filename' ); 
 has 'concat_sequence'         => ( is => 'rw', isa => 'Maybe[Str]' );

modules/MLST/ProcessFasta.pm

@@ -24,10 +24,11 @@ use MLST::CompareAlleles;
 use MLST::SequenceType;
 use MLST::OutputFasta;
 use MLST::Spreadsheet::Row;
+use MLST::Types;
 
 has 'species'             => ( is => 'ro', isa => 'Str',      required => 1 ); 
 has 'base_directory'      => ( is => 'ro', isa => 'Str',      required => 1 ); 
-has 'fasta_file'          => ( is => 'ro', isa => 'Str',      required => 1 ); 
+has 'fasta_file'          => ( is => 'ro', isa => 'MLST::File',      required => 1 ); 
 has 'makeblastdb_exec'    => ( is => 'ro', isa => 'Str',      required => 1 ); 
 has 'blastn_exec'         => ( is => 'ro', isa => 'Str',      required => 1 ); 
 has 'output_directory'    => ( is => 'ro', isa => 'Str',      required => 1 ); 

modules/MLST/SearchForFiles.pm

@@ -17,11 +17,12 @@ $search_results->profiles_filename();
 
 package MLST::SearchForFiles;
 use Moose;
+use MLST::Types;
 
 has 'species_name'      => ( is => 'ro', isa => 'Str',      required => 1 ); 
 has 'base_directory'    => ( is => 'ro', isa => 'Str',      required => 1 ); 
 
-has 'profiles_filename'     => ( is => 'ro', isa => 'Str',      lazy => 1, builder => '_build_profiles_filename');
+has 'profiles_filename'     => ( is => 'ro', isa => 'MLST::File',      lazy => 1, builder => '_build_profiles_filename');
 has 'allele_filenames'      => ( is => 'ro', isa => 'ArrayRef', lazy => 1, builder => '_build_allele_filenames');
 has 'search_base_directory' => ( is => 'ro', isa => 'Str',      lazy => 1, builder => '_build__search_base_directory');
 

modules/MLST/SequenceType.pm

@@ -17,8 +17,9 @@ $st->sequence_type();
 
 package MLST::SequenceType;
 use Moose;
+use MLST::Types;
 
-has 'profiles_filename'     => ( is => 'ro', isa => 'Str',        required => 1 ); 
+has 'profiles_filename'     => ( is => 'ro', isa => 'MLST::File',        required => 1 ); 
 has 'sequence_names'        => ( is => 'ro', isa => 'ArrayRef',   required => 1 ); 
 
 has 'allele_to_number'     => ( is => 'ro', isa => 'HashRef',    lazy => 1, builder => '_build_allele_to_number' ); 

modules/MLST/Types.pm

@@ -2,11 +2,16 @@ package MLST::Types;
 use Moose;
 use Moose::Util::TypeConstraints;
 use MLST::Validate::Executable;
+use MLST::Validate::File;
 
 subtype 'MLST::Executable',
   as 'Str',
   where { MLST::Validate::Executable->new()->does_executable_exist($_) };
 
+subtype 'MLST::File',
+  as 'Str',
+  where { MLST::Validate::File->new()->does_file_exist($_) };
+
 no Moose;
 no Moose::Util::TypeConstraints;
 __PACKAGE__->meta->make_immutable;

modules/MLST/Validate/File.pm

@@ -0,0 +1,22 @@
+=head1 NAME
+
+File - Does a file exist?
+
+=head1 SYNOPSIS
+
+=cut
+
+package MLST::Validate::File;
+use Moose;
+
+sub does_file_exist
+{
+  my($self, $file) = @_;
+  return 1 if(-e $file);
+
+  return 0;
+}
+
+no Moose;
+__PACKAGE__->meta->make_immutable;
+1;

t/Input/NormaliseFasta.t

@@ -0,0 +1,32 @@
+#!/usr/bin/env perl
+use strict;
+use warnings;
+use File::Temp;
+use Bio::SeqIO;
+
+BEGIN { unshift(@INC, './modules') }
+BEGIN {
+    use Test::Most;
+    use_ok('MLST::NormaliseFasta');
+}
+
+my $tmpdirectory_obj = File::Temp->newdir(CLEANUP => 1);
+my $tmpdirectory = $tmpdirectory_obj->dirname();
+
+ok((my $output_fasta = MLST::NormaliseFasta->new(
+  fasta_filename    => 't/data/contigs.fa',
+  working_directory => $tmpdirectory
+)),'Initalise file wihtout pipe characters in sequence names');
+is($output_fasta->processed_fasta_filename(),'t/data/contigs.fa', 'file without pipe characters shouldnt change at all');
+
+
+ok(($output_fasta = MLST::NormaliseFasta->new(
+  fasta_filename    => 't/data/contigs_pipe_character_in_seq_name.fa',
+  working_directory => $tmpdirectory
+)),'Initalise file with pipe characters in filename');
+is($output_fasta->processed_fasta_filename(), $tmpdirectory.'/contigs_pipe_character_in_seq_name.fa', 'file without pipe characters shouldnt change at all');
+ok((my $in_fasta_obj =  Bio::SeqIO->new( -file => $tmpdirectory.'/contigs_pipe_character_in_seq_name.fa' , -format => 'Fasta')), 'Open temp fasta file');
+is($in_fasta_obj->next_seq()->id, '1', 'seq name now 1');
+is($in_fasta_obj->next_seq()->id, '2', 'seq name now 2');
+is($in_fasta_obj->next_seq()->id, '3', 'seq name now 3');
+done_testing();

t/Output/MultipleFastas.t

@@ -0,0 +1,83 @@
+#!/usr/bin/env perl
+use strict;
+use warnings;
+use File::Temp;
+
+BEGIN { unshift(@INC, './modules') }
+BEGIN {
+    use Test::Most;
+    use_ok('MLST::MultipleFastas');
+}
+
+my $tmpdirectory_obj = File::Temp->newdir(CLEANUP => 1);
+my $tmpdirectory = $tmpdirectory_obj->dirname();
+
+ok((my $multiple_fastas = MLST::MultipleFastas->new(
+  species               => "E.coli",
+  base_directory        => 't/data',
+  raw_input_fasta_files => ['t/data/contigs.fa'],
+  makeblastdb_exec      => 'makeblastdb',
+  blastn_exec           => 'blastn',
+  output_directory      => $tmpdirectory,
+  output_fasta_files    => 1,
+  spreadsheet_basename  => 'mlst_results',
+  parallel_processes    => 1
+)),'Initialise single valid fasta');
+ok(($multiple_fastas->create_result_files),'create all the results files for a single valid fasta');
+compare_files('t/data/expected_mlst_results.genomic.csv', $tmpdirectory.'/mlst_results.genomic.csv');
+compare_files('t/data/expected_mlst_results.allele.csv', $tmpdirectory.'/mlst_results.allele.csv');
+compare_files('t/data/expected_concatenated_alleles.fa', $tmpdirectory.'/concatenated_alleles.fa');
+
+$tmpdirectory_obj = File::Temp->newdir(CLEANUP => 1);
+$tmpdirectory = $tmpdirectory_obj->dirname();
+ok(($multiple_fastas = MLST::MultipleFastas->new(
+  species               => "E.coli",
+  base_directory        => 't/data',
+  raw_input_fasta_files => ['t/data/contigs.fa','t/data/contigs_pipe_character_in_seq_name.fa'],
+  makeblastdb_exec      => 'makeblastdb',
+  blastn_exec           => 'blastn',
+  output_directory      => $tmpdirectory,
+  output_fasta_files    => 1,
+  spreadsheet_basename  => 'mlst_results',
+  parallel_processes    => 1
+)),'Initialise 2 files, one with pipe char and no hits');
+ok(($multiple_fastas->create_result_files),'create all the results files for two fastas');
+compare_files('t/data/expected_two_mlst_results.genomic.csv', $tmpdirectory.'/mlst_results.genomic.csv');
+compare_files('t/data/expected_two_mlst_results.allele.csv', $tmpdirectory.'/mlst_results.allele.csv');
+compare_files('t/data/expected_two_concatenated_alleles.fa', $tmpdirectory.'/concatenated_alleles.fa');
+
+
+$tmpdirectory_obj = File::Temp->newdir(CLEANUP => 1);
+$tmpdirectory = $tmpdirectory_obj->dirname();
+ok(($multiple_fastas = MLST::MultipleFastas->new(
+  species               => "E.coli",
+  base_directory        => 't/data',
+  raw_input_fasta_files => ['t/data/contigs.fa','t/data/contigs_pipe_character_in_seq_name.fa','t/data/contigs_one_unknown.tfa'],
+  makeblastdb_exec      => 'makeblastdb',
+  blastn_exec           => 'blastn',
+  output_directory      => $tmpdirectory,
+  output_fasta_files    => 1,
+  spreadsheet_basename  => 'mlst_results',
+  parallel_processes    => 1
+)),'Initialise 3 files where 1 has near matches');
+ok(($multiple_fastas->create_result_files),'create all the results files for three fastas');
+compare_files('t/data/expected_three_mlst_results.genomic.csv', $tmpdirectory.'/mlst_results.genomic.csv');
+compare_files('t/data/expected_three_mlst_results.allele.csv', $tmpdirectory.'/mlst_results.allele.csv');
+compare_files('t/data/expected_three_concatenated_alleles.fa', $tmpdirectory.'/concatenated_alleles.fa');
+compare_files('t/data/expected_three_contigs_one_unknown.unknown_allele.adk-2.fa', $tmpdirectory.'/contigs_one_unknown.unknown_allele.adk-2.fa');
+compare_files('t/data/expected_three_contigs_one_unknown.unknown_allele.recA-1.fa', $tmpdirectory.'/contigs_one_unknown.unknown_allele.recA-1.fa');
+
+
+done_testing();
+
+sub compare_files
+{
+  my($expected_file, $actual_file) = @_;
+  ok((-e $actual_file),' results file exist');
+  local $/ = undef;
+  open(EXPECTED, $expected_file);
+  open(ACTUAL, $actual_file);
+  my $expected_line = <EXPECTED>;
+  my $actual_line = <ACTUAL>;
+  is($expected_line,$actual_line, 'Content matches expected');
+}