# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
-my ( $options, $in, $out, $tmp_dir, $tmp_fh1, $tmp_fh2, $fh, $record, $entry, @args1, @args2, $arg_str1, $arg_str2, $clusters );
-
-$tmp_dir = Maasha::Biopieces::get_tmpdir();
+my ( $options, $in, $out, $tmp_dir, $tmp_file, $tmp_fh, $record, $entry, @args, $arg_str, @fields, $line );
$options = Maasha::Biopieces::parse_options(
[
- { long => 'identity', short => 'i', type => 'float', mandatory => 'no', default => "0.9", allowed => undef, disallowed => undef },
- { long => 'word_size', short => 'w', type => 'uint', mandatory => 'no', default => 7, allowed => undef, disallowed => 0 },
- { long => 'fast_clust', short => 'f', type => 'flag', mandatory => 'no', default => undef, allowed => undef, disallowed => undef },
-
- { long => 'tmp_dir', short => 't', type => 'dir!', mandatory => 'no', default => $tmp_dir, allowed => undef, disallowed => undef },
+ { long => 'identity', short => 'i', type => 'float', mandatory => 'no', default => "0.9", allowed => undef, disallowed => undef },
+ { long => 'library', short => 'l', type => 'file!', mandatory => 'no', default => undef, allowed => undef, disallowed => undef },
+ { long => 'no_sort', short => 'n', type => 'flag', mandatory => 'no', default => undef, allowed => undef, disallowed => undef },
]
);
$in = Maasha::Biopieces::read_stream( $options->{ "stream_in" } );
$out = Maasha::Biopieces::write_stream( $options->{ "stream_out" } );
-$tmp_fh1 = Maasha::Filesys::file_write_open( "$tmp_dir/cluster.fasta" );
-$tmp_fh2 = Maasha::Filesys::file_write_open( "$tmp_dir/cluster.stream" );
+$tmp_dir = Maasha::Biopieces::get_tmpdir();
+$tmp_file = "$tmp_dir/uclust.fasta";
+$tmp_fh = Maasha::Filesys::file_write_open( $tmp_file );
while ( $record = Maasha::Biopieces::get_record( $in ) )
{
- if ( $entry = Maasha::Fasta::biopiece2fasta( $record ) )
- {
- Maasha::Fasta::put_entry( $entry, $tmp_fh1 );
+ if ( $entry = Maasha::Fasta::biopiece2fasta( $record ) ) {
+ Maasha::Fasta::put_entry( $entry, $tmp_fh );
}
- Maasha::Biopieces::put_record( $record, $tmp_fh2 );
+ Maasha::Biopieces::put_record( $record, $out );
}
-close $tmp_fh1;
-close $tmp_fh2;
-
-push @args1, "--sort $tmp_dir/cluster.fasta";
-push @args1, "--output $tmp_dir/cluster.fasta.sort";
-push @args1, "--tmpdir $options->{ 'tmp_dir' }";
-push @args1, "--quiet" if not $options->{ 'verbose' };
-push @args1, "> /dev/null 2>&1" if not $options->{ 'verbose' };
+close $tmp_fh;
-push @args2, "--input $tmp_dir/cluster.fasta.sort";
-push @args2, "--id $options->{ 'identity' }";
-push @args2, "--tmpdir $options->{ 'tmp_dir' }";
-push @args2, "--uc $tmp_dir/cluster.uc";
-push @args2, "--quiet" if not $options->{ 'verbose' };
-push @args2, "> /dev/null 2>&1" if not $options->{ 'verbose' };
+uclust_sort( $tmp_file, $tmp_dir, $options->{ 'verbose' } ) if not $options->{ 'no_sort' };
-$arg_str1 = join " ", @args1;
-$arg_str2 = join " ", @args2;
+push @args, "--input $tmp_file";
+push @args, "--id $options->{ 'identity' }";
+push @args, "--tmpdir $tmp_dir";
+push @args, "--uc $tmp_file.out";
+push @args, "--lib $options->{ 'library' }" if $options->{ 'library' };
+push @args, "--libonly" if $options->{ 'library' };
+push @args, "--quiet" if not $options->{ 'verbose' };
+push @args, "> /dev/null 2>&1" if not $options->{ 'verbose' };
-Maasha::Common::run( "uclust", $arg_str1 );
-Maasha::Common::run( "uclust", $arg_str2 );
+$arg_str = join " ", @args;
-$clusters = parse_clusters( "$tmp_dir/cluster.uc" );
+Maasha::Common::run( "uclust", $arg_str );
-$tmp_fh2 = Maasha::Filesys::file_read_open( "$tmp_dir/cluster.stream" );
+$tmp_fh = Maasha::Filesys::file_read_open( "$tmp_file.out" );
-while ( $record = Maasha::Biopieces::get_record( $tmp_fh2 ) )
+while ( $line = <$tmp_fh> )
{
- if ( exists $clusters->{ $record->{ 'SEQ_NAME' } } ) {
- $record->{ 'CLUSTER' } = $clusters->{ $record->{ 'SEQ_NAME' } };
- }
+ next if $line =~ /^#/;
+
+ chomp $line;
+
+ @fields = split "\t", $line;
+
+ $record = {
+ REC_TYPE => 'UCLUST',
+ TYPE => $fields[ 0 ],
+ CLUSTER => $fields[ 1 ],
+ SIZE => $fields[ 2 ],
+ IDENT => $fields[ 3 ],
+ STRAND => $fields[ 4 ],
+ Q_BEG => $fields[ 5 ],
+ S_BEG => $fields[ 6 ],
+ CIGAR => $fields[ 7 ],
+ SEQ_NAME => $fields[ 8 ],
+ };
Maasha::Biopieces::put_record( $record, $out );
}
+close $tmp_fh;
+
Maasha::Biopieces::close_stream( $in );
Maasha::Biopieces::close_stream( $out );
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> SUBROUTINES <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
-sub parse_clusters
+sub uclust_sort
{
# Martin A. Hansen, January 2010.
- # Parses a uclust uc cluster file and returns a hash with
- # sequence name as key and cluster number as value.
+ # Sorts a FASTA file according to sequence length
+ # - longest first - using uclust.
- my ( $file, # cluster file
+ my ( $file, # file to sort
+ $tmp_dir, # temporary directory for sorting
+ $verbose, # verbose flag - OPTIONAL
) = @_;
- # Returns a hash.
-
- my ( $fh, $line, %clusters, $seq_name, $cluster );
-
- $fh = Maasha::Filesys::file_read_open( $file );
-
- while ( $line = <$fh> )
- {
- next if $line =~ /^#/;
-
- chomp $line;
+ # Returns nothing.
+
+ my ( @args, $arg_str );
- ( undef, $cluster, undef, undef, undef, undef, undef, undef, $seq_name ) = split "\t", $line;
+ push @args, "--mergesort $file";
+ push @args, "--output $file.sort";
+ push @args, "--tmpdir $tmp_dir";
+ push @args, "--quiet" if not $verbose;
+ push @args, "> /dev/null 2>&1" if not $verbose;
- $clusters{ $seq_name } = $cluster;
- }
+ $arg_str = join " ", @args;
- close $fh;
+ Maasha::Common::run( "uclust", $arg_str );
- return wantarray ? %clusters : \%clusters;
+ rename "$file.sort", $file;
}