# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
+use warnings;
use strict;
use Maasha::Biopieces;
use Maasha::Filesys;
use Maasha::UCSC::BED;
+use constant {
+ chrom => 0, # BED field names
+ chromStart => 1,
+ chromEnd => 2,
+ name => 3,
+ score => 4,
+ strand => 5,
+ thickStart => 6,
+ thickEnd => 7,
+ itemRgb => 8,
+ blockCount => 9,
+ blockSizes => 10,
+ blockStarts => 11,
+};
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
-my ( $options, $in, $out, $tmp_dir, $bed_file, $tag_file, $fh_in, $fh_out,
- $cols, $record, $bed_entry, $file_hash, $chr, $strand );
+my ( $options, $in, $out, $tmp_dir, $fh_in, $fh_out, $cols, $record, $bed_entry, %fh_hash, %file_hash, $file, $strand, $id, $tag_file, $bed_file );
$options = Maasha::Biopieces::parse_options(
[
$in = Maasha::Biopieces::read_stream( $options->{ "stream_in" } );
$out = Maasha::Biopieces::write_stream( $options->{ "stream_out" } );
+`cd "$ENV{ BP_C }/Maasha/src" && make`; # FIXME OMG this is crufty!
+
$tmp_dir = Maasha::Biopieces::get_tmpdir();
-$bed_file = "$tmp_dir/assemble_tag_contigs.bed";
-$fh_out = Maasha::Filesys::file_write_open( $bed_file );
-$cols = 6; # we only need the first 6 BED columns
+$cols = 6; # we only need the first 6 BED columns
while ( $record = Maasha::Biopieces::get_record( $in ) )
{
if ( $bed_entry = Maasha::UCSC::BED::biopiece2bed( $record, $cols ) )
{
- $strand = $record->{ 'STRAND' } || '+';
+ if ( not exists $fh_hash{ $bed_entry->[ chrom ] . $bed_entry->[ strand ] } )
+ {
+ $file = "$tmp_dir/$bed_entry->[ chrom ]$bed_entry->[ strand ]";
+
+ $file_hash{ $bed_entry->[ chrom ] . $bed_entry->[ strand ] } = $file;
+ $fh_hash{ $bed_entry->[ chrom ] . $bed_entry->[ strand ] } = Maasha::Filesys::file_write_open( $file );
+ }
+
+ $fh_out = $fh_hash{ $bed_entry->[ chrom ] . $bed_entry->[ strand ] };
Maasha::UCSC::BED::bed_entry_put( $bed_entry, $fh_out, $cols, $options->{ 'check' } );
}
}
-close $fh_out;
+$id = 0;
-$file_hash = Maasha::UCSC::BED::bed_file_split_on_chr( $bed_file, $tmp_dir, $cols );
-
-unlink $bed_file;
-
-foreach $chr ( sort keys %{ $file_hash } )
+foreach $file ( sort keys %file_hash )
{
- $bed_file = $file_hash->{ $chr };
+ $bed_file = $file_hash{ $file };
$tag_file = "$bed_file.tc";
+ $strand = substr $file, -1, 1;
+
Maasha::Common::run( "bed2tag_contigs", "< $bed_file > $tag_file" );
$fh_in = Maasha::Filesys::file_read_open( $tag_file );
while ( $bed_entry = Maasha::UCSC::BED::bed_entry_get( $fh_in, $cols, $options->{ 'check' } ) )
{
+ $bed_entry->[ name ] = sprintf( "TC%08d", $id );
+ $bed_entry->[ strand ] = $strand;
+
if ( $record = Maasha::UCSC::BED::bed2biopiece( $bed_entry ) ) {
Maasha::Biopieces::put_record( $record, $out );
}
+
+ $id++;
}
close $fh_in;