use Maasha::GFF;
use Maasha::TwoBit;
use Maasha::Solid;
+use Maasha::Solexa;
use Maasha::SQL;
use Maasha::Gwiki;
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> RUN SCRIPT <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
-my $t0 = gettimeofday();
-
run_script( $script );
-my $t1 = gettimeofday();
-
-print STDERR "Program: $script" . ( " " x ( 25 - length( $script ) ) ) . sprintf( "Run time: %.4f\n", ( $t1 - $t0 ) );
-
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> SUBROUTINES <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
# Returns nothing.
- my ( $options, $in, $out );
+ my ( $t0, $t1, $options, $in, $out );
+
+ $t0 = gettimeofday();
$options = get_options( $script );
elsif ( $script eq "oligo_freq" ) { script_oligo_freq( $in, $out, $options ) }
elsif ( $script eq "create_weight_matrix" ) { script_create_weight_matrix( $in, $out, $options ) }
elsif ( $script eq "calc_bit_scores" ) { script_calc_bit_scores( $in, $out, $options ) }
+ elsif ( $script eq "calc_fixedstep" ) { script_calc_fixedstep( $in, $out, $options ) }
elsif ( $script eq "reverse_seq" ) { script_reverse_seq( $in, $out, $options ) }
elsif ( $script eq "complement_seq" ) { script_complement_seq( $in, $out, $options ) }
elsif ( $script eq "remove_indels" ) { script_remove_indels( $in, $out, $options ) }
elsif ( $script eq "write_solid" ) { script_write_solid( $in, $out, $options ) }
elsif ( $script eq "head_records" ) { script_head_records( $in, $out, $options ) }
elsif ( $script eq "remove_keys" ) { script_remove_keys( $in, $out, $options ) }
+ elsif ( $script eq "remove_adaptor" ) { script_remove_adaptor( $in, $out, $options ) }
elsif ( $script eq "rename_keys" ) { script_rename_keys( $in, $out, $options ) }
elsif ( $script eq "uniq_vals" ) { script_uniq_vals( $in, $out, $options ) }
elsif ( $script eq "merge_vals" ) { script_merge_vals( $in, $out, $options ) }
+ elsif ( $script eq "merge_records" ) { script_merge_records( $in, $out, $options ) }
elsif ( $script eq "grab" ) { script_grab( $in, $out, $options ) }
elsif ( $script eq "compute" ) { script_compute( $in, $out, $options ) }
elsif ( $script eq "flip_tab" ) { script_flip_tab( $in, $out, $options ) }
close $in if defined $in;
close $out;
+
+ $t1 = gettimeofday();
+
+ print STDERR "Program: $script" . ( " " x ( 25 - length( $script ) ) ) . sprintf( "Run time: %.4f\n", ( $t1 - $t0 ) ) if $options->{ 'verbose' };
}
# Returns hash
- my ( %options, @options, $opt, @genomes );
+ my ( %options, @options, $opt, @genomes, $real );
if ( $script eq "print_usage" )
{
{
@options = qw(
data_in|i=s
+ samples|s=s
num|n=s
);
}
@options = qw(
data_in|i=s
num|n=s
+ format|f=s
quality|q=s
);
}
percent|p
);
}
+ elsif ( $script eq "calc_fixedstep" )
+ {
+ @options = qw(
+ score|S
+ log10|L
+ );
+ }
elsif ( $script eq "transliterate_seq" )
{
@options = qw(
save_keys|K=s
);
}
+ elsif ( $script eq "remove_adaptor" )
+ {
+ @options = qw(
+ adaptor|a=s
+ mismatches|m=s
+ remove|r=s
+ offset|o=s
+ );
+ }
elsif ( $script eq "rename_keys" )
{
@options = qw(
delimit|d=s
);
}
+ elsif ( $script eq "merge_records" )
+ {
+ @options = qw(
+ keys|k=s
+ merge|m=s
+ );
+ }
elsif ( $script eq "grab" )
{
@options = qw(
priority|p=f
use_score|u
visibility|v=s
- wiggle|w
color|c=s
chunk_size|C=s
);
$options{ "feats" } = [ split ",", $options{ "feats" } ] if defined $options{ "feats" };
$options{ "frames" } = [ split ",", $options{ "frames" } ] if defined $options{ "frames" };
$options{ "formats" } = [ split ",", $options{ "formats" } ] if defined $options{ "formats" };
+ $options{ "samples" } = [ split ",", $options{ "samples" } ] if defined $options{ "samples" };
# ---- check arguments ----
map { Maasha::Common::error( qq(Argument to --cols must be a whole numbers - not "$_") ) if $_ !~ /^\d+$/ } @{ $options{ "cols" } } if $options{ "cols" };
-# print STDERR Dumper( \%options );
+ # print STDERR Dumper( \%options );
+
+ $real = "beg|end|word_size|wrap|chunk_size|tile_size|len|prefix_length|mismatches|offset|num|skip|cpus|window_size|step_size";
foreach $opt ( keys %options )
{
{
Maasha::Common::error( qq(Argument to --$opt must be a valid file or fileglob expression - not "$options{ $opt }") );
}
- elsif ( $opt =~ /beg|end|word_size|wrap|chunk_size|tile_size|len|prefix_length|num|skip|cpus|window_size|step_size/ and $options{ $opt } !~ /^\d+$/ )
+ elsif ( $opt =~ /$real/ and $options{ $opt } !~ /^\d+$/ )
{
Maasha::Common::error( qq(Argument to --$opt must be a whole number - not "$options{ $opt }") );
}
Maasha::Common::error( qq(Genome $options{ $opt } not found in "$ENV{ 'BP_DATA' }/genomes/") );
}
}
- elsif ( $opt eq "terminal" and not $options{ $opt } =~ /^(svg|post|dumb)/ )
+ elsif ( $opt eq "terminal" and not $options{ $opt } =~ /^(svg|post|dumb|x11)/ )
{
Maasha::Common::error( qq(Bad --$opt argument "$options{ $opt }") );
}
- elsif ( $opt eq "table" and $options{ $opt } =~ /-\./ )
+ elsif ( $opt eq "table" and $options{ $opt } =~ /(-|\.)/ )
+ {
+ Maasha::Common::error( qq(Character '$1' is not allowed in table name: $options{ $opt }) );
+ }
+ elsif ( $opt eq "merge" and $options{ $opt } !~ /^(AandB|AorB|BorA|AnotB|BnotA)$/ )
+ {
+ Maasha::Common::error( qq(Argument to --$opt must be AandB, AorB, BorA, AnotB, or BnotA - not "$options{ $opt }") );
+ }
+ elsif ( $opt eq "format" and $script eq "read_solexa" and $options{ $opt } !~ /octal|decimal/ )
{
- Maasha::Common::error( qq(Character '$options{ $opt }' is not allowed in table names) );
+ Maasha::Common::error( qq(Argument to --$opt must be octal or decimal - not "$options{ $opt }") );
+ }
+ elsif ( $opt eq "remove" and $script eq "remove_adaptor" and $options{ $opt } !~ /before|after|skip/ )
+ {
+ Maasha::Common::error( qq(Argument to --$opt must be before, after, or skip - not "$options{ $opt }") );
}
}
# Returns nothing.
- my ( $record, @files, $file, $entries, $entry, $num );
+ my ( $record, $file, $data_in, $num );
while ( $record = get_record( $in ) ) {
put_record( $record, $out );
foreach $file ( @{ $options->{ "files" } } )
{
- $entries = Maasha::UCSC::psl_get_entries( $file );
+ $data_in = Maasha::Common::read_open( $file );
- foreach $entry ( @{ $entries } )
+ while ( $record = Maasha::UCSC::psl_get_entry( $data_in ) )
{
- put_record( $entry, $out );
+ put_record( $record, $out );
goto NUM if $options->{ "num" } and $num == $options->{ "num" };
{
# Martin A. Hansen, Juli 2008.
- # Read fixedStep wiggle format from stream or file.
+ # Read fixedstep wiggle format from stream or file.
my ( $in, # handle to in stream
$out, # handle to out stream
if ( $head =~ /^chrom=([^ ]+) start=(\d+) step=(\d+)$/ )
{
- $record->{ "CHR" } = $1;
- $record->{ "CHR_BEG" } = $2;
- $record->{ "STEP" } = $3;
- $record->{ "VALS" } = join ",", @{ $entry };
+ $record->{ "REC_TYPE" } = "fixed_step";
+ $record->{ "CHR" } = $1;
+ $record->{ "CHR_BEG" } = $2;
+ $record->{ "STEP" } = $3;
+ $record->{ "VALS" } = join ";", @{ $entry };
}
put_record( $record, $out );
# Returns nothing.
- my ( $data_in, $file, $num, $records, $record, $soft_index, $fh, @platforms, $plat_table, @samples, $sample, $old_end );
+ my ( $data_in, $file, $num, $records, $record, $soft_index, $fh, @platforms, $plat_table, @samples, $sample, $old_end, $skip );
while ( $record = get_record( $in ) ) {
put_record( $record, $out );
foreach $file ( @{ $options->{ "files" } } )
{
+ print STDERR "Creating index for file: $file\n" if $options->{ "verbose" };
+
$soft_index = Maasha::NCBI::soft_index_file( $file );
$fh = Maasha::Common::read_open( $file );
- @platforms = grep { $_->[ 0 ] =~ /PLATFORM/ } @{ $soft_index };
+ @platforms = grep { $_->{ "SECTION" } =~ /PLATFORM/ } @{ $soft_index };
+
+ print STDERR "Getting platform tables for file: $file\n" if $options->{ "verbose" };
- $plat_table = Maasha::NCBI::soft_get_platform( $fh, $platforms[ 0 ]->[ 1 ], $platforms[ -1 ]->[ 2 ] );
+ $plat_table = Maasha::NCBI::soft_get_platform( $fh, $platforms[ 0 ]->{ "LINE_BEG" }, $platforms[ -1 ]->{ "LINE_END" } );
- @samples = grep { $_->[ 0 ] =~ /SAMPLE/ } @{ $soft_index };
+ @samples = grep { $_->{ "SECTION" } =~ /SAMPLE/ } @{ $soft_index };
- $old_end = $platforms[ -1 ]->[ 2 ];
+ $old_end = $platforms[ -1 ]->{ "LINE_END" };
foreach $sample ( @samples )
{
- $records = Maasha::NCBI::soft_get_sample( $fh, $plat_table, $sample->[ 1 ] - $old_end - 1, $sample->[ 2 ] - $old_end - 1 );
+ $skip = 0;
+ $skip = 1 if ( $options->{ "samples" } and grep { $sample->{ "SECTION" } !~ /$_/ } @{ $options->{ "samples" } } );
+
+ print STDERR "Getting samples for dataset: $sample->{ 'SECTION' }\n" if $options->{ "verbose" } and not $skip;
+
+ $records = Maasha::NCBI::soft_get_sample( $fh, $plat_table, $sample->{ "LINE_BEG" } - $old_end - 1, $sample->{ "LINE_END" } - $old_end - 1, $skip );
foreach $record ( @{ $records } )
{
$num++;
}
- $old_end = $sample->[ 2 ];
+ $old_end = $sample->{ "LINE_END" };
}
close $fh;
# Returns nothing.
- my ( $record, $file, $base_name, $data_in, $line, $num, @fields, @seqs, @scores, $i, $seq, $seq_count );
+ my ( $record, $file, $data_in, $entry, $num, @seqs, @scores, $i );
+ $options->{ "format" } ||= "octal";
$options->{ "quality" } ||= 20;
while ( $record = get_record( $in ) ) {
foreach $file ( @{ $options->{ "files" } } )
{
- $data_in = Maasha::Common::read_open( $file );
- $base_name = Maasha::Common::get_basename( $file );
- $base_name =~ s/\..*//;
-
- $seq_count = 0;
+ $data_in = Maasha::Common::read_open( $file );
- while ( $line = <$data_in> )
+ if ( $options->{ "format" } eq "octal" )
{
- @fields = split /:/, $line;
- @seqs = split //, $fields[ 5 ];
- @scores = split / /, $fields[ -1 ];
+ while ( $entry = Maasha::Solexa::solexa_get_entry_octal( $data_in ) )
+ {
+ $record = Maasha::Solexa::solexa2biopiece( $entry, $options->{ "quality" } );
- for ( $i = 0; $i < @scores; $i++ ) {
- $seqs[ $i ] = lc $seqs[ $i ] if $scores[ $i ] < $options->{ "quality" };
- }
+ put_record( $record, $out );
- $seq = join "", @seqs;
+ goto NUM if $options->{ "num" } and $num == $options->{ "num" };
- $record->{ "SEQ_NAME" } = sprintf( "%s_ID%08d", $base_name, $seq_count );
- $record->{ "SEQ" } = $seq;
- $record->{ "SEQ_LEN" } = length $seq;
- $record->{ "SCORE_MEAN" } = sprintf ( "%.2f", Maasha::Calc::mean( \@scores ) );
+ $num++;
+ }
+ }
+ else
+ {
+ while ( $entry = Maasha::Solexa::solexa_get_entry_decimal( $data_in ) )
+ {
+ $record = Maasha::Solexa::solexa2biopiece( $entry, $options->{ "quality" } );
- put_record( $record, $out );
+ put_record( $record, $out );
- goto NUM if $options->{ "num" } and $num == $options->{ "num" };
+ goto NUM if $options->{ "num" } and $num == $options->{ "num" };
- $seq_count++;
- $num++;
+ $num++;
+ }
}
close $data_in;
}
+sub script_calc_fixedstep
+{
+ # Martin A. Hansen, September 2008.
+
+ # Calculates fixedstep entries from data in the stream.
+
+ my ( $in, # handle to in stream
+ $out, # handle to out stream
+ $options, # options hash
+ ) = @_;
+
+ # Returns nothing.
+
+ my ( $record, %fh_hash, $fh_in, $fh_out, $chr, $chr, $beg, $end, $q_id, $block, $entry, $clones, $beg_block, $max, $i );
+
+ while ( $record = get_record( $in ) )
+ {
+ $record->{ "CHR" } = $record->{ "S_ID" } if not defined $record->{ "CHR" };
+ $record->{ "CHR_BEG" } = $record->{ "S_BEG" } if not defined $record->{ "CHR_BEG" };
+ $record->{ "CHR_END" } = $record->{ "S_END" } if not defined $record->{ "CHR_END" };
+
+ if ( $record->{ "CHR" } and defined $record->{ "CHR_BEG" } and $record->{ "CHR_END" } )
+ {
+ $fh_hash{ $record->{ "CHR" } } = Maasha::Common::write_open( "$BP_TMP/$record->{ 'CHR' }" ) if not exists $fh_hash{ $record->{ "CHR" } };
+
+ $fh_out = $fh_hash{ $record->{ "CHR" } };
+
+ Maasha::UCSC::bed_put_entry( $record, $fh_out, 5 );
+ }
+ }
+
+ map { close $_ } keys %fh_hash;
+
+ foreach $chr ( sort keys %fh_hash )
+ {
+ Maasha::Common::run( "bedSort", "$BP_TMP/$chr $BP_TMP/$chr" );
+
+ $fh_in = Maasha::Common::read_open( "$BP_TMP/$chr" );
+
+ undef $block;
+
+ while ( $entry = Maasha::UCSC::bed_get_entry( $fh_in, 5 ) )
+ {
+ $chr = $entry->{ 'CHR' };
+ $beg = $entry->{ 'CHR_BEG' };
+ $end = $entry->{ 'CHR_END' };
+ $q_id = $entry->{ 'Q_ID' };
+
+ if ( $options->{ "score" } ) {
+ $clones = $entry->{ 'SCORE' };
+ } elsif ( $q_id =~ /_(\d+)$/ ) {
+ $clones = $1;
+ } else {
+ $clones = 1;
+ }
+
+ if ( $block )
+ {
+ if ( $beg > $max )
+ {
+ map { $_ = sprintf( "%.4f", Maasha::Calc::log10( $_ ) ) } @{ $block } if $options->{ "log10" };
+
+ $record->{ "CHR" } = $chr;
+ $record->{ "CHR_BEG" } = $beg_block;
+ $record->{ "STEP" } = 1;
+ $record->{ "VALS" } = join ";", @{ $block };
+ $record->{ "REC_TYPE" } = "fixed_step";
+
+ put_record( $record, $out );
+
+ undef $block;
+ }
+ else
+ {
+ for ( $i = $beg - $beg_block; $i < ( $beg - $beg_block ) + ( $end - $beg ); $i++ ) {
+ $block->[ $i ] += $clones;
+ }
+
+ $max = Maasha::Calc::max( $max, $end );
+ }
+ }
+
+ if ( not $block )
+ {
+ $beg_block = $beg;
+ $max = $end;
+
+ for ( $i = 0; $i < ( $end - $beg ); $i++ ) {
+ $block->[ $i ] += $clones;
+ }
+ }
+ }
+
+ close $fh_in;
+
+ map { $_ = sprintf( "%.4f", Maasha::Calc::log10( $_ ) ) } @{ $block } if $options->{ "log10" };
+
+ $record->{ "CHR" } = $chr;
+ $record->{ "CHR_BEG" } = $beg_block;
+ $record->{ "STEP" } = 1;
+ $record->{ "VALS" } = join ";", @{ $block };
+ $record->{ "REC_TYPE" } = "fixed_step";
+
+ put_record( $record, $out );
+
+ unlink "$BP_TMP/$chr";
+ }
+}
+
+
sub script_reverse_seq
{
# Martin A. Hansen, August 2007.
$phastcons_file = Maasha::Config::genome_phastcons( $options->{ "genome" } );
$phastcons_index = Maasha::Config::genome_phastcons_index( $options->{ "genome" } );
- $index = Maasha::UCSC::phastcons_index_retrieve( $phastcons_index );
+ $index = Maasha::UCSC::fixedstep_index_retrieve( $phastcons_index );
$fh_phastcons = Maasha::Common::read_open( $phastcons_file );
if ( defined $options->{ "chr" } and defined $options->{ "beg" } and ( defined $options->{ "end" } or defined $options->{ "len" } ) )
$options->{ "end" } = $options->{ "beg" } + $options->{ "len" } - 1;
}
- $scores = Maasha::UCSC::phastcons_index_lookup( $index, $fh_phastcons, $options->{ "chr" }, $options->{ "beg" }, $options->{ "end" }, $options->{ "flank" } );
+ $scores = Maasha::UCSC::fixedstep_index_lookup( $index, $fh_phastcons, $options->{ "chr" }, $options->{ "beg" }, $options->{ "end" }, $options->{ "flank" } );
$record->{ "CHR" } = $options->{ "chr" };
$record->{ "CHR_BEG" } = $options->{ "beg" } - $options->{ "flank" };
}
}
- Maasha::Common::run( "blastall", "-p $options->{ 'program' } -e $options->{ 'e_val' } -a $options->{ 'cpus' } -m 8 -i $tmp_in -d $options->{ 'database' } -F $options->{ 'filter' } -o $tmp_out > /dev/null 2>&1", 1 );
+ if ( $options->{ 'verbose' } )
+ {
+ Maasha::Common::run(
+ "blastall",
+ join( " ",
+ "-p $options->{ 'program' }",
+ "-e $options->{ 'e_val' }",
+ "-a $options->{ 'cpus' }",
+ "-m 8",
+ "-i $tmp_in",
+ "-d $options->{ 'database' }",
+ "-F $options->{ 'filter' }",
+ "-o $tmp_out",
+ ),
+ 1
+ );
+ }
+ else
+ {
+ Maasha::Common::run(
+ "blastall",
+ join( " ",
+ "-p $options->{ 'program' }",
+ "-e $options->{ 'e_val' }",
+ "-a $options->{ 'cpus' }",
+ "-m 8",
+ "-i $tmp_in",
+ "-d $options->{ 'database' }",
+ "-F $options->{ 'filter' }",
+ "-o $tmp_out",
+ "> /dev/null 2>&1"
+ ),
+ 1
+ );
+ }
unlink $tmp_in;
# Returns nothing.
- my ( $genome, $tmp_in, $tmp_out, $fh_in, $fh_out, $record, $line, @fields, $entry );
+ my ( $genome, $tmp_in, $tmp_out, $fh_in, $fh_out, $record, $line, @fields, $entry, $count, $args );
$options->{ "seed_size" } ||= 10;
$options->{ "mismatches" } ||= 2;
$options->{ "gap_size" } ||= 0;
$options->{ "cpus" } ||= 1;
- $options->{ "in_file" } = "$ENV{ 'BP_DATA' }/genomes/$options->{ 'genome' }/fasta/$options->{ 'genome' }.fna" if $options->{ 'genome' };
+ if ( $options->{ "genome" } ) {
+ $options->{ "in_file" } = "$ENV{ 'BP_DATA' }/genomes/$options->{ 'genome' }/fasta/$options->{ 'genome' }.fna";
+ }
$tmp_in = "$BP_TMP/soap_query.seq";
$tmp_out = "$BP_TMP/soap.result";
$fh_out = Maasha::Common::write_open( $tmp_in );
+
+ $count = 0;
while ( $record = get_record( $in ) )
{
- if ( $entry = record2fasta( $record ) ) {
+ if ( $entry = record2fasta( $record ) )
+ {
Maasha::Fasta::put_entry( $entry, $fh_out );
+
+ $count++;
}
put_record( $record, $out );
close $fh_out;
- Maasha::Common::run( "soap", "-s $options->{ 'seed_size' } -r 2 -a $tmp_in -v $options->{ 'mismatches' } -g $options->{ 'gap_size' } -p $options->{ 'cpus' } -d $options->{ 'in_file' } -o $tmp_out > /dev/null 2>&1", 1 );
+ if ( $count > 0 )
+ {
+ $args = join( " ",
+ "-s $options->{ 'seed_size' }",
+ "-r 2",
+ "-a $tmp_in",
+ "-v $options->{ 'mismatches' }",
+ "-g $options->{ 'gap_size' }",
+ "-p $options->{ 'cpus' }",
+ "-d $options->{ 'in_file' }",
+ "-o $tmp_out",
+ );
- unlink $tmp_in;
+ $args .= " > /dev/null 2>&1" if not $options->{ 'verbose' };
- $fh_out = Maasha::Common::read_open( $tmp_out );
+ Maasha::Common::run( "soap", $args, 1 );
- undef $record;
+ unlink $tmp_in;
- while ( $line = <$fh_out> )
- {
- chomp $line;
+ $fh_out = Maasha::Common::read_open( $tmp_out );
- @fields = split /\t/, $line;
+ undef $record;
- $record->{ "REC_TYPE" } = "SOAP";
- $record->{ "Q_ID" } = $fields[ 0 ];
- $record->{ "SCORE" } = $fields[ 3 ];
- $record->{ "STRAND" } = $fields[ 6 ];
- $record->{ "S_ID" } = $fields[ 7 ];
- $record->{ "S_BEG" } = $fields[ 8 ] - 1; # soap is 1-based
- $record->{ "S_END" } = $fields[ 8 ] + $fields[ 5 ] - 2;
+ while ( $line = <$fh_out> )
+ {
+ chomp $line;
- put_record( $record, $out );
- }
+ @fields = split /\t/, $line;
- close $fh_out;
+ $record->{ "REC_TYPE" } = "SOAP";
+ $record->{ "Q_ID" } = $fields[ 0 ];
+ $record->{ "SCORE" } = $fields[ 3 ];
+ $record->{ "STRAND" } = $fields[ 6 ];
+ $record->{ "S_ID" } = $fields[ 7 ];
+ $record->{ "S_BEG" } = $fields[ 8 ] - 1; # soap is 1-based
+ $record->{ "S_END" } = $fields[ 8 ] + $fields[ 5 ] - 2;
+
+ put_record( $record, $out );
+ }
+
+ close $fh_out;
+ }
unlink $tmp_out;
}
$vals = $record->{ 'VALS' };
- $vals =~ tr/,/\n/;
+ $vals =~ tr/;/\n/;
print $fh "$vals\n";
}
}
+sub script_remove_adaptor
+{
+ # Martin A. Hansen, August 2008.
+
+ # Find and remove adaptor from sequences in the stream.
+
+ my ( $in, # handle to in stream
+ $out, # handle to out stream
+ $options, # options hash
+ ) = @_;
+
+ # Returns nothing.
+
+ my ( $record, $adaptor, $seq, $adaptor_len, $seq_len, $offset, $max_match, $max_mismatch, $pos );
+
+ $options->{ "remove" } ||= "after";
+
+ $max_mismatch = $options->{ "mismatches" } || 0;
+ $offset = $options->{ "offset" };
+
+ if ( not defined $offset ) {
+ $offset = 0;
+ } else {
+ $offset--;
+ }
+
+ $adaptor = uc $options->{ "adaptor" };
+ $adaptor_len = length $adaptor;
+
+ while ( $record = get_record( $in ) )
+ {
+ if ( $record->{ "SEQ" } )
+ {
+ $seq = uc $record->{ "SEQ" };
+ $seq_len = length $seq;
+
+ $pos = Maasha::Common::index_m( $seq, $adaptor, $seq_len, $adaptor_len, $offset, $max_mismatch );
+
+ $record->{ "ADAPTOR_POS" } = $pos;
+
+ if ( $pos >= 0 and $options->{ "remove" } ne "skip" )
+ {
+ if ( $options->{ "remove" } eq "after" )
+ {
+ $record->{ "SEQ" } = substr $record->{ "SEQ" }, 0, $pos;
+ $record->{ "SEQ_LEN" } = $pos;
+ }
+ else
+ {
+ $record->{ "SEQ" } = substr $record->{ "SEQ" }, $pos + $adaptor_len;
+ $record->{ "SEQ_LEN" } = length $record->{ "SEQ" };
+ }
+ }
+
+ put_record( $record, $out );
+ }
+ else
+ {
+ put_record( $record, $out );
+ }
+ }
+}
+
+
sub script_rename_keys
{
# Martin A. Hansen, August 2007.
}
+sub script_merge_records
+{
+ # Martin A. Hansen, July 2008.
+
+ # Merges records in the stream based on identical values of two given keys.
+
+ my ( $in, # handle to in stream
+ $out, # handle to out stream
+ $options, # options hash
+ ) = @_;
+
+ # Returns nothing.
+
+ my ( $merge, $record, $file1, $file2, $fh1, $fh2, $key1, $key2, @keys1, @keys2, @vals1, @vals2,
+ $num1, $num2, $num, $cmp, $i );
+
+ $merge = $options->{ "merge" } || "AandB";
+
+ $file1 = "$BP_TMP/merge_records1.tmp";
+ $file2 = "$BP_TMP/merge_records2.tmp";
+
+ $fh1 = Maasha::Common::write_open( $file1 );
+ $fh2 = Maasha::Common::write_open( $file2 );
+
+ $key1 = $options->{ "keys" }->[ 0 ];
+ $key2 = $options->{ "keys" }->[ 1 ];
+
+ $num = $key2 =~ s/n$//;
+ $num1 = 0;
+ $num2 = 0;
+
+ while ( $record = get_record( $in ) )
+ {
+ if ( exists $record->{ $key1 } )
+ {
+ @keys1 = $key1;
+ @vals1 = $record->{ $key1 };
+
+ delete $record->{ $key1 };
+
+ map { push @keys1, $_; push @vals1, $record->{ $_ } } keys %{ $record };
+
+ print $fh1 join( "\t", @vals1 ), "\n";
+
+ $num1++;
+ }
+ elsif ( exists $record->{ $key2 } )
+ {
+ @keys2 = $key2;
+ @vals2 = $record->{ $key2 };
+
+ delete $record->{ $key2 };
+
+ map { push @keys2, $_; push @vals2, $record->{ $_ } } keys %{ $record };
+
+ print $fh2 join( "\t", @vals2 ), "\n";
+
+ $num2++;
+ }
+ }
+
+ close $fh1;
+ close $fh2;
+
+ if ( $num )
+ {
+ Maasha::Common::run( "sort", "-k 1,1n $file1 > $file1.sort" ) and rename "$file1.sort", $file1;
+ Maasha::Common::run( "sort", "-k 1,1n $file2 > $file2.sort" ) and rename "$file2.sort", $file2;
+ }
+ else
+ {
+ Maasha::Common::run( "sort", "-k 1,1 $file1 > $file1.sort" ) and rename "$file1.sort", $file1;
+ Maasha::Common::run( "sort", "-k 1,1 $file2 > $file2.sort" ) and rename "$file2.sort", $file2;
+ }
+
+ $fh1 = Maasha::Common::read_open( $file1 );
+ $fh2 = Maasha::Common::read_open( $file2 );
+
+ @vals1 = Maasha::Common::get_fields( $fh1 );
+ @vals2 = Maasha::Common::get_fields( $fh2 );
+
+ while ( $num1 > 0 and $num2 > 0 )
+ {
+ undef $record;
+
+ if ( $num ) {
+ $cmp = $vals1[ 0 ] <=> $vals2[ 0 ];
+ } else {
+ $cmp = $vals1[ 0 ] cmp $vals2[ 0 ];
+ }
+
+ if ( $cmp < 0 )
+ {
+ if ( $merge =~ /^(AorB|AnotB)$/ )
+ {
+ for ( $i = 0; $i < @keys1; $i++ ) {
+ $record->{ $keys1[ $i ] } = $vals1[ $i ];
+ }
+
+ put_record( $record, $out );
+ }
+
+ @vals1 = Maasha::Common::get_fields( $fh1 );
+ $num1--;
+ }
+ elsif ( $cmp > 0 )
+ {
+ if ( $merge =~ /^(BorA|BnotA)$/ )
+ {
+ for ( $i = 0; $i < @keys2; $i++ ) {
+ $record->{ $keys2[ $i ] } = $vals2[ $i ];
+ }
+
+ put_record( $record, $out );
+ }
+
+ @vals2 = Maasha::Common::get_fields( $fh2 );
+ $num2--;
+ }
+ else
+ {
+ if ( $merge =~ /^(AandB|AorB|BorA)$/ )
+ {
+ for ( $i = 0; $i < @keys1; $i++ ) {
+ $record->{ $keys1[ $i ] } = $vals1[ $i ];
+ }
+
+ for ( $i = 1; $i < @keys2; $i++ ) {
+ $record->{ $keys2[ $i ] } = $vals2[ $i ];
+ }
+
+ put_record( $record, $out );
+ }
+
+ @vals1 = Maasha::Common::get_fields( $fh1 );
+ @vals2 = Maasha::Common::get_fields( $fh2 );
+ $num1--;
+ $num2--;
+ }
+ }
+
+ close $fh1;
+ close $fh2;
+
+ unlink $file1;
+ unlink $file2;
+
+ if ( $num1 > 0 and $merge =~ /^(AorB|AnotB)$/ )
+ {
+ undef $record;
+
+ for ( $i = 0; $i < @keys1; $i++ ) {
+ $record->{ $keys1[ $i ] } = $vals1[ $i ];
+ }
+
+ put_record( $record, $out );
+ }
+
+ if ( $num2 > 0 and $merge =~ /^(BorA|BnotA)$/ )
+ {
+ undef $record;
+
+ for ( $i = 0; $i < @keys2; $i++ ) {
+ $record->{ $keys2[ $i ] } = $vals2[ $i ];
+ }
+
+ put_record( $record, $out );
+ }
+}
+
+
sub script_grab
{
# Martin A. Hansen, August 2007.
}
}
- $result = { "count_records" => $count };
+ $result = { "RECORDS_COUNT" => $count };
$fh = write_stream( $options->{ "data_out" } );
$fh_out = Maasha::Common::write_open( $tmp_file );
+ $cache = 0;
$num = 0;
while ( $record = get_record( $in ) )
if ( $cache )
{
- $num = 0;
+ $num = 0;
$fh_in = Maasha::Common::read_open( $tmp_file );
while ( $record = get_record( $in ) )
{
- $data_hash{ $record->{ $options->{ "key" } } }++ if $record->{ $options->{ "key" } };
+ $data_hash{ $record->{ $options->{ "key" } } }++ if defined $record->{ $options->{ "key" } };
put_record( $record, $out ) if not $options->{ "no_stream" };
}
while ( $record = get_record( $in ) )
{
- $data_hash{ $record->{ $options->{ "key" } } }++ if $record->{ $options->{ "key" } };
+ $data_hash{ $record->{ $options->{ "key" } } }++ if defined $record->{ $options->{ "key" } };
put_record( $record, $out ) if not $options->{ "no_stream" };
}
# Calculate the mean of values of given keys.
+ # This routine has developed into the most ugly hack. Do something!
+
my ( $in, # handle to in stream
$out, # handle to out stream
$options, # options hash
# Returns nothing.
- my ( $record, $file, $wib_file, $wig_file, $wib_dir, $fh_in, $fh_out, $i, $first, $format, $args, $type, $columns, $append, %fh_hash,
- $chr, $beg, $end, $block, $line, $max, $beg_block, $entry, $q_id, $clones );
+ my ( $record, $file, $wib_file, $wig_file, $wib_dir, $fh_out, $i, $first, $format, $type, $columns, $append, $vals );
$options->{ "short_label" } ||= $options->{ 'table' };
$options->{ "long_label" } ||= $options->{ 'table' };
$i = 0;
- if ( $options->{ 'wiggle' } )
+ $fh_out = Maasha::Common::write_open( $file );
+
+ while ( $record = get_record( $in ) )
{
- $options->{ "visibility" } = "full";
+ put_record( $record, $out ) if not $options->{ "no_stream" };
- while ( $record = get_record( $in ) )
+ if ( $record->{ "REC_TYPE" } eq "fixed_step" )
{
- put_record( $record, $out ) if not $options->{ "no_stream" };
-
- $record->{ "CHR" } = $record->{ "S_ID" } if not defined $record->{ "CHR" };
- $record->{ "CHR_BEG" } = $record->{ "S_BEG" } if not defined $record->{ "CHR_BEG" };
- $record->{ "CHR_END" } = $record->{ "S_END" } if not defined $record->{ "CHR_END" };
+ $vals = $record->{ "VALS" };
+ $vals =~ tr/;/\n/;
- $fh_hash{ $record->{ "CHR" } } = Maasha::Common::write_open( "$BP_TMP/$record->{ 'CHR' }" ) if not exists $fh_hash{ $record->{ "CHR" } };
+ print $fh_out "fixedStep chrom=$record->{ 'CHR' } start=$record->{ 'CHR_BEG' } step=$record->{ 'STEP' }\n";
+ print $fh_out "$vals\n";
- $fh_out = $fh_hash{ $record->{ "CHR" } };
-
- Maasha::UCSC::bed_put_entry( $record, $fh_out, 5 );
+ $format = "WIGGLE" if not $format;
}
-
- map { close $_ } keys %fh_hash;
-
- $fh_out = Maasha::Common::write_open( $file );
-
- foreach $chr ( sort keys %fh_hash )
+ elsif ( $record->{ "REC_TYPE" } eq "PSL" )
{
- Maasha::Common::run( "bedSort", "$BP_TMP/$chr $BP_TMP/$chr" );
-
- $fh_in = Maasha::Common::read_open( "$BP_TMP/$chr" );
-
- undef $block;
-
- while ( $entry = Maasha::UCSC::bed_get_entry( $fh_in, 5 ) )
- {
- $chr = $entry->{ 'CHR' };
- $beg = $entry->{ 'CHR_BEG' };
- $end = $entry->{ 'CHR_END' };
- $q_id = $entry->{ 'Q_ID' };
-
- if ( $q_id =~ /_(\d+)$/ ) {
- $clones = $1;
- } else {
- $clones = 1;
- }
-
- if ( $block )
- {
- if ( $beg > $max )
- {
- Maasha::UCSC::fixedstep_put_entry( $chr, $beg_block, $block, $fh_out );
- undef $block;
- }
- else
- {
- for ( $i = $beg - $beg_block; $i < ( $beg - $beg_block ) + ( $end - $beg ); $i++ ) {
- $block->[ $i ] += $clones;
- }
-
- $max = Maasha::Calc::max( $max, $end );
- }
- }
-
- if ( not $block )
- {
- $beg_block = $beg;
- $max = $end;
-
- for ( $i = 0; $i < ( $end - $beg ); $i++ ) {
- $block->[ $i ] += $clones;
- }
- }
- }
-
- close $fh_in;
-
- Maasha::UCSC::fixedstep_put_entry( $chr, $beg_block, $block, $fh_out );
+ Maasha::UCSC::psl_put_header( $fh_out ) if $first;
+ Maasha::UCSC::psl_put_entry( $record, $fh_out );
+
+ $first = 0;
- unlink "$BP_TMP/$chr";
+ $format = "PSL" if not $format;
}
+ elsif ( $record->{ "REC_TYPE" } eq "BED" and $record->{ "SEC_STRUCT" } )
+ {
+ # chrom chromStart chromEnd name score strand size secStr conf
- close $fh_out;
-
- $wig_file = "$options->{ 'table' }.wig";
- $wib_file = "$options->{ 'table' }.wib";
-
- $wib_dir = "$ENV{ 'HOME' }/ucsc/wib";
-
- Maasha::Common::dir_create_if_not_exists( $wib_dir );
-
- # Maasha::Common::run( "wigEncode", "$file $wig_file $wib_file > /dev/null 2>&1" );
-
- `cd $BP_TMP && wigEncode $file $wig_file $wib_file > /dev/null 2>&1`;
- Maasha::Common::run( "mv", "$BP_TMP/$wib_file $wib_dir" );
-
- unlink $file;
-
- $file = $wig_file;
+ print $fh_out join ( "\t",
+ $record->{ "CHR" },
+ $record->{ "CHR_BEG" },
+ $record->{ "CHR_END" } + 1,
+ $record->{ "Q_ID" },
+ $record->{ "SCORE" },
+ $record->{ "STRAND" },
+ $record->{ "SIZE" },
+ $record->{ "SEC_STRUCT" },
+ $record->{ "CONF" },
+ ), "\n";
- $format = "WIGGLE";
- }
- else
- {
- $fh_out = Maasha::Common::write_open( $file );
-
- while ( $record = get_record( $in ) )
+ $format = "BED_SS" if not $format;
+ }
+ elsif ( $record->{ "REC_TYPE" } eq "BED" )
{
- put_record( $record, $out ) if not $options->{ "no_stream" };
+ Maasha::UCSC::bed_put_entry( $record, $fh_out, $record->{ "BED_COLS" } );
- if ( $record->{ "REC_TYPE" } eq "PSL" )
- {
- Maasha::UCSC::psl_put_header( $fh_out ) if $first;
- Maasha::UCSC::psl_put_entry( $record, $fh_out );
-
- $first = 0;
+ $format = "BED" if not $format;
+ $columns = $record->{ "BED_COLS" } if not $columns;
+ }
+ elsif ( $record->{ "REC_TYPE" } eq "PATSCAN" and $record->{ "CHR" } )
+ {
+ Maasha::UCSC::bed_put_entry( $record, $fh_out, 6 );
- $format = "PSL" if not $format;
- }
- elsif ( $record->{ "REC_TYPE" } eq "BED" and $record->{ "SEC_STRUCT" } )
- {
- # chrom chromStart chromEnd name score strand size secStr conf
-
- print $fh_out join ( "\t",
- $record->{ "CHR" },
- $record->{ "CHR_BEG" },
- $record->{ "CHR_END" } + 1,
- $record->{ "Q_ID" },
- $record->{ "SCORE" },
- $record->{ "STRAND" },
- $record->{ "SIZE" },
- $record->{ "SEC_STRUCT" },
- $record->{ "CONF" },
- ), "\n";
-
- $format = "BED_SS" if not $format;
- }
- elsif ( $record->{ "REC_TYPE" } eq "BED" )
- {
- Maasha::UCSC::bed_put_entry( $record, $fh_out, $record->{ "BED_COLS" } );
+ $format = "BED" if not $format;
+ $columns = 6 if not $columns;
+ }
+ elsif ( $record->{ "REC_TYPE" } eq "BLAST" and $record->{ "S_ID" } =~ /^chr/ )
+ {
+ $record->{ "CHR" } = $record->{ "S_ID" };
+ $record->{ "CHR_BEG" } = $record->{ "S_BEG" };
+ $record->{ "CHR_END" } = $record->{ "S_END" };
+ $record->{ "SCORE" } = $record->{ "BIT_SCORE" } * 1000;
- $format = "BED" if not $format;
- $columns = $record->{ "BED_COLS" } if not $columns;
- }
- elsif ( $record->{ "REC_TYPE" } eq "PATSCAN" and $record->{ "CHR" } )
- {
- Maasha::UCSC::bed_put_entry( $record, $fh_out, 6 );
+ $format = "BED" if not $format;
+ $columns = 6 if not $columns;
- $format = "BED" if not $format;
- $columns = 6 if not $columns;
- }
- elsif ( $record->{ "REC_TYPE" } eq "BLAST" and $record->{ "S_ID" } =~ /^chr/ )
- {
- $record->{ "CHR" } = $record->{ "S_ID" };
- $record->{ "CHR_BEG" } = $record->{ "S_BEG" };
- $record->{ "CHR_END" } = $record->{ "S_END" };
- $record->{ "SCORE" } = $record->{ "BIT_SCORE" } * 1000;
+ Maasha::UCSC::bed_put_entry( $record, $fh_out );
+ }
+ elsif ( $record->{ "REC_TYPE" } eq "VMATCH" and $record->{ "S_ID" } =~ /^chr/i )
+ {
+ $record->{ "CHR" } = $record->{ "S_ID" };
+ $record->{ "CHR_BEG" } = $record->{ "S_BEG" };
+ $record->{ "CHR_END" } = $record->{ "S_END" };
+ $record->{ "SCORE" } = $record->{ "SCORE" } || 999;
+ $record->{ "SCORE" } = int( $record->{ "SCORE" } );
- $format = "BED" if not $format;
- $columns = 6 if not $columns;
+ $format = "BED" if not $format;
+ $columns = 6 if not $columns;
- Maasha::UCSC::bed_put_entry( $record, $fh_out );
- }
- elsif ( $record->{ "REC_TYPE" } eq "VMATCH" and $record->{ "S_ID" } =~ /^chr/i )
- {
- $record->{ "CHR" } = $record->{ "S_ID" };
- $record->{ "CHR_BEG" } = $record->{ "S_BEG" };
- $record->{ "CHR_END" } = $record->{ "S_END" };
- $record->{ "SCORE" } = $record->{ "SCORE" } || 999;
- $record->{ "SCORE" } = int( $record->{ "SCORE" } );
+ Maasha::UCSC::bed_put_entry( $record, $fh_out, 6 );
+ }
- $format = "BED" if not $format;
- $columns = 6 if not $columns;
+ if ( $i == $options->{ "chunk_size" } )
+ {
+ close $fh_out;
- Maasha::UCSC::bed_put_entry( $record, $fh_out, 6 );
+ if ( $format eq "BED" ) {
+ Maasha::UCSC::bed_upload_to_ucsc( $BP_TMP, $file, $options, $append );
+ } elsif ( $format eq "PSL" ) {
+ Maasha::UCSC::psl_upload_to_ucsc( $file, $options, $append );
}
- if ( $i == $options->{ "chunk_size" } )
- {
- close $fh_out;
-
- if ( $format eq "BED" ) {
- Maasha::UCSC::bed_upload_to_ucsc( $BP_TMP, $file, $options, $append );
- } elsif ( $format eq "PSL" ) {
- Maasha::UCSC::psl_upload_to_ucsc( $file, $options, $append );
- }
-
- unlink $file;
+ unlink $file;
- $first = 1;
+ $first = 1;
- $append = 1;
+ $append = 1;
- $fh_out = Maasha::Common::write_open( $file );
- }
-
- $i++;
+ $fh_out = Maasha::Common::write_open( $file );
}
+
+ $i++;
}
close $fh_out;
}
elsif ( $format eq "WIGGLE" )
{
+ $options->{ "visibility" } = "full";
+
+ $wig_file = "$options->{ 'table' }.wig";
+ $wib_file = "$options->{ 'table' }.wib";
+
+ $wib_dir = "$ENV{ 'HOME' }/ucsc/wib";
+
+ Maasha::Common::dir_create_if_not_exists( $wib_dir );
+
+ if ( $options->{ 'verbose' } ) {
+ `cd $BP_TMP && wigEncode $file $wig_file $wib_file`;
+ } else {
+ `cd $BP_TMP && wigEncode $file $wig_file $wib_file > /dev/null 2>&1`;
+ }
+
+ Maasha::Common::run( "mv", "$BP_TMP/$wib_file $wib_dir" );
+
+ unlink $file;
+
+ $file = $wig_file;
+
$type = "wig 0";
Maasha::UCSC::wiggle_upload_to_ucsc( $BP_TMP, $wib_dir, $file, $options );
{
# Martin A. Hansen, July 2008.
- # Cleans out any unused temporary files and direcotries in BP_TMP.
+ # Cleans out any unused temporary files and directories in BP_TMP.
# Returns nothing.
foreach $dir ( @dirs )
{
- if ( $dir =~ /(.+)_(\d+)_(\d+)_bp_tmp/ )
+ if ( $dir =~ /^$tmpdir\/(.+)_(\d+)_(\d+)_bp_tmp$/ )
{
$user = $1;
$sid = $2;
$pid = $3;
- if ( not Maasha::Common::process_running( $pid ) )
+ if ( $user eq Maasha::Common::get_user() )
{
- # print STDERR "Removing stale dir: $dir\n";
- Maasha::Common::dir_remove( $dir );
- }
- elsif ( $pid == $curr_pid )
- {
- # print STDERR "Removing current dir: $dir\n";
- Maasha::Common::dir_remove( $dir );
+ if ( not Maasha::Common::process_running( $pid ) )
+ {
+ # print STDERR "Removing stale dir: $dir\n";
+ Maasha::Common::dir_remove( $dir );
+ }
+ elsif ( $pid == $curr_pid )
+ {
+ # print STDERR "Removing current dir: $dir\n";
+ # Maasha::Common::dir_remove( $dir );
+ }
}
}
}