ngs_server 0.1 → 0.2

data/ext/vcftools/perl/.svn/prop-base/fill-aa.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/fill-an-ac.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/fill-ref-md5.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/tab-to-vcf.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/test.t.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-annotate.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-compare.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-concat.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-convert.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-fix-newlines.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-isec.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-merge.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-query.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-shuffle-cols.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-sort.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-stats.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-subset.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-to-tab.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/prop-base/vcf-validator.svn-base

@@ -0,0 +1,5 @@
+K 14
+svn:executable
+V 1
+*
+END

data/ext/vcftools/perl/.svn/text-base/ChangeLog.svn-base

@@ -0,0 +1,84 @@
+2011-04-04 14:00  petr.danecek@sanger
+    * VCFtools now support VCFv4.1
+    * fill-ref-md5: New tool backfilling sequence MD5s into VCF header
+    * Renamed merge-vcf, compare-vcf etc. to consistent naming vcf-merge, vcf-compare
+    * vcf-merge: Now merging also GL and other Number=[AG] tags
+    * vcf-compare: Comparing indel haplotypes
+
+2011-02-21 12:31  petr.danecek@sanger
+    * vcf-stats: new -s option to speed up parsing when stats computed for selected samples only
+    * merge-vcf: allow to merge arbitrary chunks; -c option now deprecated, use -r instead
+    * compare-vcf: change in output format and more detailed comparison
+
+2011-02-17 17:36  petr.danecek@sanger
+    * vcf-stats: allow querying stats of individual samples
+
+2011-02-16 12:07  petr.danecek@sanger
+    * vcf-stats: major revision
+    * vcf-annotate: more filtering options
+
+2011-02-04 14:43  petr
+    * merge-vcf: if possible, calculate AC,AN even for sites without genotypes
+
+2011-02-03 15:04  petr
+    * merge-vcf: fixed a bug introduced by the previous fix.
+
+2011-02-02 21:02  petr
+    * merge-vcf: fixed a bug in merging indel ALTs. Only VCFs without samples were affected. 
+
+2011-01-28 15:38  petr
+    * vcf-subset: new option for printing rows with calls private to the subset group
+
+2011-01-24 13:38  petr
+    * Vcf.pm: uppercase floating point number expressions (such as
+    1.0382033E-6) now pass validation
+
+2011-01-20 08:28  petr
+    * vcf-concat: print header also for empty VCFs with the -s option
+
+2011-01-04 08:59  petr
+    * vcf-isec, vcf-sort, Vcf.pm: replaced "zcat" by "gunzip -c"
+
+2010-12-22 14:18  petr
+    * vcf-annotate: New --SnpCluster option
+    * Vcf.pm: new sub add_filter()
+
+2010-12-15 13:44  petr
+    * vcf-isec: By default output records from all files with unique positions
+    (duplicate records from the same file still should be printed). With the -o
+    switch, only positions from the left-most file will be printed.
+
+2010-12-09 14:48  petr
+    * query-vcf: Output 'True' for Flag tags when present and . when absent
+    * vcf-annotate: Fix: the command line eats quotes when they are not escaped
+
+2010-12-08 12:06  petr
+    * Vcf.pm: throw an error when tabix fails.
+    * query-vcf: enable streaming of files when region is not specified.
+
+2010-12-02 11:53  petr
+    * Vcf.pm: allow ALT alleles which are not present in samples
+    * vcf-isec: Multiple files can be created simultaneously with all possible
+    isec combinations. Suitable for Venn Diagram analysis.
+    * merge-vcf: Do not remove ALT alleles if no samples are present
+    * merge-vcf: Do FILTER merging more intelligently.
+    * merge-vcf: Join the QUAL column: use average value weighted by the number of samples.
+
+2010-11-28 08:34  petr
+    * vcf-concat: Partial sort
+    * vcf-validator: Added -u option
+    * VcfStats.pm: dump_counts
+
+2010-11-27 13:04  petr
+    * vcf-subset: Filter variants by type
+
+2010-11-26 09:08  petr
+    * vcf-annotate: Added possibility to read header descriptions from a file
+
+2010-11-24 13:25  petr
+    * Fix in Vcf.pm:fill_ref_alt_mapping. VCF files processed with merge-vcf were
+    affected when containing IDs in the ALT column.
+
+2010-11-23 13:12  petr
+    * Major revamp of Vcf.pm to allow better inheritance. Problems likely.
+

data/ext/vcftools/perl/.svn/text-base/FaSlice.pm.svn-base

@@ -0,0 +1,214 @@
+# Author: petr.danecek@sanger
+#
+
+=head1 NAME
+
+FaSlice.pm.  Module for cached access to fasta sequences, employs samtools faidx. 
+
+=head1 SYNOPSIS
+
+use FaSlice;
+
+my $fa = FaSlice->new(file=>'ref.fa');
+$fa->get_base(1,12345);
+$fa->get_slice(1,12345,54321);
+
+=cut
+
+
+package FaSlice;
+
+use strict;
+use warnings;
+use Carp;
+
+=head2 new
+
+    About   : Creates new FaSlice object.
+    Usage   : my $fa = FaSlice->new(file=>'ref.fa');
+    Args    : file   .. the fasta file
+              oob    .. out-of-bounds requests: one of 'throw' (throws), 'N' (fills the missing bases by Ns), or '' (returns empty string, default)
+              size   .. size of the cached chunk read by samtools faidx (1_000_000)
+
+=cut
+
+sub new
+{
+    my ($class,@args) = @_;
+    my $self = @args ? {@args} : {};
+    bless $self, ref($class) || $class;
+    if ( !$$self{file} ) { $self->throw("Missing the parameter file\n"); }
+    $$self{chr}  = undef;
+    $$self{from} = undef;
+    $$self{to}   = undef;
+    if ( !$$self{size} ) { $$self{size}=1_000_000; }
+    $$self{ncache_missed} = 0;
+    $$self{nqueries} = 0;
+    if ( !exists($$self{oob}) ) { $$self{oob}=''; }
+    if ( $$self{oob} ne '' && $$self{oob} ne 'throw' && $$self{oob} ne 'N' ) { $self->throw("The value of oob not recognised: [$$self{oob}]"); }
+    $self->chromosome_naming($$self{file});
+    return $self;
+}
+
+sub throw
+{
+    my ($self,@msg) = @_;
+    confess(@msg);
+}
+
+sub cmd
+{
+    my ($self,$cmd) = @_;
+    my @out = `$cmd`;
+    if ( $? )
+    {
+        my @msg = ();
+        push @msg, qq[The command "$cmd" returned non-zero status $?];
+        if ( $! )
+        {
+            push @msg, ": $!\n";
+        }
+        else
+        {
+            push @msg, ".\n";
+        }
+        if ( scalar @out )
+        {
+            push @msg, @out;
+        }
+        $self->throw(@msg);
+    }
+    return (@out);
+}
+
+# Read the first file of the fasta file and make a guess: Are all chromosomes
+#   names as 'chr1','chr2',etc or just '1','2',...?
+# Future TODO: more robust chromosome name mapping?
+sub chromosome_naming
+{
+    my ($self,$fa_file) = @_;
+    open(my $fh,'<',"$fa_file.fai") or $self->throw("$fa_file.fai: $!");
+    my $line=<$fh>;
+    if ( !($line=~/^(chr)?\S+\t/) ) { chomp($line); $self->throw("FIXME: the sequence names not in '>(chr)?\\S+' format [$line] ... $fa_file.fai\n"); }
+    close($fh); 
+    $$self{chr_naming} = defined $1 ? $1 : '';
+}
+
+
+sub read_chunk
+{
+    my ($self,$chr,$pos) = @_;
+    $$self{chr}  = $chr;
+    $chr =~ s/^chr//;
+    $chr = $$self{chr_naming}.$chr;
+    my $to = $pos + $$self{size};
+    my $cmd = "samtools faidx $$self{file} $chr:$pos-$to";
+    my @out = $self->cmd($cmd) or $self->throw("$cmd: $!");
+    my $line = shift(@out);
+    if ( !($line=~/^>$chr:(\d+)-(\d+)/) ) { $self->throw("Could not parse: $line"); }
+    $$self{from} = $1;
+    my $chunk = '';
+    while ($line=shift(@out))
+    {
+        chomp($line);
+        $chunk .= $line;
+    }
+    $$self{to} = $$self{from} + length($chunk) - 1;
+    $$self{chunk} = $chunk;
+    $$self{ncache_missed}++;
+    return;
+}
+
+=head2 get_base
+
+    About   : Retrieves base at the given chromosome and position
+    Usage   : my $fa = FaSlice->new(file=>'ref.fa'); $fa->get_base(1,12345);
+    Args    : chromosome
+              1-based coordinate
+
+=cut
+
+sub get_base
+{
+    my ($self,$chr,$pos) = @_;
+    if ( !$$self{chr} || $chr ne $$self{chr} || $pos<$$self{from} || $pos>$$self{to} )
+    {
+        $self->read_chunk($chr,$pos);
+    }
+    $$self{nqueries}++;
+    my $idx = $pos - $$self{from};
+    if ( $$self{from}>$$self{to} ) 
+    { 
+        if ( $$self{oob} eq '' ) { return ''; }
+        elsif ( $$self{oob} eq 'N' ) { return 'N'; }
+        $self->throw("No such site $chr:$pos in $$self{file}\n"); 
+    }
+    return substr($$self{chunk},$idx,1);
+}
+
+
+=head2 get_slice
+
+    About   : Retrieves region
+    Usage   : my $fa = FaSlice->new(file=>'ref.fa'); $fa->get_base(1,12345,54321);
+    Args    : chromosome
+              1-based coordinate
+
+=cut
+
+sub get_slice
+{
+    my ($self,$chr,$from,$to) = @_;
+    if ( $to-$from >= $$self{size} ) { $self->throw("Too big region requested, $from-$to >= $$self{size}\n"); }
+    if ( $from>$to ) { $self->throw("Expected $from>$to\n"); }
+    if ( !$$self{chr} || $chr ne $$self{chr} || $from<$$self{from} || $to>$$self{to} )
+    {
+        $self->read_chunk($chr,$from);
+    }
+    $$self{nqueries}++;
+
+    if ( $$self{from}>$$self{to} || $$self{from}>$from || $$self{to}<$to )
+    {
+        if ( $$self{oob} eq 'throw' ) { $self->throw("The region out of bounds $chr:$from-$to in $$self{file}\n"); }
+        elsif ( $$self{oob} eq '' ) { return ''; }
+
+        if ( $$self{from}>$$self{to} ) { return 'N' x ($to-$from+1); }
+        if ( $$self{from}>$to ) { $self->throw("FIXME: this shouldn't happen $chr:$from-$to .. $$self{from},$$self{to} .. $$self{file}"); }
+
+        my $lfill = '';
+        my $rfill = '';
+        if ( $$self{from}>$from ) { $lfill = 'N' x ($$self{from}-$from); $from=$$self{from}; }
+        if ( $$self{to}<$to ) { $rfill = 'N' x ($to-$$self{to}); $to=$$self{to}; }
+        return $lfill . substr($$self{chunk},$from-$$self{from},$to-$from+1) . $rfill;
+    }
+    return substr($$self{chunk},$from-$$self{from},$to-$from+1);
+}
+
+
+# http://www.illumina.com/documents/products/technotes/technote_topbot.pdf
+sub illumina_alleles_TOP_to_ref
+{
+    my ($self,$a1,$a2,$chr,$pos,$ref) = @_;
+    my %map = (A=>'T', C=>'G', G=>'C', T=>'A');
+    my %top = ( 
+            A=>{A=>-2,C=> 1,G=> 1,T=>-1}, 
+            C=>{A=> 1,C=>-2,G=>-1,T=> 0}, 
+            G=>{A=> 1,C=>-1,G=>-2,T=> 0}, 
+            T=>{A=>-1,C=> 0,G=> 0,T=>-2} ); 
+
+    my $stat = $top{$a1}{$a2};
+    if ( $stat==-2 ) { $self->throw("Expected two different bases, got $a1 and $a2.\n"); }
+    if ( $stat==-1 )
+    {
+        # Now we should do the sequence walking to see if the reference is TOP or BOT,
+        #   but we do not this in ill-to-vcf: C/G would become G/C and A/T would become T/A.
+        return ($a1,$a2);
+    }
+    if ( $stat==0 ) { $self->throw("Expected Illumina TOP, got $a1 and $a2.\n"); }
+    if ( $ref eq $a1 or $ref eq $a2 ) { return ($a1,$a2); }
+    return ($map{$a1},$map{$a2});
+}
+
+
+1;
+

data/ext/vcftools/perl/.svn/text-base/Makefile.svn-base

@@ -0,0 +1,12 @@
+
+BIN = vcf-compare fill-aa vcf-annotate vcf-merge vcf-isec vcf-stats vcf-to-tab fill-an-ac \
+    vcf-query vcf-convert vcf-subset vcf-validator vcf-concat vcf-sort
+MOD = FaSlice.pm Vcf.pm
+
+install:
+	    @for i in $(BIN); do cp $(CURDIR)/$$i $(BINDIR)/$$i; done; \
+        for i in $(MOD); do cp $(CURDIR)/$$i $(MODDIR)/$$i; done; 
+
+clean:
+		@for i in $(BIN); do rm -f $(BINDIR)/$$i; done; \
+        for i in $(MOD); do rm -f $(MODDIR)/$$i; done;