ngs_server 0.1 → 0.2

data/ext/vcftools/examples/merge-test-b.vcf

@@ -0,0 +1,17 @@
+##fileformat=VCFv4.0
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FORMAT=<ID=GL,Number=G,Type=Float,Description="Genotype Likelihood">
+##FILTER=<ID=q20,Description="Mapping quality below 20">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	B
+1	3062915	.	GTTT	GT	376	q20	DP=14	GT:GQ:DP:GL	0/1:376:14:-10,0,-10
+1	3106154	.	CAAAA	C	677	PASS	DP=15	GT:GQ:DP:GL	0/1:277:15:-10,0,-10
+1	3157410	.	GA	G	249	PASS	DP=11	GT:GQ:DP	0/1:49:11
+1	3162006	.	GAA	G	663	PASS	DP=19	GT:GQ:DP	0/1:589:19
+1	3177144	.	GT	G	460	PASS	DP=24	GT:GQ:DP	0/1:236:24
+1	3184885	.	TAAA	T	598	PASS	DP=16	GT:GQ:DP	0/1:435:16
+2	3188209	.	GA	G	162	.	DP=15	GT:GQ:DP	0/1:162:15
+3	3199812	.	G	GTT,GT	353	PASS	DP=19	GT:GQ:DP	1/2:188:19
+4	3212016	.	CTT	C	677	q20	DP=15	GT:GQ:DP	0/1:158:15

data/ext/vcftools/examples/merge-test-c.vcf

@@ -0,0 +1,15 @@
+##fileformat=VCFv4.0
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	C
+1	3062915	.	GTTT	G	388	.	DP=10	GT:GQ:DP	0/1:149:10
+1	3106154	.	CAAA	C	269	.	DP=9	GT:GQ:DP	1/1:25:9
+1	3157410	.	GA	G	212	.	DP=10	GT:GQ:DP	0/1:52:10
+1	3162006	.	GAA	G	558	.	DP=17	GT:GQ:DP	0/1:163:17
+1	3177144	.	GT	G	211	.	DP=14	GT:GQ:DP	0/1:151:14
+1	3199812	.	G	GT	.	.	.	GT	1/1
+2	3212016	.	CTT	C	613	.	DP=11	GT:GQ:DP	0/1:41:11
+3	3242491	.	TT	T	.	.	.	GT	1/1
+4	3291771	.	T	TAA,TAAA	336	.	DP=12	GT:GQ:DP	1/2:2:12

data/ext/vcftools/examples/merge-test.vcf.out

@@ -0,0 +1,31 @@
+Using column name 'A' for merge-test-a.vcf.gz:A
+Using column name 'B' for merge-test-b.vcf.gz:B
+Using column name 'C' for merge-test-c.vcf.gz:C
+##fileformat=VCFv4.1
+##FILTER=<ID=q10,Description="Quality below 10">
+##FORMAT=<ID=GL,Number=G,Type=Float,Description="Genotype Likelihood">
+##FILTER=<ID=q20,Description="Mapping quality below 20">
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##INFO=<ID=SF,Number=.,Type=String,Description="Source File (index to sourceFiles, f when filtered)">
+##INFO=<ID=AC,Number=.,Type=Integer,Description="Allele count in genotypes">
+##INFO=<ID=AN,Number=1,Type=Integer,Description="Total number of alleles in called genotypes">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	A	B	C
+1	3062915	.	GTTT	G,GT	856.67	q20;q10	AC=2,1;AN=6;DP=35;SF=0f,1f,2	GT:GL:DP:GQ	0/1:-20,-5,-20,.,.,.:35:409	0/2:-10,.,.,0,.,-10:14:376	0/1:.:10:149
+1	3106154	.	CAAAA	CA,C	912.67	PASS	AC=3,1;AN=6;DP=32;SF=0,1,2	GT:GL:DP:GQ	0/1:.:32:245	0/2:-10,.,.,0,.,-10:15:277	1/1:.:9:25
+1	3157410	.	GA	G	363.00	q10	AC=4;AN=6;DP=21;SF=0f,1,2	GT:DP:GQ	1/1:21:21	0/1:11:49	0/1:10:52
+1	3162006	.	GAA	G	745.67	PASS	AC=3;AN=6;DP=22;SF=0,1,2	GT:DP:GQ	0/1:22:212	0/1:19:589	0/1:17:163
+1	3177144	.	GT	G	466.00	PASS	AC=3;AN=6;DP=30;SF=0,1,2	GT:DP:GQ	0/1:30:150	0/1:24:236	0/1:14:151
+1	3184885	.	TAAAA	TA,T	422.00	PASS	AC=2,1;AN=4;DP=10;SF=0,1	GT:DP:GQ	1/2:10:12	0/1:16:435	.
+1	3199812	.	G	GT	.	.	AC=2;AN=2;SF=2	GT	.	.	1/1
+2	3188209	.	GA	G	162.00	.	AC=1;AN=2;DP=15;SF=1	GT:DP:GQ	.	0/1:15:162	.
+2	3199812	.	G	GTT,GT	481.00	PASS	AC=1,1;AN=2;DP=26;SF=0	GT:DP:GQ	1/2:26:322	.	.
+2	3212016	.	CTT	C	613.00	.	AC=1;AN=2;DP=11;SF=2	GT:DP:GQ	.	.	0/1:11:41
+3	3199812	.	G	GTT,GT	353.00	PASS	AC=1,1;AN=2;DP=19;SF=1	GT:DP:GQ	.	1/2:19:188	.
+3	3212016	.	CTT	C,CT	565.00	PASS	AC=1,1;AN=2;DP=26;SF=0	GT:DP:GQ	1/2:26:91	.	.
+3	3242491	.	TT	T	.	.	AC=2;AN=2;SF=2	GT	.	.	1/1
+4	3212016	.	CTT	C	677.00	q20	AC=1;AN=2;DP=15;SF=1f	GT:DP:GQ	.	0/1:15:158	.
+4	3258448	.	TACACACAC	T	325.00	PASS	AC=1;AN=2;DP=31;SF=0	GT:DP:GQ	0/1:31:325	.	.
+4	3291771	.	T	TAA,TAAA	336.00	.	AC=1,1;AN=2;DP=12;SF=2	GT:DP:GQ	.	.	1/2:12:2

data/ext/vcftools/examples/perl-api-1.pl

@@ -0,0 +1,46 @@
+#!/usr/bin/env perl
+#
+# Example code for generating a minimal VCF file using the perl API
+#
+# Author: pd3@sanger
+#
+
+use strict;
+use warnings;
+use Carp;
+use Vcf;
+
+my $sample = 'Sample1';
+my $vcf_out = Vcf->new();
+$vcf_out->add_columns($sample);
+$vcf_out->add_header_line({key=>'FORMAT',ID=>'GT',Number=>'1',Type=>'String',Description=>"Genotype"});
+$vcf_out->add_header_line({key=>'ALT',ID=>'DEL',Description=>"Deletion"});
+$vcf_out->add_header_line({key=>'ALT',ID=>'DEL:ME:ALU',Description=>"Deletion of ALU element"});
+$vcf_out->add_header_line({key=>'ALT',ID=>'DEL:ME:L1',Description=>"Deletion of L1 element"});
+$vcf_out->add_header_line({key=>'ALT',ID=>'DUP',Description=>"Duplication"});
+$vcf_out->add_header_line({key=>'INFO',ID=>'DP',Number=>1,Type=>'Integer',Description=>"Total Depth"});
+$vcf_out->add_header_line({key=>'INFO',ID=>'H2',Number=>0,Type=>'Flag',Description=>"HapMap2 membership"});
+
+print $vcf_out->format_header();
+
+my $pos = 1;
+for my $gt qw(A/A C/C <DEL>/C <DEL:ME:ALU>/<DEL:ME:ALU> <DEL:ME:L1>/<DEL:ME:L1> <DUP>/<DUP>)
+{
+    $pos++;
+
+    my %out;
+    $out{CHROM}  = '1';
+    $out{POS}    = $pos;
+    $out{ID}     = '.';
+    $out{ALT}    = [];
+    $out{REF}    = 'C';
+    $out{QUAL}   = '.';
+    $out{FILTER} = ['.'];
+    $out{INFO}   = { DP=>3, H2=>undef };
+    $out{FORMAT} = ['GT'];
+    $out{gtypes}{$sample}{GT} = $gt;
+
+    $vcf_out->format_genotype_strings(\%out);
+    print $vcf_out->format_line(\%out);
+}
+

data/ext/vcftools/examples/query-test.out

@@ -0,0 +1,6 @@
+1:100100	ref=G	alt=C	qual=0	1	A=G|C	B=G/C
+1:100200	ref=G	alt=C	qual=0	1	A=G|C	B=G/G
+1:100300	ref=G	alt=C	qual=0	1	A=C/C	B=./.
+1:100400	ref=C	alt=G,T	qual=35	1	A=G/G	B=C/T
+1:100500	ref=A	alt=G	qual=0	1	A=G/G	B=A/A
+1:100600	ref=C	alt=G	qual=0	1	A=G/G	B=C/C

data/ext/vcftools/examples/shuffle-test.vcf

@@ -0,0 +1,12 @@
+##fileformat=VCFv4.0
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	B	A
+1	100100	.	G	C	0	.	DP=1	GT:GQ:DP	0/1:40:1	0|1:40:1
+1	100200	.	G	C	0	.	DP=1	GT:GQ:DP	0/0:40:1	0|1:40:1
+1	100300	.	G	C	0	.	DP=1	GT:GQ:DP	./.:40:1	1/1:40:1
+1	100400	.	C	G,T	35	.	DP=1	GT:GQ:DP	0/2:40:1	1/1:41:1
+1	100500	.	A	G	0	.	DP=1	GT:GQ:DP	0/0:40:1	1/1:40:1
+1	100600	.	C	G	0	.	DP=1	GT:GQ:DP	0/0:40:1	1/1:40:1

data/ext/vcftools/examples/subset.SNPs.out

@@ -0,0 +1,10 @@
+##fileformat=VCFv4.0
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FILTER=<ID=q10,Description="Quality below 10">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	A
+1	100	.	G	C	1806	q10	DP=35	GT:GQ:DP	1/1:409:35
+1	120	.	T	G	628	q10	DP=21	GT:GQ:DP	0|1|1:21:21
+1	130	.	GAA	GTA	1016	PASS	DP=22	GT:GQ:DP	0/1:212:22

data/ext/vcftools/examples/subset.indels.out

@@ -0,0 +1,18 @@
+##fileformat=VCFv4.0
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FILTER=<ID=q10,Description="Quality below 10">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	A
+1	130	.	GAA	G	1016	PASS	DP=22	GT:GQ:DP	1/0:212:22
+1	140	.	GT	G	727	PASS	DP=30	GT:GQ:DP	0/1:150:30
+1	150	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10
+1	160	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10
+2	100	.	GTTT	G	1806	q10	DP=35	GT:GQ:DP	0/1:409:35
+2	110	.	CAAA	C	1792	PASS	DP=32	GT:GQ:DP	0/1:245:32
+2	120	.	GA	G	628	q10	DP=21	GT:GQ:DP	1/1:21:21
+2	130	.	GAA	G	1016	PASS	DP=22	GT:GQ:DP	0/1:212:22
+2	140	.	GT	G	727	PASS	DP=30	GT:GQ:DP	0/1:150:30
+2	150	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10
+2	160	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10

data/ext/vcftools/examples/subset.vcf

@@ -0,0 +1,21 @@
+##fileformat=VCFv4.0
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FILTER=<ID=q10,Description="Quality below 10">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	A	B
+1	100	.	G	C	1806	q10	DP=35	GT:GQ:DP	1/1:409:35	1/1:409:35
+1	110	.	C	.	1792	PASS	DP=32	GT:GQ:DP	0/0:245:32	0/0:245:32
+1	120	.	T	G	628	q10	DP=21	GT:GQ:DP	0|1|1:21:21	0|1|1:21:21
+1	130	.	GAA	G,GTA	1016	PASS	DP=22	GT:GQ:DP	1/2:212:22	1/2:212:22
+1	140	.	GT	G	727	PASS	DP=30	GT:GQ:DP	0/1:150:30	0/1:150:30
+1	150	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10	1/2:12:10
+1	160	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10	1/2:12:10
+2	100	.	GTTT	G	1806	q10	DP=35	GT:GQ:DP	0/1:409:35	0/1:409:35
+2	110	.	CAAA	C	1792	PASS	DP=32	GT:GQ:DP	0/1:245:32	0/1:245:32
+2	120	.	GA	G	628	q10	DP=21	GT:GQ:DP	1/1:21:21	1/1:21:21
+2	130	.	GAA	G	1016	PASS	DP=22	GT:GQ:DP	0/1:212:22	0/1:212:22
+2	140	.	GT	G	727	PASS	DP=30	GT:GQ:DP	0/1:150:30	0/1:150:30
+2	150	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10	1/2:12:10
+2	160	.	TAAAA	TA,T	246	PASS	DP=10	GT:GQ:DP	1/2:12:10	1/2:12:10

data/ext/vcftools/examples/valid-3.3.vcf

@@ -0,0 +1,30 @@
+##fileformat=VCFv3.3
+##fileDate=20090805
+##phasing=partial
+##test meta
+##INFO=NS,1,Integer,"Number of Samples With Data"
+##INFO=DP,1,Integer,"Total Depth"
+##INFO=AN,1,Integer,"Total number of alleles in called genotypes"
+##INFO=AC,-1,Integer,"Allele count in genotypes, for each ALT allele, in the same order as listed"
+##INFO=AF,-1,Float,"Allele Frequency"
+##INFO=AA,1,String,"Ancestral Allele"
+##INFO=DB,0,Flag,"dbSNP membership, build 129"
+##INFO=H2,0,Flag,"HapMap2 membership"
+##FILTER=q10,"Quality below 10"
+##FILTER=s50,"Less than 50% of samples have data"
+##FORMAT=GT,1,String,"Genotype"
+##FORMAT=GQ,1,Integer,"Genotype Quality"
+##FORMAT=DP,1,Integer,"Read Depth"
+##FORMAT=HQ,2,Integer,"Haplotype Quality"
+##FORMAT=DS,1,Float,"Alternative Allele Dosage"
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	NA00001	NA00002	NA00003
+19	111	.	A	C	9.6	.	.	GT:HQ	0|0:10,10	0|0:10,10	0\1:3,3
+19	112	.	A	G	10	.	.	GT:HQ	0|0:10,10	0|0:10,10	0\1:3,3
+20	14370	rs6054257	G	A	29	0	NS=3;DP=14;AF=0.5;DB;H2	GT:GQ:DP:HQ	0|0:48:1:51,51	1|0:48:8:51,51	1/1:43:5:-1,-1
+20	17330	.	T	A	3	q10	NS=3;DP=11;AF=0.017	GT:GQ:DP:HQ	0|0:49:3:58,50	0|1:3:5:65,3	0/0:41:3:-1,-1
+20	1110696	rs6040355	A	G,T	67	0	NS=2;DP=10;AF=0.333,0.667;AA=T;DB	GT:GQ:DP:HQ	1|2:21:6:23,27	2|1:2:0:18,2	2/2:35:4:-1,-1
+20	1230237	.	T	.	47	0	NS=3;DP=13;AA=T	GT:GQ:DP:HQ	0|0:54:-1:56,60	0|0:48:4:51,51	0/0:61:2:-1,-1
+20	1234567	microsat1	G	D4,IGA	50	0	NS=3;DP=9;AA=G;AN=6;AC=3,1	GT:GQ:DP	0/1:-1:4	0/2:17:2	1/1:40:3
+20	1235237	.	T	.	-1	.	.	GT	0\0	0|0	./.
+X	10	rsTest	A	T	10	.	.	GT:DS	0:0.1	0/1:0.5	0|1:0.5
+X	11	rsTest2	T	A,G	10	q10;s50	.	GT:DP:GQ	0:3:10	.:5:20	0:3:10

data/ext/vcftools/examples/valid-4.0.vcf

@@ -0,0 +1,34 @@
+##fileformat=VCFv4.0
+##fileDate=20090805
+##source=myImputationProgramV3.1
+##reference=1000GenomesPilot-NCBI36
+##phasing=partial
+##INFO=<ID=NS,Number=1,Type=Integer,Description="Number of Samples With Data">
+##INFO=<ID=AN,Number=1,Type=Integer,Description="Total number of alleles in called genotypes">
+##INFO=<ID=AC,Number=.,Type=Integer,Description="Allele count in genotypes, for each ALT allele, in the same order as listed">
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##INFO=<ID=AF,Number=.,Type=Float,Description="Allele Frequency">
+##INFO=<ID=AA,Number=1,Type=String,Description="Ancestral Allele">
+##INFO=<ID=DB,Number=0,Type=Flag,Description="dbSNP membership, build 129">
+##INFO=<ID=H2,Number=0,Type=Flag,Description="HapMap2 membership">
+##FILTER=<ID=q10,Description="Quality below 10">
+##FILTER=<ID=s50,Description="Less than 50% of samples have data">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FORMAT=<ID=HQ,Number=2,Type=Integer,Description="Haplotype Quality">
+##ALT=<ID=DEL:ME:ALU,Description="Deletion of ALU element">
+##ALT=<ID=CNV,Description="Copy number variable region">
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	NA00001	NA00002	NA00003
+19	111	.	A	C	9.6	.	.	GT:HQ	0|0:10,10	0|0:10,10	0/1:3,3
+19	112	.	A	G	10	.	.	GT:HQ	0|0:10,10	0|0:10,10	0/1:3,3
+20	14370	rs6054257	G	A	29	PASS	NS=3;DP=14;AF=0.5;DB;H2	GT:GQ:DP:HQ	0|0:48:1:51,51	1|0:48:8:51,51	1/1:43:5:.,.
+20	17330	.	T	A	3	q10	NS=3;DP=11;AF=0.017	GT:GQ:DP:HQ	0|0:49:3:58,50	0|1:3:5:65,3	0/0:41:3:.,.
+20	1110696	rs6040355	A	G,T	67	PASS	NS=2;DP=10;AF=0.333,0.667;AA=T;DB	GT:GQ:DP:HQ	1|2:21:6:23,27	2|1:2:0:18,2	2/2:35:4:.,.
+20	1230237	.	T	.	47	PASS	NS=3;DP=13;AA=T	GT:GQ:DP:HQ	0|0:54:.:56,60	0|0:48:4:51,51	0/0:61:2:.,.
+20	1234567	microsat1	G	GA,GAC	50	PASS	NS=3;DP=9;AA=G;AN=6;AC=3,1	GT:GQ:DP	0/1:.:4	0/2:17:2	1/1:40:3
+20	1235237	.	T	.	.	.	.	GT	0/0	0|0	./.
+X	9	.	A	T	12.1	.	.	GT	0	0/1	1/0
+X	10	rsTest	AC	A,ATG	10	PASS	.	GT	0	0/1	0|2
+X	11	rsTest2	T	A,<DEL:ME:ALU>	10	q10;s50	.	GT:DP:GQ	.:3:10	./.	0|2:3
+X	12	.	T	A	13	.	.	GT	0	1/0	1/1

data/ext/vcftools/examples/valid-4.0.vcf.stats

@@ -0,0 +1,104 @@
+$VAR1 = {
+          'samples' => {
+                         'NA00002' => {
+                                        'indel_count' => 2,
+                                        'indel' => {
+                                                     '2' => 1,
+                                                     '-1' => 1
+                                                   },
+                                        'het_RA_count' => 6,
+                                        'snp_count' => 5,
+                                        'count' => 11,
+                                        'hom_RR_count' => 4,
+                                        'ref' => 10,
+                                        'missing' => 1,
+                                        'private' => 1,
+                                        'phased' => 7,
+                                        'het_AA_count' => 1,
+                                        'snp' => {
+                                                   'A>G' => 1,
+                                                   'A>T' => 2,
+                                                   'T>A' => 2,
+                                                   'G>A' => 1
+                                                 },
+                                        'unphased' => 4,
+                                        'ref_count' => 10
+                                      },
+                         'NA00001' => {
+                                        'het_RA_count' => 1,
+                                        'indel_count' => 1,
+                                        'indel' => {
+                                                     '1' => 1
+                                                   },
+                                        'snp_count' => 1,
+                                        'count' => 11,
+                                        'hom_RR_count' => 9,
+                                        'ref' => 10,
+                                        'missing' => 1,
+                                        'phased' => 6,
+                                        'het_AA_count' => 1,
+                                        'snp' => {
+                                                   'A>T' => 1,
+                                                   'A>G' => 1
+                                                 },
+                                        'unphased' => 5,
+                                        'ref_count' => 10
+                                      },
+                         'NA00003' => {
+                                        'hom_AA_count' => 4,
+                                        'other_count' => 2,
+                                        'indel_count' => 1,
+                                        'indel' => {
+                                                     '1' => 1
+                                                   },
+                                        'het_RA_count' => 5,
+                                        'snp_count' => 6,
+                                        'count' => 11,
+                                        'hom_RR_count' => 2,
+                                        'ref' => 7,
+                                        'missing' => 1,
+                                        'private' => 3,
+                                        'phased' => 2,
+                                        'other' => 2,
+                                        'snp' => {
+                                                   'A>C' => 1,
+                                                   'A>T' => 2,
+                                                   'A>G' => 1,
+                                                   'T>A' => 1,
+                                                   'G>A' => 1
+                                                 },
+                                        'ref_count' => 7,
+                                        'unphased' => 9
+                                      }
+                       },
+          'all' => {
+                     'other_count' => 2,
+                     'indel_count' => 2,
+                     'indel' => {
+                                  '1' => 1,
+                                  '2' => 1,
+                                  '-1' => 1
+                                },
+                     'snp_count' => 8,
+                     'count' => 12,
+                     'ref' => 2,
+                     'nalt_0' => 2,
+                     'nalt_1' => 7,
+                     'shared' => {
+                                   '1' => 4,
+                                   '0' => 2,
+                                   '3' => 2,
+                                   '2' => 4
+                                 },
+                     'other' => 2,
+                     'nalt_2' => 3,
+                     'snp' => {
+                                'A>C' => 1,
+                                'A>T' => 2,
+                                'A>G' => 2,
+                                'T>A' => 3,
+                                'G>A' => 1
+                              },
+                     'ref_count' => 2
+                   }
+        };

data/ext/vcftools/examples/valid-4.1.vcf

@@ -0,0 +1,37 @@
+##fileformat=VCFv4.1
+##fileDate=20090805
+##source=myImputationProgramV3.1
+##phasing=partial
+##INFO=<ID=NS,Number=1,Type=Integer,Description="Number of Samples With Data">
+##INFO=<ID=DP,Number=1,Type=Integer,Description="Total Depth">
+##INFO=<ID=AF,Number=A,Type=Float,Description="Allele Frequency">
+##INFO=<ID=AA,Number=1,Type=String,Description="Ancestral Allele">
+##INFO=<ID=DB,Number=0,Type=Flag,Description="dbSNP membership, build 129">
+##INFO=<ID=H2,Number=0,Type=Flag,Description="HapMap2 membership">
+##INFO=<ID=SVTYPE,Number=1,Type=String,Description="Type of structural variant">
+##FILTER=<ID=q10,Description="Quality below 10 (\"with quotes\")">
+##FILTER=<ID=s50,Description="Less than 50% of samples have data (with \\ backslash)">
+##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
+##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype Quality">
+##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Read Depth">
+##FORMAT=<ID=HQ,Number=2,Type=Integer,Description="Haplotype Quality">
+##reference=file:/lustre/scratch105/projects/g1k/ref/main_project/human_g1k_v37.fasta
+##contig=<ID=19,length=59128983,md5=1aacd71f30db8e561810913e0b72636d,species="Homo Sapiens">
+##contig=<ID=20,length=63025520,md5=0dec9660ec1efaaf33281c0d5ea2560f,species="Homo Sapiens">
+##SAMPLE=<ID=Blood,Genomes=Germline,Mixture=1.,Description="Patient germline genome">
+##SAMPLE=<ID=TissueSample,Genomes=Germline;Tumor,Mixture=.3;.7,Description="Patient germline genome;Patient tumor genome">
+##PEDIGREE=<Derived=ID2,Original=ID1>
+##PEDIGREE=<Child=CHILD-GENOME-ID,Mother=MOTHER-GENOME-ID,Father=FATHER-GENOME-ID>
+##pedigreeDB=url
+#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	NA00001	NA00002	NA00003
+19	14370	rs6054257	G	A	29	PASS	NS=3;DP=14;AF=0.5;DB;H2	GT:GQ:DP:HQ	0|0:48:1:51,51	1|0:48:8:51,51	1/1:43:5:.,.
+20	17330	.	T	A	3	q10	NS=3;DP=11;AF=0.017	GT:GQ:DP:HQ	0|0:49:3:58,50	0|1:3:5:65,3	0/0:41:3
+20	1110696	rs6040355	A	G,T	67	PASS	NS=2;DP=10;AF=0.333,0.667;AA=T;DB	GT:GQ:DP:HQ	1|2:21:6:23,27	2|1:2:0:18,2	2/2:35:4
+20	1230237	.	T	.	47	PASS	NS=3;DP=13;AA=T	GT:GQ:DP:HQ	0|0:54:7:56,60	0|0:48:4:51,51	0/0:61:2
+20	1234567	microsat1	GTC	G,GTCTC	50	PASS	NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/2:17:2	1/1:40:3
+20	2234567	.	C	[13:123457[ACGC	50	PASS	SVTYPE=BND;NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/1:17:2	1/1:40:3
+20	2234568	.	C	.CT	50	PASS	SVTYPE=BND;NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/1:17:2	1/1:40:3
+20	2234569	.	C	TC.	50	PASS	SVTYPE=BND;NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/1:17:2	1/1:40:3
+20	3234569	.	C	<INV>	50	PASS	SVTYPE=BND;NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/1:17:2	1/1/0:40:3
+20	4234569	.	N	.[13:123457[	50	PASS	SVTYPE=BND;NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/1:17:2	./.:40:3
+20	5234569	.	N	[13:123457[.	50	PASS	SVTYPE=BND;NS=3;DP=9;AA=G	GT:GQ:DP	0/1:35:4	0/1:17:2	1/1:40:3

data/ext/vcftools/extconf.rb

@@ -0,0 +1,2 @@
+
+puts "nothing to do"

data/ext/vcftools/perl/.svn/all-wcprops

@@ -0,0 +1,149 @@
+K 25
+svn:wc:ra_dav:version-url
+V 35
+/svnroot/vcftools/!svn/ver/618/perl
+END
+test.t
+K 25
+svn:wc:ra_dav:version-url
+V 42
+/svnroot/vcftools/!svn/ver/601/perl/test.t
+END
+FaSlice.pm
+K 25
+svn:wc:ra_dav:version-url
+V 46
+/svnroot/vcftools/!svn/ver/618/perl/FaSlice.pm
+END
+vcf-query
+K 25
+svn:wc:ra_dav:version-url
+V 45
+/svnroot/vcftools/!svn/ver/536/perl/vcf-query
+END
+vcf-subset
+K 25
+svn:wc:ra_dav:version-url
+V 46
+/svnroot/vcftools/!svn/ver/578/perl/vcf-subset
+END
+vcf-concat
+K 25
+svn:wc:ra_dav:version-url
+V 46
+/svnroot/vcftools/!svn/ver/512/perl/vcf-concat
+END
+fill-ref-md5
+K 25
+svn:wc:ra_dav:version-url
+V 48
+/svnroot/vcftools/!svn/ver/502/perl/fill-ref-md5
+END
+vcf-fix-newlines
+K 25
+svn:wc:ra_dav:version-url
+V 52
+/svnroot/vcftools/!svn/ver/535/perl/vcf-fix-newlines
+END
+vcf-annotate
+K 25
+svn:wc:ra_dav:version-url
+V 48
+/svnroot/vcftools/!svn/ver/617/perl/vcf-annotate
+END
+ChangeLog
+K 25
+svn:wc:ra_dav:version-url
+V 45
+/svnroot/vcftools/!svn/ver/486/perl/ChangeLog
+END
+Vcf.pm
+K 25
+svn:wc:ra_dav:version-url
+V 42
+/svnroot/vcftools/!svn/ver/617/perl/Vcf.pm
+END
+vcf-convert
+K 25
+svn:wc:ra_dav:version-url
+V 47
+/svnroot/vcftools/!svn/ver/341/perl/vcf-convert
+END
+vcf-isec
+K 25
+svn:wc:ra_dav:version-url
+V 44
+/svnroot/vcftools/!svn/ver/606/perl/vcf-isec
+END
+vcf-validator
+K 25
+svn:wc:ra_dav:version-url
+V 49
+/svnroot/vcftools/!svn/ver/449/perl/vcf-validator
+END
+vcf-compare
+K 25
+svn:wc:ra_dav:version-url
+V 47
+/svnroot/vcftools/!svn/ver/617/perl/vcf-compare
+END
+tab-to-vcf
+K 25
+svn:wc:ra_dav:version-url
+V 46
+/svnroot/vcftools/!svn/ver/579/perl/tab-to-vcf
+END
+vcf-to-tab
+K 25
+svn:wc:ra_dav:version-url
+V 46
+/svnroot/vcftools/!svn/ver/544/perl/vcf-to-tab
+END
+fill-an-ac
+K 25
+svn:wc:ra_dav:version-url
+V 46
+/svnroot/vcftools/!svn/ver/341/perl/fill-an-ac
+END
+vcf-sort
+K 25
+svn:wc:ra_dav:version-url
+V 44
+/svnroot/vcftools/!svn/ver/403/perl/vcf-sort
+END
+fill-aa
+K 25
+svn:wc:ra_dav:version-url
+V 43
+/svnroot/vcftools/!svn/ver/341/perl/fill-aa
+END
+vcf-shuffle-cols
+K 25
+svn:wc:ra_dav:version-url
+V 52
+/svnroot/vcftools/!svn/ver/341/perl/vcf-shuffle-cols
+END
+VcfStats.pm
+K 25
+svn:wc:ra_dav:version-url
+V 47
+/svnroot/vcftools/!svn/ver/476/perl/VcfStats.pm
+END
+vcf-stats
+K 25
+svn:wc:ra_dav:version-url
+V 45
+/svnroot/vcftools/!svn/ver/477/perl/vcf-stats
+END
+vcf-merge
+K 25
+svn:wc:ra_dav:version-url
+V 45
+/svnroot/vcftools/!svn/ver/608/perl/vcf-merge
+END
+Makefile
+K 25
+svn:wc:ra_dav:version-url
+V 44
+/svnroot/vcftools/!svn/ver/601/perl/Makefile
+END