bio-vcf 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- data/Gemfile +16 -0
- data/Gemfile.lock +78 -0
- data/LICENSE.txt +20 -0
- data/README.md +120 -0
- data/Rakefile +58 -0
- data/VERSION +1 -0
- data/bin/bio-vcf +121 -0
- data/features/step_definitions/bio-vcf_steps.rb +0 -0
- data/features/support/env.rb +13 -0
- data/lib/bio-vcf.rb +15 -0
- data/lib/bio-vcf/vcf.rb +3 -0
- data/lib/bio-vcf/vcfgenotypefield.rb +43 -0
- data/lib/bio-vcf/vcfheader.rb +42 -0
- data/lib/bio-vcf/vcfline.rb +11 -0
- data/lib/bio-vcf/vcfrecord.rb +55 -0
- data/test/data/input/somaticsniper.vcf +87 -0
- metadata +163 -0
data/Gemfile
ADDED
@@ -0,0 +1,16 @@
|
|
1
|
+
source "http://rubygems.org"
|
2
|
+
# Add dependencies required to use your gem here.
|
3
|
+
# Example:
|
4
|
+
# gem "activesupport", ">= 2.3.5"
|
5
|
+
|
6
|
+
# Add dependencies to develop your gem here.
|
7
|
+
# Include everything needed to run rake, tests, features, etc.
|
8
|
+
group :development do
|
9
|
+
gem "minitest", "~> 5.0.7"
|
10
|
+
gem "rspec", "~> 2.8.0"
|
11
|
+
gem "cucumber", ">= 0"
|
12
|
+
gem "jeweler", "~> 1.8.4", :git => "https://github.com/technicalpickles/jeweler.git"
|
13
|
+
gem "bundler", ">= 1.0.21"
|
14
|
+
# gem "bio", ">= 1.4.2"
|
15
|
+
gem "rdoc", "~> 3.12"
|
16
|
+
end
|
data/Gemfile.lock
ADDED
@@ -0,0 +1,78 @@
|
|
1
|
+
GIT
|
2
|
+
remote: https://github.com/technicalpickles/jeweler.git
|
3
|
+
revision: f7e0a55a207d83f56637dd8fbabf26a803410faf
|
4
|
+
specs:
|
5
|
+
jeweler (1.8.7)
|
6
|
+
builder
|
7
|
+
bundler (~> 1.0)
|
8
|
+
git (>= 1.2.5)
|
9
|
+
github_api (= 0.10.1)
|
10
|
+
highline (>= 1.6.15)
|
11
|
+
nokogiri (= 1.5.10)
|
12
|
+
rake
|
13
|
+
rdoc
|
14
|
+
|
15
|
+
GEM
|
16
|
+
remote: http://rubygems.org/
|
17
|
+
specs:
|
18
|
+
addressable (2.3.5)
|
19
|
+
builder (3.2.2)
|
20
|
+
cucumber (1.3.2)
|
21
|
+
builder (>= 2.1.2)
|
22
|
+
diff-lcs (>= 1.1.3)
|
23
|
+
gherkin (~> 2.12.0)
|
24
|
+
multi_json (~> 1.3)
|
25
|
+
diff-lcs (1.1.3)
|
26
|
+
faraday (0.8.8)
|
27
|
+
multipart-post (~> 1.2.0)
|
28
|
+
gherkin (2.12.1)
|
29
|
+
multi_json (~> 1.3)
|
30
|
+
git (1.2.6)
|
31
|
+
github_api (0.10.1)
|
32
|
+
addressable
|
33
|
+
faraday (~> 0.8.1)
|
34
|
+
hashie (>= 1.2)
|
35
|
+
multi_json (~> 1.4)
|
36
|
+
nokogiri (~> 1.5.2)
|
37
|
+
oauth2
|
38
|
+
hashie (2.0.5)
|
39
|
+
highline (1.6.19)
|
40
|
+
httpauth (0.2.0)
|
41
|
+
json (1.8.0)
|
42
|
+
jwt (0.1.8)
|
43
|
+
multi_json (>= 1.5)
|
44
|
+
minitest (5.0.7)
|
45
|
+
multi_json (1.8.0)
|
46
|
+
multi_xml (0.5.5)
|
47
|
+
multipart-post (1.2.0)
|
48
|
+
nokogiri (1.5.10)
|
49
|
+
oauth2 (0.9.2)
|
50
|
+
faraday (~> 0.8)
|
51
|
+
httpauth (~> 0.2)
|
52
|
+
jwt (~> 0.1.4)
|
53
|
+
multi_json (~> 1.0)
|
54
|
+
multi_xml (~> 0.5)
|
55
|
+
rack (~> 1.2)
|
56
|
+
rack (1.5.2)
|
57
|
+
rake (10.1.0)
|
58
|
+
rdoc (3.12.2)
|
59
|
+
json (~> 1.4)
|
60
|
+
rspec (2.8.0)
|
61
|
+
rspec-core (~> 2.8.0)
|
62
|
+
rspec-expectations (~> 2.8.0)
|
63
|
+
rspec-mocks (~> 2.8.0)
|
64
|
+
rspec-core (2.8.0)
|
65
|
+
rspec-expectations (2.8.0)
|
66
|
+
diff-lcs (~> 1.1.2)
|
67
|
+
rspec-mocks (2.8.0)
|
68
|
+
|
69
|
+
PLATFORMS
|
70
|
+
ruby
|
71
|
+
|
72
|
+
DEPENDENCIES
|
73
|
+
bundler (>= 1.0.21)
|
74
|
+
cucumber
|
75
|
+
jeweler (~> 1.8.4)!
|
76
|
+
minitest (~> 5.0.7)
|
77
|
+
rdoc (~> 3.12)
|
78
|
+
rspec (~> 2.8.0)
|
data/LICENSE.txt
ADDED
@@ -0,0 +1,20 @@
|
|
1
|
+
Copyright (c) 2013 Pjotr Prins
|
2
|
+
|
3
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
4
|
+
a copy of this software and associated documentation files (the
|
5
|
+
"Software"), to deal in the Software without restriction, including
|
6
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
7
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
8
|
+
permit persons to whom the Software is furnished to do so, subject to
|
9
|
+
the following conditions:
|
10
|
+
|
11
|
+
The above copyright notice and this permission notice shall be
|
12
|
+
included in all copies or substantial portions of the Software.
|
13
|
+
|
14
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
15
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
16
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
17
|
+
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
18
|
+
LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
19
|
+
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
20
|
+
WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,120 @@
|
|
1
|
+
# bio-vcf
|
2
|
+
|
3
|
+
[![Build Status](https://secure.travis-ci.org/pjotrp/bioruby-vcf.png)](http://travis-ci.org/pjotrp/bioruby-vcf)
|
4
|
+
|
5
|
+
Yet another VCF parser. This one may give better performance and
|
6
|
+
useful command line filtering.
|
7
|
+
|
8
|
+
The VCF format is commonly used for variant calling between NGS
|
9
|
+
samples. The fast parser needs to carry some state, recorded for each
|
10
|
+
file in VcfHeader, which contains the VCF file header. Individual
|
11
|
+
lines (variant calls) first go through a raw parser returning an array
|
12
|
+
of fields. Further (lazy) parsing is handled through VcfRecord.
|
13
|
+
|
14
|
+
Health warning: Early days, your mileage may vary because I add
|
15
|
+
features as I go along! If something is not working, check out the
|
16
|
+
code. It is easy to add features.
|
17
|
+
|
18
|
+
## Installation
|
19
|
+
|
20
|
+
```sh
|
21
|
+
gem install bio-vcf
|
22
|
+
```
|
23
|
+
|
24
|
+
## Quick start
|
25
|
+
|
26
|
+
## Command line interface (CLI)
|
27
|
+
|
28
|
+
Get the version of the VCF file
|
29
|
+
|
30
|
+
```ruby
|
31
|
+
bio-vcf -q --eval-once header.version < file.vcf
|
32
|
+
4.1
|
33
|
+
```
|
34
|
+
|
35
|
+
Get the column headers
|
36
|
+
|
37
|
+
```ruby
|
38
|
+
bio-vcf -q -eval-once 'header.column_names.join(",")' < file.vcf
|
39
|
+
CHROM,POS,ID,REF,ALT,QUAL,FILTER,INFO,FORMAT,NORMAL,TUMOR
|
40
|
+
```
|
41
|
+
|
42
|
+
The 'fields' array contains unprocessed data (strings). Print first
|
43
|
+
five raw fields
|
44
|
+
|
45
|
+
```ruby
|
46
|
+
bio-vcf --eval 'fields[0..4].join("\t")' < file.vcf
|
47
|
+
```
|
48
|
+
|
49
|
+
Add a filter to display the fields on chromosome 12
|
50
|
+
|
51
|
+
```ruby
|
52
|
+
bio-vcf --filter 'fields[0]=="12"' --eval 'fields[0..4].join("\t")' < file.vcf
|
53
|
+
```
|
54
|
+
|
55
|
+
It gets better when we start using processed data, represented by an
|
56
|
+
object named 'rec'. Position is a value, so we can filter a range
|
57
|
+
|
58
|
+
```ruby
|
59
|
+
bio-vcf --filter 'rec.chrom=="12" and rec.pos>96_641_270 and rec.pos<96_641_276' < file.vcf
|
60
|
+
```
|
61
|
+
|
62
|
+
With subfields defined by rec.format
|
63
|
+
|
64
|
+
```ruby
|
65
|
+
bio-vcf --filter 'rec.tumor.ss != 2' < file.vcf
|
66
|
+
```
|
67
|
+
|
68
|
+
Output
|
69
|
+
|
70
|
+
```ruby
|
71
|
+
bio-vcf --filter 'rec.tumor.gq>30' --eval '[rec.ref,rec.alt,rec.tumor.bcount,rec.tumor.gq,rec.normal.gq].join("\t")' < file.vcf
|
72
|
+
```
|
73
|
+
|
74
|
+
Show the count of the bases that were scored as somatic
|
75
|
+
|
76
|
+
```ruby
|
77
|
+
bio-vcf --eval 'rec.alt+"\t"+rec.tumor.bcount.split(",")[["A","C","G","T"].index(rec.alt)]+"\t"+rec.tumor.gq.to_s' < file.vcf
|
78
|
+
```
|
79
|
+
|
80
|
+
Actually, we have a convenience implementation for bcount, so this is the same
|
81
|
+
|
82
|
+
```ruby
|
83
|
+
bio-vcf --eval 'rec.alt+"\t"+rec.tumor.bcount[rec.alt].to_s+"\t"+rec.tumor.gq.to_s' < file.vcf
|
84
|
+
```
|
85
|
+
|
86
|
+
Filter on the somatic results that were scored at least 4 times
|
87
|
+
|
88
|
+
```ruby
|
89
|
+
bio-vcf --filter 'rec.alt.size==1 and rec.tumor.bcount[rec.alt]>4' < test.vcf
|
90
|
+
```
|
91
|
+
|
92
|
+
Similar for base quality scores
|
93
|
+
|
94
|
+
```ruby
|
95
|
+
bio-vcf --filter 'rec.alt.size==1 and rec.tumor.amq[rec.alt]>30' < test.vcf
|
96
|
+
```
|
97
|
+
|
98
|
+
|
99
|
+
## Project home page
|
100
|
+
|
101
|
+
Information on the source tree, documentation, examples, issues and
|
102
|
+
how to contribute, see
|
103
|
+
|
104
|
+
http://github.com/pjotrp/bioruby-vcf
|
105
|
+
|
106
|
+
## Cite
|
107
|
+
|
108
|
+
If you use this software, please cite one of
|
109
|
+
|
110
|
+
* [BioRuby: bioinformatics software for the Ruby programming language](http://dx.doi.org/10.1093/bioinformatics/btq475)
|
111
|
+
* [Biogem: an effective tool-based approach for scaling up open source software development in bioinformatics](http://dx.doi.org/10.1093/bioinformatics/bts080)
|
112
|
+
|
113
|
+
## Biogems.info
|
114
|
+
|
115
|
+
This Biogem is published at (http://biogems.info/index.html#bio-vcf)
|
116
|
+
|
117
|
+
## Copyright
|
118
|
+
|
119
|
+
Copyright (c) 2014 Pjotr Prins. See LICENSE.txt for further details.
|
120
|
+
|
data/Rakefile
ADDED
@@ -0,0 +1,58 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
|
3
|
+
require 'rubygems'
|
4
|
+
require 'bundler'
|
5
|
+
begin
|
6
|
+
Bundler.setup(:default, :development)
|
7
|
+
rescue Bundler::BundlerError => e
|
8
|
+
$stderr.puts e.message
|
9
|
+
$stderr.puts "Run `bundle install` to install missing gems"
|
10
|
+
exit e.status_code
|
11
|
+
end
|
12
|
+
require 'rake'
|
13
|
+
|
14
|
+
require 'jeweler'
|
15
|
+
Jeweler::Tasks.new do |gem|
|
16
|
+
# gem is a Gem::Specification... see http://docs.rubygems.org/read/chapter/20 for more options
|
17
|
+
gem.name = "bio-vcf"
|
18
|
+
gem.homepage = "http://github.com/pjotrp/bioruby-vcf"
|
19
|
+
gem.license = "MIT"
|
20
|
+
gem.summary = %Q{VCF parser}
|
21
|
+
gem.description = %Q{Smart parser for VCF format}
|
22
|
+
gem.email = "pjotr.public01@thebird.nl"
|
23
|
+
gem.authors = ["Pjotr Prins"]
|
24
|
+
# dependencies defined in Gemfile
|
25
|
+
end
|
26
|
+
Jeweler::RubygemsDotOrgTasks.new
|
27
|
+
|
28
|
+
# require 'rspec/core'
|
29
|
+
# require 'rspec/core/rake_task'
|
30
|
+
# RSpec::Core::RakeTask.new(:spec) do |spec|
|
31
|
+
# spec.pattern = FileList['spec/**/*_spec.rb']
|
32
|
+
# end
|
33
|
+
|
34
|
+
# RSpec::Core::RakeTask.new(:rcov) do |spec|
|
35
|
+
# spec.pattern = 'spec/**/*_spec.rb'
|
36
|
+
# spec.rcov = true
|
37
|
+
# end
|
38
|
+
|
39
|
+
require 'rake/testtask'
|
40
|
+
|
41
|
+
Rake::TestTask.new do |t|
|
42
|
+
t.pattern = "spec/*_spec.rb"
|
43
|
+
end
|
44
|
+
|
45
|
+
require 'cucumber/rake/task'
|
46
|
+
Cucumber::Rake::Task.new(:features)
|
47
|
+
|
48
|
+
task :default => :spec
|
49
|
+
|
50
|
+
require 'rdoc/task'
|
51
|
+
Rake::RDocTask.new do |rdoc|
|
52
|
+
version = File.exist?('VERSION') ? File.read('VERSION') : ""
|
53
|
+
|
54
|
+
rdoc.rdoc_dir = 'rdoc'
|
55
|
+
rdoc.title = "bio-vcf #{version}"
|
56
|
+
rdoc.rdoc_files.include('README*')
|
57
|
+
rdoc.rdoc_files.include('lib/**/*.rb')
|
58
|
+
end
|
data/VERSION
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
0.0.1
|
data/bin/bio-vcf
ADDED
@@ -0,0 +1,121 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
#
|
3
|
+
# BioRuby vcf plugin
|
4
|
+
# Author:: Pjotr Prins
|
5
|
+
#
|
6
|
+
# Copyright (C) 2014 Pjotr Prins <pjotr.prins@thebird.nl>
|
7
|
+
|
8
|
+
USAGE = "Vcf parser"
|
9
|
+
|
10
|
+
gempath = File.dirname(File.dirname(__FILE__))
|
11
|
+
$: << File.join(gempath,'lib')
|
12
|
+
|
13
|
+
VERSION_FILENAME=File.join(gempath,'VERSION')
|
14
|
+
version = File.new(VERSION_FILENAME).read.chomp
|
15
|
+
|
16
|
+
require 'bio-vcf'
|
17
|
+
require 'optparse'
|
18
|
+
|
19
|
+
# Uncomment when using the bio-logger
|
20
|
+
# require 'bio-logger'
|
21
|
+
# log = Bio::Log::LoggerPlus.new 'vcf'
|
22
|
+
# log.outputters = Bio::Log::Outputter.stderr
|
23
|
+
# Bio::Log::CLI.logger('stderr')
|
24
|
+
# Bio::Log::CLI.trace('info')
|
25
|
+
|
26
|
+
options = { show_help: false}
|
27
|
+
opts = OptionParser.new do |o|
|
28
|
+
o.banner = "Usage: #{File.basename($0)} [options] filename\ne.g. #{File.basename($0)} --rdf < test/data/input/somaticsniper.vcf"
|
29
|
+
|
30
|
+
o.on_tail('--filter cmd',String, 'Evaluate filter on each record') do |cmd|
|
31
|
+
options[:filter] = cmd
|
32
|
+
end
|
33
|
+
o.on_tail('-e cmd', '--eval cmd',String, 'Evaluate command on each record') do |cmd|
|
34
|
+
options[:eval] = cmd
|
35
|
+
end
|
36
|
+
o.on_tail('--eval-once cmd',String, 'Evaluate command once') do |cmd|
|
37
|
+
options[:eval_once] = true
|
38
|
+
options[:eval] = cmd
|
39
|
+
end
|
40
|
+
o.on("--rdf", "Generate RDF") do |b|
|
41
|
+
options[:rdf] = true
|
42
|
+
end
|
43
|
+
|
44
|
+
# Uncomment the following when using the bio-logger
|
45
|
+
# o.separator ""
|
46
|
+
# o.on("--logger filename",String,"Log to file (default stderr)") do | name |
|
47
|
+
# Bio::Log::CLI.logger(name)
|
48
|
+
# end
|
49
|
+
#
|
50
|
+
# o.on("--trace options",String,"Set log level (default INFO, see bio-logger)") do | s |
|
51
|
+
# Bio::Log::CLI.trace(s)
|
52
|
+
# end
|
53
|
+
#
|
54
|
+
o.on("-q", "--quiet", "Run quietly") do |q|
|
55
|
+
# Bio::Log::CLI.trace('error')
|
56
|
+
options[:quiet] = true
|
57
|
+
end
|
58
|
+
#
|
59
|
+
# o.on("-v", "--verbose", "Run verbosely") do |v|
|
60
|
+
# Bio::Log::CLI.trace('info')
|
61
|
+
# end
|
62
|
+
#
|
63
|
+
# o.on("--debug", "Show debug messages") do |v|
|
64
|
+
# Bio::Log::CLI.trace('debug')
|
65
|
+
# end
|
66
|
+
|
67
|
+
o.separator ""
|
68
|
+
o.on_tail('-h', '--help', 'display this help and exit') do
|
69
|
+
options[:show_help] = true
|
70
|
+
end
|
71
|
+
end
|
72
|
+
|
73
|
+
include BioVcf
|
74
|
+
|
75
|
+
begin
|
76
|
+
opts.parse!(ARGV)
|
77
|
+
|
78
|
+
$stderr.print "vcf #{version} (biogem Ruby #{RUBY_VERSION}) by Pjotr Prins 2014\n" if !options[:quiet]
|
79
|
+
|
80
|
+
|
81
|
+
if options[:show_help]
|
82
|
+
print opts
|
83
|
+
print USAGE
|
84
|
+
exit 1
|
85
|
+
end
|
86
|
+
|
87
|
+
$stderr.print "Options: ",options,"\n" if !options[:quiet]
|
88
|
+
|
89
|
+
header = VcfHeader.new
|
90
|
+
|
91
|
+
STDIN.each_line do | line |
|
92
|
+
if line =~ /^##fileformat=/
|
93
|
+
# ---- We have a new file header
|
94
|
+
header = VcfHeader.new
|
95
|
+
header.add(line)
|
96
|
+
STDIN.each_line do | headerline |
|
97
|
+
if headerline !~ /^#/
|
98
|
+
line = headerline
|
99
|
+
break # end of header
|
100
|
+
end
|
101
|
+
header.add(headerline)
|
102
|
+
end
|
103
|
+
end
|
104
|
+
# ---- Parse VCF record line
|
105
|
+
# fields = VcfLine.parse(line,header.columns)
|
106
|
+
fields = VcfLine.parse(line)
|
107
|
+
rec = VcfRecord.new(fields,header)
|
108
|
+
if !options[:filter] or (options[:filter] and eval(options[:filter]))
|
109
|
+
if options[:eval]
|
110
|
+
print eval(options[:eval])
|
111
|
+
exit(1) if options[:eval_once]
|
112
|
+
else
|
113
|
+
# Default behaviour
|
114
|
+
print fields.join("\t")
|
115
|
+
end
|
116
|
+
print "\n"
|
117
|
+
end
|
118
|
+
end
|
119
|
+
|
120
|
+
end
|
121
|
+
|
File without changes
|
@@ -0,0 +1,13 @@
|
|
1
|
+
require 'bundler'
|
2
|
+
begin
|
3
|
+
Bundler.setup(:default, :development)
|
4
|
+
rescue Bundler::BundlerError => e
|
5
|
+
$stderr.puts e.message
|
6
|
+
$stderr.puts "Run `bundle install` to install missing gems"
|
7
|
+
exit e.status_code
|
8
|
+
end
|
9
|
+
|
10
|
+
$LOAD_PATH.unshift(File.dirname(__FILE__) + '/../../lib')
|
11
|
+
require 'bio-vcf'
|
12
|
+
|
13
|
+
require 'rspec/expectations'
|
data/lib/bio-vcf.rb
ADDED
@@ -0,0 +1,15 @@
|
|
1
|
+
# Please require your code below, respecting the naming conventions in the
|
2
|
+
# bioruby directory tree.
|
3
|
+
#
|
4
|
+
# For example, say you have a plugin named bio-plugin, the only uncommented
|
5
|
+
# line in this file would be
|
6
|
+
#
|
7
|
+
# require 'bio/bio-plugin/plugin'
|
8
|
+
#
|
9
|
+
# In this file only require other files. Avoid other source code.
|
10
|
+
|
11
|
+
require 'bio-vcf/vcf'
|
12
|
+
require 'bio-vcf/vcfheader'
|
13
|
+
require 'bio-vcf/vcfline'
|
14
|
+
require 'bio-vcf/vcfgenotypefield'
|
15
|
+
require 'bio-vcf/vcfrecord'
|
data/lib/bio-vcf/vcf.rb
ADDED
@@ -0,0 +1,43 @@
|
|
1
|
+
module BioVcf
|
2
|
+
|
3
|
+
class VcfNucleotides
|
4
|
+
def initialize list
|
5
|
+
@list = list
|
6
|
+
end
|
7
|
+
|
8
|
+
def [] idx
|
9
|
+
idx = ["A","C","G","T"].index(idx) if idx.kind_of?(String)
|
10
|
+
return 0 if idx == nil
|
11
|
+
@list[idx].to_i
|
12
|
+
end
|
13
|
+
end
|
14
|
+
|
15
|
+
class VcfGenotypeField
|
16
|
+
def initialize s, format, header
|
17
|
+
@values = s.split(/:/)
|
18
|
+
@format = format
|
19
|
+
@header = header
|
20
|
+
end
|
21
|
+
|
22
|
+
def bcount
|
23
|
+
VcfNucleotides.new(@values[@format['BCOUNT']].split(','))
|
24
|
+
end
|
25
|
+
|
26
|
+
def bq
|
27
|
+
VcfNucleotides.new(@values[@format['BQ']].split(','))
|
28
|
+
end
|
29
|
+
|
30
|
+
def amq
|
31
|
+
VcfNucleotides.new(@values[@format['AMQ']].split(','))
|
32
|
+
end
|
33
|
+
|
34
|
+
|
35
|
+
def method_missing(m, *args, &block)
|
36
|
+
v = @values[@format[m.to_s.upcase]]
|
37
|
+
v = v.to_i if v =~ /^\d+$/
|
38
|
+
v = v.to_f if v =~ /^\d+\.\d+$/
|
39
|
+
v
|
40
|
+
end
|
41
|
+
|
42
|
+
end
|
43
|
+
end
|
@@ -0,0 +1,42 @@
|
|
1
|
+
|
2
|
+
module BioVcf
|
3
|
+
|
4
|
+
module VcfHeaderParser
|
5
|
+
def VcfHeaderParser.get_column_names(lines)
|
6
|
+
lines.each do | line |
|
7
|
+
if line =~ /^#[^#]/
|
8
|
+
names = line.split
|
9
|
+
names[0].sub!(/^#/,'')
|
10
|
+
return names
|
11
|
+
end
|
12
|
+
end
|
13
|
+
nil
|
14
|
+
end
|
15
|
+
end
|
16
|
+
|
17
|
+
class VcfHeader
|
18
|
+
|
19
|
+
attr_reader :lines
|
20
|
+
|
21
|
+
def initialize
|
22
|
+
@lines = []
|
23
|
+
end
|
24
|
+
|
25
|
+
def add line
|
26
|
+
@lines << line.strip
|
27
|
+
end
|
28
|
+
|
29
|
+
def version
|
30
|
+
@version ||= lines[0].scan(/##fileformat=VCFv(\d+\.\d+)/)[0][0]
|
31
|
+
end
|
32
|
+
|
33
|
+
def column_names
|
34
|
+
@column_names ||= VcfHeaderParser::get_column_names(@lines)
|
35
|
+
end
|
36
|
+
|
37
|
+
def columns
|
38
|
+
@column ||= column_names.size
|
39
|
+
end
|
40
|
+
end
|
41
|
+
|
42
|
+
end
|
@@ -0,0 +1,11 @@
|
|
1
|
+
module BioVcf
|
2
|
+
module VcfLine
|
3
|
+
|
4
|
+
# Split a line into fields and check size
|
5
|
+
def VcfLine.parse line,expected_size=nil
|
6
|
+
fields = line.strip.split(/\t/)
|
7
|
+
raise "Expected #{expected_size} fields but got #{fields.size} in "+fields.to_s if expected_size and fields.size != expected_size
|
8
|
+
fields
|
9
|
+
end
|
10
|
+
end
|
11
|
+
end
|
@@ -0,0 +1,55 @@
|
|
1
|
+
module BioVcf
|
2
|
+
|
3
|
+
module VcfRecordParser
|
4
|
+
def VcfRecordParser.get_format s
|
5
|
+
h = {}
|
6
|
+
s.split(/:/).each_with_index { |v,i| h[v] = i }
|
7
|
+
h
|
8
|
+
end
|
9
|
+
end
|
10
|
+
|
11
|
+
class VcfRecord
|
12
|
+
|
13
|
+
def initialize fields, header
|
14
|
+
@header = header
|
15
|
+
@fields = fields
|
16
|
+
end
|
17
|
+
|
18
|
+
def chrom
|
19
|
+
@fields[0]
|
20
|
+
end
|
21
|
+
|
22
|
+
def pos
|
23
|
+
@pos ||= @fields[1].to_i
|
24
|
+
end
|
25
|
+
|
26
|
+
def ids
|
27
|
+
@ids ||= @fields[2].split(';')
|
28
|
+
end
|
29
|
+
|
30
|
+
def id
|
31
|
+
ids[0]
|
32
|
+
end
|
33
|
+
|
34
|
+
def ref
|
35
|
+
@refs ||= @fields[3]
|
36
|
+
end
|
37
|
+
|
38
|
+
def alt
|
39
|
+
@alt ||= @fields[4]
|
40
|
+
end
|
41
|
+
|
42
|
+
def format
|
43
|
+
@format ||= VcfRecordParser.get_format(@fields[8])
|
44
|
+
end
|
45
|
+
|
46
|
+
def normal
|
47
|
+
@normal ||= VcfGenotypeField.new(@fields[9],format,@header)
|
48
|
+
end
|
49
|
+
|
50
|
+
def tumor
|
51
|
+
@tumor ||= VcfGenotypeField.new(@fields[10],format,@header)
|
52
|
+
end
|
53
|
+
|
54
|
+
end
|
55
|
+
end
|
@@ -0,0 +1,87 @@
|
|
1
|
+
##fileformat=VCFv4.1
|
2
|
+
##fileDate=20140121
|
3
|
+
##phasing=none
|
4
|
+
##reference=file:///data/GENOMES/human_GATK_GRCh37/GRCh37_gatk.fasta
|
5
|
+
##FORMAT=<ID=GT,Number=1,Type=String,Description="Genotype">
|
6
|
+
##FORMAT=<ID=IGT,Number=1,Type=String,Description="Genotype when called independently (only filled if called in joint prior mode)">
|
7
|
+
##FORMAT=<ID=DP,Number=1,Type=Integer,Description="Total read depth">
|
8
|
+
##FORMAT=<ID=DP4,Number=4,Type=Integer,Description="# high-quality ref-forward bases, ref-reverse, alt-forward and alt-reverse bases">
|
9
|
+
##FORMAT=<ID=BCOUNT,Number=4,Type=Integer,Description="Occurrence count for each base at this site (A,C,G,T)">
|
10
|
+
##FORMAT=<ID=GQ,Number=1,Type=Integer,Description="Genotype quality">
|
11
|
+
##FORMAT=<ID=JGQ,Number=1,Type=Integer,Description="Joint genotype quality (only filled if called in join prior mode)">
|
12
|
+
##FORMAT=<ID=VAQ,Number=1,Type=Integer,Description="Variant allele quality">
|
13
|
+
##FORMAT=<ID=BQ,Number=.,Type=Integer,Description="Average base quality">
|
14
|
+
##FORMAT=<ID=MQ,Number=1,Type=Integer,Description="Average mapping quality across all reads">
|
15
|
+
##FORMAT=<ID=AMQ,Number=.,Type=Integer,Description="Average mapping quality for each allele present in the genotype">
|
16
|
+
##FORMAT=<ID=SS,Number=1,Type=Integer,Description="Variant status relative to non-adjacent Normal, 0=wildtype,1=germline,2=somatic,3=LOH,4=unknown">
|
17
|
+
##FORMAT=<ID=SSC,Number=1,Type=Integer,Description="Somatic Score">
|
18
|
+
#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT NORMAL TUMOR
|
19
|
+
1 1636394 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:17:12,5,0,0:0,17,0,0:75:36:0:56:37:37:0:. 0/1:0/1:6:3,1,1,1:0,4,0,2:29:36:29:60,60:37:37,37:2:36
|
20
|
+
1 36217006 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:21:9,12,0,0:0,21,0,0:90:31:0:60:37:37:0:. 0/1:0/1:6:2,2,0,2:0,4,0,2:24:31:24:60,53:37:37,37:2:31
|
21
|
+
1 46527674 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:6,3,0,0:0,0,9,0:54:37:0:60:37:37:0:. 0/1:0/1:4:2,0,0,2:2,0,2,0:30:37:30:60,60:37:37,37:2:34
|
22
|
+
1 108417572 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:5,2,0,0:0,7,0,0:48:42:0:41:37:37:0:. 0/1:0/1:4:2,0,1,1:0,2,0,2:30:42:35:60,60:37:37,37:2:31
|
23
|
+
1 155170305 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:5,4,0,0:0,9,0,0:54:34:0:59:37:37:0:. 0/1:0/1:6:3,1,2,0:0,4,0,2:27:34:27:44,60:37:37,37:2:32
|
24
|
+
1 155449089 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:5,4,0,0:0,0,9,0:54:37:0:60:37:37:0:. 0/1:0/1:4:1,1,0,2:2,0,2,0:30:37:30:60,60:37:37,37:2:34
|
25
|
+
1 169847826 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:6,3,0,0:0,0,9,0:54:37:0:60:37:37:0:. 0/1:0/1:4:2,0,2,0:2,0,2,0:30:37:30:60,60:37:37,37:2:34
|
26
|
+
1 203098164 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:3,7,0,0:0,10,0,0:57:40:0:53:37:37:0:. 0/1:0/1:3:1,0,2,0:0,1,0,2:1:40:33:60,60:37:37,37:2:37
|
27
|
+
2 39213209 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:16:8,8,0,0:0,0,16,0:75:42:0:56:37:37:0:. 0/1:0/1:4:2,0,1,1:2,0,2,0:30:42:35:60,59:37:37,37:2:42
|
28
|
+
2 86691250 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:6,2,0,0:0,0,8,0:51:42:0:59:37:37:0:. 0/1:0/1:4:2,0,1,1:2,0,2,0:30:42:35:60,60:37:37,37:2:33
|
29
|
+
2 88874243 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:17:12,5,0,0:0,0,17,0:78:34:0:59:37:37:0:. 0/1:0/1:5:2,1,0,2:2,0,3,0:27:34:27:60,60:37:37,37:2:34
|
30
|
+
2 121728044 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:6,3,0,0:0,0,9,0:54:42:0:55:37:37:0:. 0/1:0/1:4:1,1,1,1:2,0,2,0:35:42:35:60,57:37:37,37:2:36
|
31
|
+
2 170062591 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:4,5,0,0:0,0,9,0:54:39:0:60:37:37:0:. 0/1:0/1:5:2,1,1,1:2,0,3,0:32:39:32:60,60:37:37,37:2:35
|
32
|
+
2 216257844 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:2,5,0,0:0,0,7,0:48:40:0:60:37:37:0:. 0/1:0/1:3:1,0,0,2:2,0,1,0:1:40:33:60,60:37:37,37:2:30
|
33
|
+
2 222322623 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:15:6,9,0,0:0,15,0,0:72:33:0:60:37:37:0:. 0/1:0/1:7:4,1,1,1:0,5,0,2:26:33:26:60,60:37:37,37:2:33
|
34
|
+
3 25675413 . A T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:6,2,0,0:8,0,0,0:51:32:0:59:37:37:0:. 0/1:0/1:4:1,1,0,2:2,0,0,2:6:32:25:34,37:37:37,37:2:30
|
35
|
+
3 36779638 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:13:8,5,0,0:0,13,0,0:66:36:0:60:37:37:0:. 0/1:0/1:6:1,3,1,1:0,4,0,2:29:36:29:60,60:37:37,37:2:36
|
36
|
+
3 123458847 . T C . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:4,6,0,0:0,0,0,10:54:39:0:41:37:37:0:. 0/1:0/1:5:1,2,1,1:0,2,0,3:32:39:32:55,45:37:37,37:2:35
|
37
|
+
3 124351308 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:7,1,0,0:0,0,8,0:51:42:0:58:37:37:0:. 0/1:0/1:4:1,1,1,1:2,0,2,0:35:42:35:60,60:37:37,37:2:33
|
38
|
+
3 142171996 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:8,2,0,0:0,10,0,0:57:37:0:59:37:37:0:. 0/1:0/1:4:0,2,2,0:0,2,0,2:30:37:30:60,60:37:37,37:2:35
|
39
|
+
3 189526168 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:23:11,12,0,0:0,0,23,0:96:36:0:59:37:37:0:. 0/1:0/1:6:1,3,1,1:2,0,4,0:29:36:29:60,60:37:37,37:2:36
|
40
|
+
4 82058553 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:6,1,0,0:0,7,0,0:48:37:0:60:37:37:0:. 0/1:0/1:4:2,0,2,0:0,2,0,2:30:37:30:60,60:37:37,37:2:30
|
41
|
+
4 122769998 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:7,0,0,0:0,7,0,0:48:40:0:54:37:37:0:. 0/1:0/1:3:1,0,2,0:0,1,0,2:1:40:33:60,60:37:37,37:2:30
|
42
|
+
5 13850856 . G C . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:3,5,0,0:0,0,8,0:51:40:0:51:37:37:0:. 0/1:0/1:3:0,1,2,0:0,2,1,0:1:40:33:49,60:37:37,37:2:33
|
43
|
+
5 132038609 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:12:10,2,0,0:0,12,0,0:63:33:0:58:37:37:0:. 0/1:0/1:7:3,2,1,1:0,5,0,2:26:33:26:60,60:37:37,37:2:33
|
44
|
+
5 137756599 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:13:7,6,0,0:0,13,0,0:66:31:0:60:37:37:0:. 0/1:0/1:6:2,2,2,0:0,4,0,2:24:31:24:60,60:37:37,37:2:31
|
45
|
+
5 141974902 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:14:9,5,0,0:0,14,0,0:69:40:0:60:37:37:0:. 0/1:0/1:3:1,0,2,0:0,1,0,2:1:40:33:60,60:37:37,37:2:40
|
46
|
+
6 2749400 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:8,2,0,0:0,10,0,0:57:35:0:60:37:37:0:. 0/1:0/1:4:2,0,0,2:0,2,0,2:28:35:28:60,39:37:37,37:2:34
|
47
|
+
7 95217113 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:5,2,0,0:0,0,7,0:48:36:0:60:37:37:0:. 0/1:0/1:6:1,3,1,1:2,0,4,0:29:36:29:60,60:37:37,37:2:30
|
48
|
+
7 140434525 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:14:8,6,0,0:0,0,14,0:69:31:0:59:37:37:0:. 0/1:0/1:6:3,1,0,2:2,0,4,0:24:31:24:60,58:37:37,37:2:31
|
49
|
+
7 151856059 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:20:9,10,1,0:0,19,0,1:50:33:0:59:37:37:0:. 0/1:0/1:7:1,4,1,1:0,5,0,2:26:33:26:60,60:37:37,37:2:30
|
50
|
+
8 42958817 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:12:6,6,0,0:0,12,0,0:60:36:0:55:37:37:0:. 0/1:0/1:7:4,1,1,1:0,5,0,2:29:36:29:48,60:37:37,37:2:35
|
51
|
+
8 131070237 . A G . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:13:9,4,0,0:13,0,0,0:66:39:0:60:37:37:0:. 0/1:0/1:5:2,1,1,1:3,0,2,0:32:39:32:60,60:37:37,37:2:39
|
52
|
+
8 141711010 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:8,2,0,0:0,10,0,0:57:40:0:59:37:37:0:. 0/1:0/1:3:1,0,2,0:0,1,0,2:1:40:33:60,55:37:37,37:2:37
|
53
|
+
8 145059674 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:14:7,7,0,0:0,14,0,0:69:34:0:56:37:37:0:. 0/1:0/1:5:2,1,0,2:0,3,0,2:27:34:27:60,60:37:37,37:2:34
|
54
|
+
9 111651620 . A T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:4,3,0,0:7,0,0,0:48:92:0:60:37:37:0:. 0/1:0/1:7:1,2,3,1:3,0,0,4:58:92:85:60,60:37:37,37:2:31
|
55
|
+
9 111685156 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:13:9,4,0,0:0,0,13,0:66:37:0:60:37:37:0:. 0/1:0/1:4:1,1,2,0:2,0,2,0:12:37:30:60,37:37:37,37:2:37
|
56
|
+
10 6525571 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:0,8,0,0:0,0,8,0:51:37:0:55:37:37:0:. 0/1:0/1:4:1,1,0,2:2,0,2,0:30:37:30:60,60:37:37,37:2:32
|
57
|
+
10 97197246 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:7,3,0,0:0,10,0,0:57:39:0:58:37:37:0:. 0/1:0/1:5:3,0,1,1:0,3,0,2:32:39:32:60,53:37:37,37:2:36
|
58
|
+
11 58949455 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:12:8,4,0,0:0,0,12,0:60:33:0:54:37:37:0:. 0/1:0/1:7:3,2,1,1:2,0,5,0:26:33:26:60,36:37:37,37:2:33
|
59
|
+
11 65481082 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:17:12,5,0,0:0,0,17,0:78:45:0:58:37:37:0:. 0/1:0/1:3:0,1,1,1:2,0,1,0:1:45:34:60,60:37:37,37:2:45
|
60
|
+
11 94180424 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:23:17,6,0,0:0,0,23,0:96:34:0:56:37:37:0:. 0/1:0/1:5:2,1,2,0:2,0,3,0:27:34:27:60,59:37:37,37:2:34
|
61
|
+
11 121036021 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:12:3,9,0,0:0,12,0,0:63:31:0:59:37:37:0:. 0/1:0/1:6:4,0,0,2:0,4,0,2:24:31:24:59,60:37:37,37:2:31
|
62
|
+
12 994952 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:1,8,0,0:0,9,0,0:54:33:0:59:37:37:0:. 0/1:0/1:7:2,3,1,1:0,5,0,2:26:33:26:51,55:37:37,37:2:32
|
63
|
+
12 69233187 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:5,4,0,0:0,9,0,0:51:33:0:53:37:37:0:. 0/1:0/1:7:4,1,1,1:0,5,0,2:26:33:26:60,60:37:37,37:2:30
|
64
|
+
12 77436879 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:4,6,0,0:0,0,10,0:57:34:0:60:37:37:0:. 0/1:0/1:5:2,1,0,2:2,0,3,0:27:34:27:60,60:37:37,37:2:33
|
65
|
+
12 96641273 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:4,4,0,0:0,0,8,0:51:33:0:60:37:37:0:. 0/1:0/1:7:2,3,1,1:2,0,5,0:26:33:26:60,50:37:37,37:2:30
|
66
|
+
12 110813986 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:4,4,0,0:0,0,8,0:51:39:0:59:37:37:0:. 0/1:0/1:5:1,2,1,1:2,0,3,0:32:39:32:60,60:37:37,37:2:33
|
67
|
+
12 122825587 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:21:12,9,0,0:0,21,0,0:90:34:0:59:37:37:0:. 0/1:0/1:5:3,0,2,0:0,3,0,2:27:34:27:60,60:37:37,37:2:34
|
68
|
+
14 30135337 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:5,2,0,0:0,7,0,0:48:40:0:60:37:37:0:. 0/1:0/1:3:1,0,0,2:0,1,0,2:1:40:33:60,60:37:37,37:2:30
|
69
|
+
14 51398458 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:4,5,0,0:0,0,9,0:54:36:0:55:37:37:0:. 0/1:0/1:6:0,4,1,1:2,0,4,0:29:36:29:60,59:37:37,37:2:33
|
70
|
+
15 43170722 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:11:6,5,0,0:0,11,0,0:60:31:0:60:37:37:0:. 0/1:0/1:6:4,0,2,0:0,4,0,2:24:31:24:56,54:37:37,37:2:31
|
71
|
+
15 50862183 . C A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:7,2,0,0:0,9,0,0:54:35:0:60:37:37:0:. 0/1:0/1:4:0,2,0,2:2,2,0,0:28:35:28:45,59:37:37,37:2:33
|
72
|
+
15 64332347 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:6,1,0,0:0,7,0,0:48:39:0:60:37:37:0:. 0/1:0/1:5:3,0,1,1:0,3,0,2:32:39:32:58,56:37:37,37:2:30
|
73
|
+
15 80845030 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:1,8,0,0:0,9,0,0:54:31:0:60:37:37:0:. 0/1:0/1:6:2,2,2,0:0,4,0,2:24:31:24:60,60:37:37,37:2:30
|
74
|
+
16 1812938 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:11:8,3,0,0:0,0,11,0:60:34:0:46:37:37:0:. 0/1:0/1:5:1,2,1,1:2,0,3,0:27:34:27:46,55:37:37,37:2:33
|
75
|
+
16 3582808 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:10:6,4,0,0:0,0,10,0:57:34:0:59:37:37:0:. 0/1:0/1:5:0,3,2,0:2,0,3,0:27:34:27:60,60:37:37,37:2:33
|
76
|
+
16 14042032 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:6,1,0,0:0,7,0,0:48:37:0:60:37:37:0:. 0/1:0/1:4:0,2,2,0:0,2,0,2:30:37:30:57,60:37:37,37:2:30
|
77
|
+
16 23619204 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:5,2,0,0:0,0,7,0:48:42:0:60:37:37:0:. 0/1:0/1:4:1,1,1,1:2,0,2,0:35:42:35:60,60:37:37,37:2:31
|
78
|
+
17 41256142 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:5,3,0,0:0,0,8,0:51:40:0:60:37:37:0:. 0/1:0/1:3:0,1,2,0:2,0,1,0:1:40:33:57,60:37:37,37:2:33
|
79
|
+
17 61784013 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:9:6,3,0,0:0,9,0,0:54:45:0:60:37:37:0:. 0/1:0/1:3:0,1,1,1:0,1,0,2:1:45:34:60,60:37:37,37:2:36
|
80
|
+
18 45423074 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:16:8,8,0,0:0,16,0,0:75:46:0:53:37:37:0:. 0/1:0/1:10:3,4,3,0:0,7,0,3:39:46:39:60,60:37:37,37:2:46
|
81
|
+
18 60985432 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:14:9,4,0,1:0,0,13,0:32:42:0:60:37:37:0:. 0/1:0/1:4:2,0,1,1:2,0,2,0:30:42:35:60,57:37:37,37:2:41
|
82
|
+
19 39664512 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:14:8,6,0,0:0,14,0,0:69:42:0:60:37:37:0:. 0/1:0/1:4:0,2,1,1:0,2,0,2:30:42:35:60,60:37:37,37:2:42
|
83
|
+
19 49473085 . G C . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:4,4,0,0:0,0,8,0:51:37:0:60:37:37:0:. 0/1:0/1:4:1,1,2,0:0,2,2,0:30:37:30:48,56:37:37,37:2:32
|
84
|
+
20 34135210 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:8:2,6,0,0:0,8,0,0:51:36:0:59:37:37:0:. 0/1:0/1:6:2,2,1,1:0,4,0,2:29:36:29:60,60:37:37,37:2:32
|
85
|
+
20 35663882 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:2,5,0,0:0,0,7,0:48:37:0:56:37:37:0:. 0/1:0/1:4:1,1,0,2:2,0,2,0:30:37:30:60,60:37:37,37:2:30
|
86
|
+
X 70341572 . C T . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:11:6,5,0,0:0,11,0,0:60:33:0:56:37:37:0:. 0/1:0/1:7:3,2,1,1:0,5,0,2:26:33:26:58,60:37:37,37:2:33
|
87
|
+
X 123164862 . G A . . . GT:IGT:DP:DP4:BCOUNT:GQ:JGQ:VAQ:BQ:MQ:AMQ:SS:SSC 0/0:0/0:7:5,2,0,0:0,0,7,0:48:40:0:59:37:37:0:. 0/1:0/1:3:1,0,2,0:2,0,1,0:1:40:33:60,60:37:37,37:2:30
|
metadata
ADDED
@@ -0,0 +1,163 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: bio-vcf
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Pjotr Prins
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2014-01-23 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: minitest
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ~>
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: 5.0.7
|
22
|
+
type: :development
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ~>
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: 5.0.7
|
30
|
+
- !ruby/object:Gem::Dependency
|
31
|
+
name: rspec
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
33
|
+
none: false
|
34
|
+
requirements:
|
35
|
+
- - ~>
|
36
|
+
- !ruby/object:Gem::Version
|
37
|
+
version: 2.8.0
|
38
|
+
type: :development
|
39
|
+
prerelease: false
|
40
|
+
version_requirements: !ruby/object:Gem::Requirement
|
41
|
+
none: false
|
42
|
+
requirements:
|
43
|
+
- - ~>
|
44
|
+
- !ruby/object:Gem::Version
|
45
|
+
version: 2.8.0
|
46
|
+
- !ruby/object:Gem::Dependency
|
47
|
+
name: cucumber
|
48
|
+
requirement: !ruby/object:Gem::Requirement
|
49
|
+
none: false
|
50
|
+
requirements:
|
51
|
+
- - ! '>='
|
52
|
+
- !ruby/object:Gem::Version
|
53
|
+
version: '0'
|
54
|
+
type: :development
|
55
|
+
prerelease: false
|
56
|
+
version_requirements: !ruby/object:Gem::Requirement
|
57
|
+
none: false
|
58
|
+
requirements:
|
59
|
+
- - ! '>='
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
62
|
+
- !ruby/object:Gem::Dependency
|
63
|
+
name: jeweler
|
64
|
+
requirement: !ruby/object:Gem::Requirement
|
65
|
+
none: false
|
66
|
+
requirements:
|
67
|
+
- - ~>
|
68
|
+
- !ruby/object:Gem::Version
|
69
|
+
version: 1.8.4
|
70
|
+
type: :development
|
71
|
+
prerelease: false
|
72
|
+
version_requirements: !ruby/object:Gem::Requirement
|
73
|
+
none: false
|
74
|
+
requirements:
|
75
|
+
- - ~>
|
76
|
+
- !ruby/object:Gem::Version
|
77
|
+
version: 1.8.4
|
78
|
+
- !ruby/object:Gem::Dependency
|
79
|
+
name: bundler
|
80
|
+
requirement: !ruby/object:Gem::Requirement
|
81
|
+
none: false
|
82
|
+
requirements:
|
83
|
+
- - ! '>='
|
84
|
+
- !ruby/object:Gem::Version
|
85
|
+
version: 1.0.21
|
86
|
+
type: :development
|
87
|
+
prerelease: false
|
88
|
+
version_requirements: !ruby/object:Gem::Requirement
|
89
|
+
none: false
|
90
|
+
requirements:
|
91
|
+
- - ! '>='
|
92
|
+
- !ruby/object:Gem::Version
|
93
|
+
version: 1.0.21
|
94
|
+
- !ruby/object:Gem::Dependency
|
95
|
+
name: rdoc
|
96
|
+
requirement: !ruby/object:Gem::Requirement
|
97
|
+
none: false
|
98
|
+
requirements:
|
99
|
+
- - ~>
|
100
|
+
- !ruby/object:Gem::Version
|
101
|
+
version: '3.12'
|
102
|
+
type: :development
|
103
|
+
prerelease: false
|
104
|
+
version_requirements: !ruby/object:Gem::Requirement
|
105
|
+
none: false
|
106
|
+
requirements:
|
107
|
+
- - ~>
|
108
|
+
- !ruby/object:Gem::Version
|
109
|
+
version: '3.12'
|
110
|
+
description: Smart parser for VCF format
|
111
|
+
email: pjotr.public01@thebird.nl
|
112
|
+
executables:
|
113
|
+
- bio-vcf
|
114
|
+
extensions: []
|
115
|
+
extra_rdoc_files:
|
116
|
+
- LICENSE.txt
|
117
|
+
- README.md
|
118
|
+
files:
|
119
|
+
- Gemfile
|
120
|
+
- Gemfile.lock
|
121
|
+
- LICENSE.txt
|
122
|
+
- README.md
|
123
|
+
- Rakefile
|
124
|
+
- VERSION
|
125
|
+
- bin/bio-vcf
|
126
|
+
- features/step_definitions/bio-vcf_steps.rb
|
127
|
+
- features/support/env.rb
|
128
|
+
- lib/bio-vcf.rb
|
129
|
+
- lib/bio-vcf/vcf.rb
|
130
|
+
- lib/bio-vcf/vcfgenotypefield.rb
|
131
|
+
- lib/bio-vcf/vcfheader.rb
|
132
|
+
- lib/bio-vcf/vcfline.rb
|
133
|
+
- lib/bio-vcf/vcfrecord.rb
|
134
|
+
- test/data/input/somaticsniper.vcf
|
135
|
+
homepage: http://github.com/pjotrp/bioruby-vcf
|
136
|
+
licenses:
|
137
|
+
- MIT
|
138
|
+
post_install_message:
|
139
|
+
rdoc_options: []
|
140
|
+
require_paths:
|
141
|
+
- lib
|
142
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
143
|
+
none: false
|
144
|
+
requirements:
|
145
|
+
- - ! '>='
|
146
|
+
- !ruby/object:Gem::Version
|
147
|
+
version: '0'
|
148
|
+
segments:
|
149
|
+
- 0
|
150
|
+
hash: 751802321492035161
|
151
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
152
|
+
none: false
|
153
|
+
requirements:
|
154
|
+
- - ! '>='
|
155
|
+
- !ruby/object:Gem::Version
|
156
|
+
version: '0'
|
157
|
+
requirements: []
|
158
|
+
rubyforge_project:
|
159
|
+
rubygems_version: 1.8.23
|
160
|
+
signing_key:
|
161
|
+
specification_version: 3
|
162
|
+
summary: VCF parser
|
163
|
+
test_files: []
|