bio-tabix 0.1.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/.document +5 -0
- data/.rvmrc +41 -0
- data/Gemfile +13 -0
- data/LICENSE.txt +20 -0
- data/README.rdoc +82 -0
- data/Rakefile +59 -0
- data/VERSION +1 -0
- data/bio-tabix.gemspec +72 -0
- data/ext/tabix/Rakefile +2 -0
- data/ext/tabix/mkrf_conf.rb +73 -0
- data/lib/bio-tabix.rb +2 -0
- data/lib/bio/tabix/Version +1 -0
- data/lib/bio/tabix/binding.rb +100 -0
- data/lib/bio/tabix/library.rb +38 -0
- data/lib/bio/tabix/t_file.rb +158 -0
- data/test/helper.rb +18 -0
- data/test/test_bio-tabix.rb +7 -0
- metadata +164 -0
data/.document
ADDED
data/.rvmrc
ADDED
@@ -0,0 +1,41 @@
|
|
1
|
+
#!/usr/bin/env bash
|
2
|
+
|
3
|
+
# This is an RVM Project .rvmrc file, used to automatically load the ruby
|
4
|
+
# development environment upon cd'ing into the directory
|
5
|
+
|
6
|
+
# First we specify our desired <ruby>[@<gemset>], the @gemset name is optional.
|
7
|
+
environment_id="ruby-1.9.2-p290@tabix"
|
8
|
+
|
9
|
+
#
|
10
|
+
# First we attempt to load the desired environment directly from the environment
|
11
|
+
# file. This is very fast and efficient compared to running through the entire
|
12
|
+
# CLI and selector. If you want feedback on which environment was used then
|
13
|
+
# insert the word 'use' after --create as this triggers verbose mode.
|
14
|
+
#
|
15
|
+
if [[ -d "${rvm_path:-$HOME/.rvm}/environments" \
|
16
|
+
&& -s "${rvm_path:-$HOME/.rvm}/environments/$environment_id" ]]
|
17
|
+
then
|
18
|
+
\. "${rvm_path:-$HOME/.rvm}/environments/$environment_id"
|
19
|
+
|
20
|
+
if [[ -s "${rvm_path:-$HOME/.rvm}/hooks/after_use" ]]
|
21
|
+
then
|
22
|
+
. "${rvm_path:-$HOME/.rvm}/hooks/after_use"
|
23
|
+
fi
|
24
|
+
else
|
25
|
+
# If the environment file has not yet been created, use the RVM CLI to select.
|
26
|
+
if ! rvm --create "$environment_id"
|
27
|
+
then
|
28
|
+
echo "Failed to create RVM environment ''."
|
29
|
+
fi
|
30
|
+
fi
|
31
|
+
|
32
|
+
#
|
33
|
+
# If you use an RVM gemset file to install a list of gems (*.gems), you can have
|
34
|
+
# it be automatically loaded. Uncomment the following and adjust the filename if
|
35
|
+
# necessary.
|
36
|
+
#
|
37
|
+
# filename=".gems"
|
38
|
+
# if [[ -s "$filename" ]] ; then
|
39
|
+
# rvm gemset import "$filename" | grep -v already | grep -v listed | grep -v complete | sed '/^$/d'
|
40
|
+
# fi
|
41
|
+
|
data/Gemfile
ADDED
@@ -0,0 +1,13 @@
|
|
1
|
+
source "http://rubygems.org"
|
2
|
+
# Add dependencies required to use your gem here.
|
3
|
+
gem "ffi"
|
4
|
+
|
5
|
+
# Add dependencies to develop your gem here.
|
6
|
+
# Include everything needed to run rake, tests, features, etc.
|
7
|
+
group :development do
|
8
|
+
gem "shoulda"
|
9
|
+
gem "rdoc"
|
10
|
+
gem "bundler"
|
11
|
+
gem "jeweler"
|
12
|
+
gem "simplecov"
|
13
|
+
end
|
data/LICENSE.txt
ADDED
@@ -0,0 +1,20 @@
|
|
1
|
+
Copyright (c) 2012 Nicholas A. Thrower
|
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.rdoc
ADDED
@@ -0,0 +1,82 @@
|
|
1
|
+
= bio-tabix
|
2
|
+
|
3
|
+
Ruby binding for the tabix file indexing routines within the samtools package
|
4
|
+
http://samtools.sourceforge.net/
|
5
|
+
|
6
|
+
Tabix provides utilities for indexing and subsequently querying regions of interest from large tab delimited files.
|
7
|
+
|
8
|
+
Files are indexed on three columns: [group, pos1, pos2] and must be position sorted
|
9
|
+
|
10
|
+
This gem was modeled on the bio-samtools gem: https://github.com/helios/bioruby-samtools
|
11
|
+
|
12
|
+
== Installation
|
13
|
+
gem install bio-tabix
|
14
|
+
|
15
|
+
== Usage
|
16
|
+
=== Compress your file
|
17
|
+
The following command will run the bgzip utility.
|
18
|
+
You must supply input and output filenames. If you don't compress your
|
19
|
+
data first, it will be compressed when the file is opened appending '.bgzf' to the filename.
|
20
|
+
Bio::Tabix::TFile.compress(in_file,compressed_file)
|
21
|
+
|
22
|
+
=== Index the file
|
23
|
+
Build an index by supplying the required columns.
|
24
|
+
If you don't create an index the open method will use the default options.
|
25
|
+
The opts parameter takes a hash of index build options
|
26
|
+
- :s => sequence/group column [1]
|
27
|
+
- :b => beginning range column [2]
|
28
|
+
- :e => ending range column. Can equal :b [3]
|
29
|
+
- :meta_char => comment character [#]
|
30
|
+
- :line_skip => number of initial lines to ignore [0]
|
31
|
+
tabix_file = Bio::Tabix::TFile.build_index(compressed_file, {:s => group_col, :b => pos1_col, :e => pos2_col})
|
32
|
+
|
33
|
+
|
34
|
+
=== Open your file
|
35
|
+
Create a new TFile instance and open it.
|
36
|
+
tabix_file = Bio::Tabix::TFile.open(compressed_file)
|
37
|
+
|
38
|
+
=== Create a proc or lambda
|
39
|
+
This will be called with the value of each fileline
|
40
|
+
my_func = lambda do |line|
|
41
|
+
# convert text to array and print column index 3
|
42
|
+
puts line.split("\t")[3]
|
43
|
+
end
|
44
|
+
|
45
|
+
=== Process a region
|
46
|
+
Choose a group and range for your function and process it
|
47
|
+
tabix_file.process_region('g2', 4, 10, my_func)
|
48
|
+
|
49
|
+
== Example
|
50
|
+
=== File
|
51
|
+
g1 1 3 a1 b1
|
52
|
+
g1 2 3 a2 b2
|
53
|
+
g1 3 4 a3 b3
|
54
|
+
...
|
55
|
+
g2 1 2 A1 B1
|
56
|
+
g2 2 4 A2 B2
|
57
|
+
g2 3 5 A3 B3
|
58
|
+
g2 4 6 A4 B4
|
59
|
+
...
|
60
|
+
|
61
|
+
=== Output
|
62
|
+
A2
|
63
|
+
A3
|
64
|
+
A4
|
65
|
+
|
66
|
+
== Dependencies:
|
67
|
+
- FFI (http://github.com/ffi/ffi)
|
68
|
+
|
69
|
+
== Contributing to bio-tabix
|
70
|
+
* Check out the latest master to make sure the feature hasn't been implemented or the bug hasn't been fixed yet.
|
71
|
+
* Check out the issue tracker to make sure someone already hasn't requested it and/or contributed it.
|
72
|
+
* Fork the project.
|
73
|
+
* Start a feature/bugfix branch.
|
74
|
+
* Commit and push until you are happy with your contribution.
|
75
|
+
* Make sure to add tests for it. This is important so I don't break it in a future version unintentionally.
|
76
|
+
* Please try not to mess with the Rakefile, version, or history. If you want to have your own version, or is otherwise necessary, that is fine, but please isolate to its own commit so I can cherry-pick around it.
|
77
|
+
|
78
|
+
== Copyright
|
79
|
+
|
80
|
+
Copyright (c) 2012 Nicholas A Thrower. See LICENSE.txt for
|
81
|
+
further details.
|
82
|
+
|
data/Rakefile
ADDED
@@ -0,0 +1,59 @@
|
|
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-tabix"
|
18
|
+
gem.homepage = "http://github.com/throwern/bio-tabix"
|
19
|
+
gem.license = "MIT"
|
20
|
+
gem.summary = %Q{Ruby binding for samtools tabix}
|
21
|
+
gem.description = %Q{Tabix file indexing routines from the samtools package http://samtools.sourceforge.net/}
|
22
|
+
gem.email = "throwern@msu.edu"
|
23
|
+
gem.authors = ["throwern"]
|
24
|
+
# dependencies defined in Gemfile
|
25
|
+
gem.extensions = "ext/tabix/mkrf_conf.rb"
|
26
|
+
end
|
27
|
+
Jeweler::RubygemsDotOrgTasks.new
|
28
|
+
|
29
|
+
require 'rake/testtask'
|
30
|
+
Rake::TestTask.new(:test) do |test|
|
31
|
+
test.libs << 'lib' << 'test'
|
32
|
+
test.pattern = 'test/**/test_*.rb'
|
33
|
+
test.verbose = true
|
34
|
+
end
|
35
|
+
|
36
|
+
# require 'rcov/rcovtask'
|
37
|
+
# Rcov::RcovTask.new do |test|
|
38
|
+
# test.libs << 'test'
|
39
|
+
# test.pattern = 'test/**/test_*.rb'
|
40
|
+
# test.verbose = true
|
41
|
+
# test.rcov_opts << '--exclude "gems/*"'
|
42
|
+
# end
|
43
|
+
desc "Code coverage detail"
|
44
|
+
task :simplecov do
|
45
|
+
ENV['COVERAGE'] = "true"
|
46
|
+
Rake::Task['spec'].execute
|
47
|
+
end
|
48
|
+
|
49
|
+
task :default => :test
|
50
|
+
|
51
|
+
require 'rdoc/task'
|
52
|
+
Rake::RDocTask.new do |rdoc|
|
53
|
+
version = File.exist?('VERSION') ? File.read('VERSION') : ""
|
54
|
+
|
55
|
+
rdoc.rdoc_dir = 'rdoc'
|
56
|
+
rdoc.title = "bio-tabix #{version}"
|
57
|
+
rdoc.rdoc_files.include('README*')
|
58
|
+
rdoc.rdoc_files.include('lib/**/*.rb')
|
59
|
+
end
|
data/VERSION
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
0.1.2
|
data/bio-tabix.gemspec
ADDED
@@ -0,0 +1,72 @@
|
|
1
|
+
# Generated by jeweler
|
2
|
+
# DO NOT EDIT THIS FILE DIRECTLY
|
3
|
+
# Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
|
4
|
+
# -*- encoding: utf-8 -*-
|
5
|
+
|
6
|
+
Gem::Specification.new do |s|
|
7
|
+
s.name = "bio-tabix"
|
8
|
+
s.version = "0.1.2"
|
9
|
+
|
10
|
+
s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
|
11
|
+
s.authors = ["throwern"]
|
12
|
+
s.date = "2012-05-07"
|
13
|
+
s.description = "Tabix file indexing routines from the samtools package http://samtools.sourceforge.net/"
|
14
|
+
s.email = "throwern@msu.edu"
|
15
|
+
s.extensions = ["ext/tabix/mkrf_conf.rb"]
|
16
|
+
s.extra_rdoc_files = [
|
17
|
+
"LICENSE.txt",
|
18
|
+
"README.rdoc"
|
19
|
+
]
|
20
|
+
s.files = [
|
21
|
+
".document",
|
22
|
+
".rvmrc",
|
23
|
+
"Gemfile",
|
24
|
+
"LICENSE.txt",
|
25
|
+
"README.rdoc",
|
26
|
+
"Rakefile",
|
27
|
+
"VERSION",
|
28
|
+
"bio-tabix.gemspec",
|
29
|
+
"ext/tabix/Rakefile",
|
30
|
+
"ext/tabix/mkrf_conf.rb",
|
31
|
+
"lib/bio-tabix.rb",
|
32
|
+
"lib/bio/tabix/Version",
|
33
|
+
"lib/bio/tabix/binding.rb",
|
34
|
+
"lib/bio/tabix/library.rb",
|
35
|
+
"lib/bio/tabix/t_file.rb",
|
36
|
+
"test/helper.rb",
|
37
|
+
"test/test_bio-tabix.rb"
|
38
|
+
]
|
39
|
+
s.homepage = "http://github.com/throwern/bio-tabix"
|
40
|
+
s.licenses = ["MIT"]
|
41
|
+
s.require_paths = ["lib"]
|
42
|
+
s.rubygems_version = "1.8.24"
|
43
|
+
s.summary = "Ruby binding for samtools tabix"
|
44
|
+
|
45
|
+
if s.respond_to? :specification_version then
|
46
|
+
s.specification_version = 3
|
47
|
+
|
48
|
+
if Gem::Version.new(Gem::VERSION) >= Gem::Version.new('1.2.0') then
|
49
|
+
s.add_runtime_dependency(%q<ffi>, [">= 0"])
|
50
|
+
s.add_development_dependency(%q<shoulda>, [">= 0"])
|
51
|
+
s.add_development_dependency(%q<rdoc>, [">= 0"])
|
52
|
+
s.add_development_dependency(%q<bundler>, [">= 0"])
|
53
|
+
s.add_development_dependency(%q<jeweler>, [">= 0"])
|
54
|
+
s.add_development_dependency(%q<simplecov>, [">= 0"])
|
55
|
+
else
|
56
|
+
s.add_dependency(%q<ffi>, [">= 0"])
|
57
|
+
s.add_dependency(%q<shoulda>, [">= 0"])
|
58
|
+
s.add_dependency(%q<rdoc>, [">= 0"])
|
59
|
+
s.add_dependency(%q<bundler>, [">= 0"])
|
60
|
+
s.add_dependency(%q<jeweler>, [">= 0"])
|
61
|
+
s.add_dependency(%q<simplecov>, [">= 0"])
|
62
|
+
end
|
63
|
+
else
|
64
|
+
s.add_dependency(%q<ffi>, [">= 0"])
|
65
|
+
s.add_dependency(%q<shoulda>, [">= 0"])
|
66
|
+
s.add_dependency(%q<rdoc>, [">= 0"])
|
67
|
+
s.add_dependency(%q<bundler>, [">= 0"])
|
68
|
+
s.add_dependency(%q<jeweler>, [">= 0"])
|
69
|
+
s.add_dependency(%q<simplecov>, [">= 0"])
|
70
|
+
end
|
71
|
+
end
|
72
|
+
|
data/ext/tabix/Rakefile
ADDED
@@ -0,0 +1,73 @@
|
|
1
|
+
#(c) Copyright 2012 Nicholas A Thrower. All Rights Reserved.
|
2
|
+
# Derivative of https://github.com/helios/bioruby-samtools/blob/master/ext/mkrf_conf.rb
|
3
|
+
# create Rakefile for shared library compilation
|
4
|
+
|
5
|
+
path = File.expand_path(File.dirname(__FILE__))
|
6
|
+
|
7
|
+
path_external = File.join(path, "../../lib/bio/tabix/")
|
8
|
+
|
9
|
+
version = File.open(File.join(path_external,"Version"),'r')
|
10
|
+
Version = version.read
|
11
|
+
version.close
|
12
|
+
|
13
|
+
url = "http://samtools.svn.sourceforge.net/viewvc/samtools/trunk/tabix/?view=tar"
|
14
|
+
TabixFile = "tabix-trunk.tar"
|
15
|
+
|
16
|
+
File.open(File.join(path,"Rakefile"),"w") do |rakefile|
|
17
|
+
rakefile.write <<-RAKE
|
18
|
+
require 'rbconfig'
|
19
|
+
require 'open-uri'
|
20
|
+
require 'fileutils'
|
21
|
+
include FileUtils::Verbose
|
22
|
+
require 'rake/clean'
|
23
|
+
|
24
|
+
URL = "#{url}"
|
25
|
+
|
26
|
+
task :download do
|
27
|
+
open(URL) do |uri|
|
28
|
+
File.open("#{TabixFile}",'wb') do |fout|
|
29
|
+
fout.write(uri.read)
|
30
|
+
end #fout
|
31
|
+
end #uri
|
32
|
+
end
|
33
|
+
|
34
|
+
task :compile do
|
35
|
+
sh "tar xvf #{TabixFile}"
|
36
|
+
cd("tabix") do
|
37
|
+
#sh "patch < ../Makefile-bioruby.patch"
|
38
|
+
sh "make"
|
39
|
+
cp("libtabix.a","#{path_external}")
|
40
|
+
case Config::CONFIG['host_os']
|
41
|
+
when /linux/
|
42
|
+
sh "make libtabix.so.1"
|
43
|
+
cp("libtabix.so.1","#{path_external}")
|
44
|
+
when /darwin/
|
45
|
+
sh "make libtabix.1.dylib"
|
46
|
+
cp("libtabix.1.dylib","#{path_external}")
|
47
|
+
else raise NotImplementedError, "Tabix not supported on your platform"
|
48
|
+
end #case
|
49
|
+
cp("tabix", "#{path}/../../lib/bio/tabix")
|
50
|
+
chmod 0755, "#{path}/../../lib/bio/tabix/tabix"
|
51
|
+
cp("bgzip", "#{path}/../../lib/bio/tabix")
|
52
|
+
chmod 0755, "#{path}/../../lib/bio/tabix/bgzip"
|
53
|
+
end #cd
|
54
|
+
end
|
55
|
+
|
56
|
+
task :clean do
|
57
|
+
# cd("tabix-#{Version}") do
|
58
|
+
# sh "make clean"
|
59
|
+
# end
|
60
|
+
# rm("#{TabixFile}")
|
61
|
+
# rm_rf("tabix-#{Version}")
|
62
|
+
cd("tabix") do
|
63
|
+
sh "make clean"
|
64
|
+
end
|
65
|
+
rm("#{TabixFile}")
|
66
|
+
rm_rf("tabix")
|
67
|
+
end
|
68
|
+
|
69
|
+
task :default => [:download, :compile, :clean]
|
70
|
+
|
71
|
+
RAKE
|
72
|
+
|
73
|
+
end
|
data/lib/bio-tabix.rb
ADDED
@@ -0,0 +1 @@
|
|
1
|
+
0.2.5
|
@@ -0,0 +1,100 @@
|
|
1
|
+
# == binding.rb
|
2
|
+
# This file contains the ffi binding declarations for the tabix api
|
3
|
+
# See https://github.com/ffi/ffi and http://samtools.sourceforge.net/tabix.shtml for details
|
4
|
+
#
|
5
|
+
# == Contact
|
6
|
+
#
|
7
|
+
# Author:: Nicholas A. Thrower
|
8
|
+
# Copyright:: Copyright (c) 2012 Nicholas A Thrower
|
9
|
+
# License:: See LICENSE.txt for more details
|
10
|
+
#
|
11
|
+
|
12
|
+
# -
|
13
|
+
module Bio
|
14
|
+
# -
|
15
|
+
module Tabix
|
16
|
+
# Ruby binding for the tabix file indexing routines within the samtools package http://samtools.sourceforge.net/
|
17
|
+
module Binding
|
18
|
+
require 'bio/tabix/library'
|
19
|
+
extend FFI::Library
|
20
|
+
ffi_lib Bio::Tabix::Library.filename
|
21
|
+
|
22
|
+
# CLASSES
|
23
|
+
|
24
|
+
# Custom string storage
|
25
|
+
# member of the IterT class
|
26
|
+
class KString < FFI::Struct
|
27
|
+
layout(
|
28
|
+
:l,:size_t,
|
29
|
+
:m,:size_t,
|
30
|
+
:s,:string
|
31
|
+
)
|
32
|
+
end
|
33
|
+
|
34
|
+
# File pointers to text and index data
|
35
|
+
# created by ti_open
|
36
|
+
# used by ti_read, ti_query, ti_close
|
37
|
+
class TabixT < FFI::Struct
|
38
|
+
layout(
|
39
|
+
:fp, :pointer,
|
40
|
+
:idx, :pointer,
|
41
|
+
:fn, :string,
|
42
|
+
:fnidx, :string
|
43
|
+
)
|
44
|
+
end
|
45
|
+
|
46
|
+
# Iteratator for monitoring the query progress
|
47
|
+
# created by the ti_query method
|
48
|
+
# used by ti_read
|
49
|
+
class IterT < FFI::Struct
|
50
|
+
layout(
|
51
|
+
:from_first,:int,
|
52
|
+
:tid, :int,
|
53
|
+
:beg, :int,
|
54
|
+
:end, :int,
|
55
|
+
:n_off, :int,
|
56
|
+
:i, :int,
|
57
|
+
:finished, :int,
|
58
|
+
:curr_off, :uint64,
|
59
|
+
:str, KString,
|
60
|
+
:idx, :pointer,
|
61
|
+
:off, :pointer
|
62
|
+
)
|
63
|
+
end
|
64
|
+
# Index configuration
|
65
|
+
# used by ti_index_build2
|
66
|
+
class ConfT < FFI::Struct
|
67
|
+
layout(
|
68
|
+
:preset, :int32,
|
69
|
+
:sc, :int32,
|
70
|
+
:bc, :int32,
|
71
|
+
:ec, :int32,
|
72
|
+
:meta_char, :int32,
|
73
|
+
:line_skip, :int32
|
74
|
+
)
|
75
|
+
# convenience method to access attributes
|
76
|
+
def get_hash
|
77
|
+
{
|
78
|
+
:preset => self[:preset],
|
79
|
+
:sc => self[:sc],
|
80
|
+
:bc => self[:bc],
|
81
|
+
:ec => self[:ec],
|
82
|
+
:meta_char => self[:meta_char],
|
83
|
+
:line_skip => self[:line_skip]
|
84
|
+
}
|
85
|
+
end
|
86
|
+
end
|
87
|
+
# FUNCTIONS # PARAMETER(S) : RETURN
|
88
|
+
attach_function :ti_open, [:string, :string], :pointer # filename, idxname (or 0) : TabixT*
|
89
|
+
attach_function :ti_read, [:pointer, :pointer, :pointer], :string # TabixT*, ti_iter_t, len : string
|
90
|
+
attach_function :ti_query, [:pointer,:string,:int,:int], IterT # TabixT*, name, beg, end : IterT
|
91
|
+
attach_function :ti_close, [:pointer], :void # TabixT*
|
92
|
+
attach_function :ti_iter_destroy, [IterT], :void # ti_iter_t
|
93
|
+
attach_function :ti_index_build2, [:string,:pointer,:string], :int # filename, ti_conf_t, idxname (or 0) : 0/-1
|
94
|
+
attach_function :bgzf_is_bgzf, [:string], :int # filename, : 1/0
|
95
|
+
attach_function :ti_seqname, [:pointer,:pointer],:pointer # ti_index_t*, int*(count) : char**
|
96
|
+
attach_function :ti_index_load,[:string],:pointer # filename(no idx suffix) : ti_index_t*
|
97
|
+
attach_function :ti_get_conf,[:pointer],:pointer # ti_index_t* : ti_conf_t*
|
98
|
+
end
|
99
|
+
end
|
100
|
+
end
|
@@ -0,0 +1,38 @@
|
|
1
|
+
# == library.rb
|
2
|
+
# This file contains the Library Class for retrieving platform specific library names
|
3
|
+
#
|
4
|
+
# == Contact
|
5
|
+
#
|
6
|
+
# Author:: Nicholas A. Thrower
|
7
|
+
# Copyright:: Copyright (c) 2012 Nicholas A Thrower
|
8
|
+
# License:: See LICENSE.txt for more details
|
9
|
+
#
|
10
|
+
|
11
|
+
# -
|
12
|
+
module Bio
|
13
|
+
# -
|
14
|
+
module Tabix
|
15
|
+
# Cross-platform library naming
|
16
|
+
class Library
|
17
|
+
# return the platform specific library name
|
18
|
+
def self.filename
|
19
|
+
lib_os = case RUBY_PLATFORM
|
20
|
+
when /linux/
|
21
|
+
'so.1'
|
22
|
+
when /darwin/
|
23
|
+
'1.dylib'
|
24
|
+
else
|
25
|
+
case RUBY_DESCRIPTION
|
26
|
+
when /darwin.*java/
|
27
|
+
'1.dylib'
|
28
|
+
when /linux.*java/
|
29
|
+
'so.1'
|
30
|
+
else raise NotImplementedError, "Tabix not supported on your platform"
|
31
|
+
end
|
32
|
+
end
|
33
|
+
|
34
|
+
File.join(File.expand_path(File.dirname(__FILE__)),"libtabix.#{lib_os}")
|
35
|
+
end
|
36
|
+
end
|
37
|
+
end
|
38
|
+
end
|
@@ -0,0 +1,158 @@
|
|
1
|
+
# == t_file.rb
|
2
|
+
# This file contains the TFile class used to interact with the tabix api
|
3
|
+
#
|
4
|
+
# == Contact
|
5
|
+
#
|
6
|
+
# Author:: Nicholas A. Thrower
|
7
|
+
# Copyright:: Copyright (c) 2012 Nicholas A Thrower
|
8
|
+
# License:: See LICENSE.txt for more details
|
9
|
+
#
|
10
|
+
|
11
|
+
# -
|
12
|
+
module Bio
|
13
|
+
# -
|
14
|
+
module Tabix
|
15
|
+
# The TFile class manages compressing, indexing, opening and parsing tab delimited files.
|
16
|
+
# The file must be position sorted prior to indexing.
|
17
|
+
class TFile
|
18
|
+
require 'bio/tabix/binding'
|
19
|
+
include Bio::Tabix::Binding
|
20
|
+
# ascii or compressed file name
|
21
|
+
attr_accessor :file
|
22
|
+
# index name
|
23
|
+
attr_accessor :index
|
24
|
+
# TabixT created from open index
|
25
|
+
attr_accessor :t_file
|
26
|
+
# pointer to TabixT
|
27
|
+
attr_accessor :t_file_p
|
28
|
+
# index build options
|
29
|
+
attr_accessor :options
|
30
|
+
# compresses the fi into fo using bgzip
|
31
|
+
def self.compress(fi, fo)
|
32
|
+
`#{File.join(File.expand_path(File.dirname(__FILE__)),'bgzip')} -c #{fi} > #{fo}`
|
33
|
+
end
|
34
|
+
# Builds an index from the supplied filename and options
|
35
|
+
# - :s => sequence/group column [1]
|
36
|
+
# - :b => beginning range column [2]
|
37
|
+
# - :e => ending range column. Can equal :b. [3]
|
38
|
+
# - :meta_char => comment character [#]
|
39
|
+
# - :line_skip => number of initial lines to ignore [0]
|
40
|
+
def self.build_index(f, opts={})
|
41
|
+
conf = ConfT.new
|
42
|
+
conf[:preset]=0
|
43
|
+
conf[:sc]=opts[:s] || 1
|
44
|
+
conf[:bc]=opts[:b] || 2
|
45
|
+
conf[:ec]=opts[:e] || 3
|
46
|
+
conf[:meta_char]=('#'||opts[:c][0]).ord
|
47
|
+
conf[:line_skip]=(0||opts[:S]).to_i
|
48
|
+
unless(Bio::Tabix::Binding.bgzf_is_bgzf(f)==1)
|
49
|
+
puts "Compressing..."
|
50
|
+
self.class.compress(f,f+".bgzf")
|
51
|
+
f=f+".bgzf"
|
52
|
+
end
|
53
|
+
puts "Indexing with #{conf.get_hash}..."
|
54
|
+
Bio::Tabix::Binding.ti_index_build2(f,conf,f+".tbi")
|
55
|
+
end
|
56
|
+
# convenience method to create a new Tabix instance and open it.
|
57
|
+
def self.open(*args)
|
58
|
+
self.new(*args).open
|
59
|
+
end
|
60
|
+
# Returns a new TFile. If the file is not compressed, a new compressed
|
61
|
+
# file will be created with compress[compress]. If the index is not present
|
62
|
+
# a new index will be created with build_index[build_index].
|
63
|
+
def initialize(f, opts={})
|
64
|
+
@file = f
|
65
|
+
@options = opts
|
66
|
+
@index = file+".tbi"
|
67
|
+
return self
|
68
|
+
end
|
69
|
+
# opens the file checking for compression and corresponding index.
|
70
|
+
def open
|
71
|
+
# check existing
|
72
|
+
if(@t_file)
|
73
|
+
puts "Already open, closing and re-opening"
|
74
|
+
self.close
|
75
|
+
end
|
76
|
+
# check datafile
|
77
|
+
if file =~ /http:\/\/|ftp:\/\//
|
78
|
+
puts "Expecting remote file: #{file}"
|
79
|
+
else
|
80
|
+
raise "FileNotFound #{file}" unless(File.exist?(file))
|
81
|
+
unless(bgzf_is_bgzf(file)==1)
|
82
|
+
unless(bgzf_is_bgzf(file+".bgzf")==1)
|
83
|
+
puts "Input does not look like a bgzip compressed file. Attempting compression..."
|
84
|
+
self.class.compress(file,file+".bgzf")
|
85
|
+
end
|
86
|
+
@file = file+".bgzf"
|
87
|
+
end
|
88
|
+
end
|
89
|
+
# check index
|
90
|
+
if index =~ /http:\/\/|ftp:\/\//
|
91
|
+
puts "Expecting remote index: #{index}"
|
92
|
+
elsif !File.exist?(index)
|
93
|
+
puts "Index #{index} not found. Building..."
|
94
|
+
self.class.build_index(file,options)
|
95
|
+
end
|
96
|
+
# open
|
97
|
+
@t_file_p = ti_open(file,index)
|
98
|
+
raise "FileAcessError #{file}" if @t_file_p.null?
|
99
|
+
@t_file = TabixT.new(@t_file_p)
|
100
|
+
return self
|
101
|
+
end
|
102
|
+
# closes the TabixT file
|
103
|
+
def close
|
104
|
+
if(@t_file_p)
|
105
|
+
begin
|
106
|
+
ti_close(@t_file_p)
|
107
|
+
@t_file_p = nil
|
108
|
+
rescue
|
109
|
+
puts "Error closing file"
|
110
|
+
end
|
111
|
+
end
|
112
|
+
end
|
113
|
+
# returns an array of the group names found in the index
|
114
|
+
def groups
|
115
|
+
load_index
|
116
|
+
g_num = FFI::MemoryPointer.new(:int)
|
117
|
+
g_ptr = ti_seqname(t_file[:idx],g_num)
|
118
|
+
return [] if g_ptr.null? || g_num.null?
|
119
|
+
g_ptr.get_array_of_string(0, g_num.read_int).compact
|
120
|
+
end
|
121
|
+
# returns the header (skipped lines + comments)
|
122
|
+
def header
|
123
|
+
load_index
|
124
|
+
conf = ConfT.new(ti_get_conf(t_file[:idx]))
|
125
|
+
iter = IterT.new(ti_query(t_file_p,nil,0,1))
|
126
|
+
len = FFI::MemoryPointer.new(:int)
|
127
|
+
str = ""
|
128
|
+
while( (s = ti_read(t_file_p, iter, len)) )
|
129
|
+
break if(s[0].ord != conf[:meta_char])
|
130
|
+
str << s
|
131
|
+
str << "\n"
|
132
|
+
end
|
133
|
+
ti_iter_destroy(iter)
|
134
|
+
@header = str
|
135
|
+
end
|
136
|
+
# Iterates over the supplied region calling user_proc on each item
|
137
|
+
# a region is defined by a group name and range(pos1 - pos2)
|
138
|
+
# all overlapping intervals within the group will be processed in order
|
139
|
+
def process_region(group, pos1, pos2, user_proc)
|
140
|
+
iter = IterT.new(ti_query(t_file_p,group,pos1,pos2))
|
141
|
+
len = FFI::MemoryPointer.new(:int)
|
142
|
+
while( (s = ti_read(t_file_p, iter, len)) )
|
143
|
+
user_proc.call(s,len)
|
144
|
+
end
|
145
|
+
ti_iter_destroy(iter)
|
146
|
+
end
|
147
|
+
|
148
|
+
private
|
149
|
+
def load_index
|
150
|
+
if t_file[:idx].null?
|
151
|
+
t_file[:idx] = ti_index_load(t_file[:fn])
|
152
|
+
end
|
153
|
+
raise "Index Load Error" if t_file[:idx].null?
|
154
|
+
end
|
155
|
+
|
156
|
+
end#Index class
|
157
|
+
end#Tabix module
|
158
|
+
end#Bio module
|
data/test/helper.rb
ADDED
@@ -0,0 +1,18 @@
|
|
1
|
+
require 'rubygems'
|
2
|
+
require 'bundler'
|
3
|
+
begin
|
4
|
+
Bundler.setup(:default, :development)
|
5
|
+
rescue Bundler::BundlerError => e
|
6
|
+
$stderr.puts e.message
|
7
|
+
$stderr.puts "Run `bundle install` to install missing gems"
|
8
|
+
exit e.status_code
|
9
|
+
end
|
10
|
+
require 'test/unit'
|
11
|
+
require 'shoulda'
|
12
|
+
|
13
|
+
$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
|
14
|
+
$LOAD_PATH.unshift(File.dirname(__FILE__))
|
15
|
+
require 'bio-tabix'
|
16
|
+
|
17
|
+
class Test::Unit::TestCase
|
18
|
+
end
|
metadata
ADDED
@@ -0,0 +1,164 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: bio-tabix
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.1.2
|
5
|
+
prerelease:
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- throwern
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
date: 2012-05-07 00:00:00.000000000 Z
|
13
|
+
dependencies:
|
14
|
+
- !ruby/object:Gem::Dependency
|
15
|
+
name: ffi
|
16
|
+
requirement: !ruby/object:Gem::Requirement
|
17
|
+
none: false
|
18
|
+
requirements:
|
19
|
+
- - ! '>='
|
20
|
+
- !ruby/object:Gem::Version
|
21
|
+
version: '0'
|
22
|
+
type: :runtime
|
23
|
+
prerelease: false
|
24
|
+
version_requirements: !ruby/object:Gem::Requirement
|
25
|
+
none: false
|
26
|
+
requirements:
|
27
|
+
- - ! '>='
|
28
|
+
- !ruby/object:Gem::Version
|
29
|
+
version: '0'
|
30
|
+
- !ruby/object:Gem::Dependency
|
31
|
+
name: shoulda
|
32
|
+
requirement: !ruby/object:Gem::Requirement
|
33
|
+
none: false
|
34
|
+
requirements:
|
35
|
+
- - ! '>='
|
36
|
+
- !ruby/object:Gem::Version
|
37
|
+
version: '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: '0'
|
46
|
+
- !ruby/object:Gem::Dependency
|
47
|
+
name: rdoc
|
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: bundler
|
64
|
+
requirement: !ruby/object:Gem::Requirement
|
65
|
+
none: false
|
66
|
+
requirements:
|
67
|
+
- - ! '>='
|
68
|
+
- !ruby/object:Gem::Version
|
69
|
+
version: '0'
|
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: '0'
|
78
|
+
- !ruby/object:Gem::Dependency
|
79
|
+
name: jeweler
|
80
|
+
requirement: !ruby/object:Gem::Requirement
|
81
|
+
none: false
|
82
|
+
requirements:
|
83
|
+
- - ! '>='
|
84
|
+
- !ruby/object:Gem::Version
|
85
|
+
version: '0'
|
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: '0'
|
94
|
+
- !ruby/object:Gem::Dependency
|
95
|
+
name: simplecov
|
96
|
+
requirement: !ruby/object:Gem::Requirement
|
97
|
+
none: false
|
98
|
+
requirements:
|
99
|
+
- - ! '>='
|
100
|
+
- !ruby/object:Gem::Version
|
101
|
+
version: '0'
|
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: '0'
|
110
|
+
description: Tabix file indexing routines from the samtools package http://samtools.sourceforge.net/
|
111
|
+
email: throwern@msu.edu
|
112
|
+
executables: []
|
113
|
+
extensions:
|
114
|
+
- ext/tabix/mkrf_conf.rb
|
115
|
+
extra_rdoc_files:
|
116
|
+
- LICENSE.txt
|
117
|
+
- README.rdoc
|
118
|
+
files:
|
119
|
+
- .document
|
120
|
+
- .rvmrc
|
121
|
+
- Gemfile
|
122
|
+
- LICENSE.txt
|
123
|
+
- README.rdoc
|
124
|
+
- Rakefile
|
125
|
+
- VERSION
|
126
|
+
- bio-tabix.gemspec
|
127
|
+
- ext/tabix/Rakefile
|
128
|
+
- ext/tabix/mkrf_conf.rb
|
129
|
+
- lib/bio-tabix.rb
|
130
|
+
- lib/bio/tabix/Version
|
131
|
+
- lib/bio/tabix/binding.rb
|
132
|
+
- lib/bio/tabix/library.rb
|
133
|
+
- lib/bio/tabix/t_file.rb
|
134
|
+
- test/helper.rb
|
135
|
+
- test/test_bio-tabix.rb
|
136
|
+
homepage: http://github.com/throwern/bio-tabix
|
137
|
+
licenses:
|
138
|
+
- MIT
|
139
|
+
post_install_message:
|
140
|
+
rdoc_options: []
|
141
|
+
require_paths:
|
142
|
+
- lib
|
143
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
144
|
+
none: false
|
145
|
+
requirements:
|
146
|
+
- - ! '>='
|
147
|
+
- !ruby/object:Gem::Version
|
148
|
+
version: '0'
|
149
|
+
segments:
|
150
|
+
- 0
|
151
|
+
hash: -2955396723264547364
|
152
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
153
|
+
none: false
|
154
|
+
requirements:
|
155
|
+
- - ! '>='
|
156
|
+
- !ruby/object:Gem::Version
|
157
|
+
version: '0'
|
158
|
+
requirements: []
|
159
|
+
rubyforge_project:
|
160
|
+
rubygems_version: 1.8.24
|
161
|
+
signing_key:
|
162
|
+
specification_version: 3
|
163
|
+
summary: Ruby binding for samtools tabix
|
164
|
+
test_files: []
|