bio-cd-hit-report 0.0.3 → 0.1.0

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    MTZhMmJjOTZlZTAzYzg3ZGUwZDdiMGNmNmI2NjRkNzAzYmE0NjM2Mw==
+  data.tar.gz: !binary |-
+    ZDYyZmIyMjM0NzdlY2UzZjI2ZGQwYmQxOTQwOGNlYjQ3Mzk1Zjc5OA==
+!binary "U0hBNTEy":
+  metadata.gz: !binary |-
+    MzFjN2Q5YjA1Mzk4MGIxNGM1MDFjMWYwNjVmNGEyNmEzYmM1NjIxODFlZjJi
+    NjVmMjRmYTkwYzgwZTcyZDE0MzY5NDljZGY5NjliYTQ3ZWJlNWIyOGRlYTZi
+    ZGI0NjJjNjc5ZDI4NWNiN2MzYzUwNjM2Zjk2MzcwZjYwMWQ3NDY=
+  data.tar.gz: !binary |-
+    NWVmYjRkMmE1YTczOGY5MGFmNTFiMDdiMjlmMTdhYTEyNmI5Njg1OTQ5NmQ3
+    OTgzNTAyOGY4M2Q0ZjM0Y2JhN2FlMDRhNmQ3MGIzMDZlN2Q3Y2JmMTRiOTAz
+    MjYxOGQyNDY1NTgxOThkMjI5Zjk5MTUwMGNhOTA5YTBlZDM5OGQ=

data/Gemfile

@@ -1,15 +1,11 @@
 source "http://rubygems.org"
-# Add dependencies required to use your gem here.
-# Example:
-#   gem "activesupport", ">= 2.3.5"
 
-# Add dependencies to develop your gem here.
-# Include everything needed to run rake, tests, features, etc.
+  gem "bio", "1.4.3"
+
 group :development do
   gem "minitest"
   gem "rdoc"
-  gem "bundler"
+  gem "bundler", ">=1.3.1"
   gem "jeweler"
-  gem "bio", "1.4.2"
   gem "rdoc"
 end

data/README.md

@@ -1,10 +1,12 @@
-# bio-cd-hit-report
+[[#]] bio-cd-hit-report
 
 [![Build Status](https://secure.travis-ci.org/georgeG/bioruby-cd-hit-report.png)](http://travis-ci.org/georgeG/bioruby-cd-hit-report)
 
-A bioruby wrapper for parsing and reading CD-HIT cluster reports
+Clustering sequences with CD-HIT produces a cluster file(.clstr)
+containing sequence names and their respective clusters. This plugin
+provides methods for parsing this file. 
 
-Note: this software is under active development!
+Note: this plugin is under active development!
 
 ## Installation
 
@@ -15,17 +17,11 @@ Note: this software is under active development!
 ## Usage
 
 ```ruby
-  require 'bio-cd-hit-report'
+    require 'bio-cd-hit-report'
    
-   cluster_file = "cluster95.clstr"
+    cluster_file = "cluster95.clstr"
     report = Bio::CdHitReport.new(cluster_file)
 
-      #print the max number of sequences in a cluster for the entire dataset
-      puts report.max_members 
-
-      #print the minimum number of sequences in a cluster for the entire dataset
-      puts report.min_members 
-      
       #print total number of clusters in the report
       puts report.total_clusters  
 
@@ -37,10 +33,13 @@ Note: this software is under active development!
         puts "#{c.name} - #{c.members}" #print cluster name/id with respective sequences in the cluster
         puts c.size #print the total number of entries in the cluster
       end
-```
+      
+      #print the representative sequence for each cluster
+      report.each_cluster do |c|
+         puts c.rep_seq 
+      end
 
-The API doc is online. For more code examples see the test files in
-the source tree.
+```
         
 ## Project home page
 
@@ -64,4 +63,4 @@ This Biogem is published at [#bio-cd-hit-report](http://biogems.info/index.html)
 
 ## Copyright
 
-Copyright (c) 2012 George Githinji. See LICENSE.txt for further details.
+Copyright (c) 2013 George Githinji. See LICENSE.txt for further details.

data/VERSION

@@ -1 +1 @@
-0.0.3
+0.1.0

data/bin/bio-cd-hit-report

@@ -1,74 +1,61 @@
 #!/usr/bin/env ruby
-#
+
 # BioRuby bio-cd-hit-report Plugin BioCdHitReport
-# Author:: georgeG
+# Author:: george githinji
 # Copyright:: 2012
 
- USAGE = "Describe bio-cd-hit-report"
+require_relative '../lib/bio-cd-hit-report'
+require 'ostruct'
+require 'optparse'
 
-if ARGV.size == 0
-  print USAGE
-end
+options = OpenStruct.new
 
-require 'bio-cd-hit-report'
-require 'optparse'
+OptionParser.new do |opts|
+  opts.banner = 'USAGE: bio-cd-hit-report -i file.clstr [options] '
 
-# Uncomment when using the bio-logger 
-# require 'bio-logger'
-# Bio::Log::CLI.logger('stderr')
-# Bio::Log::CLI.trace('info')
+  opts.on('-h', 'Display this screen') do
+    puts opts
+    exit
+  end
 
-options = {:example_switch=>false,:show_help=>false}
-opts = OptionParser.new do |o|
-  o.banner = "Usage: #{File.basename($0)} [options] reponame\ne.g. #{File.basename($0)} the-perfect-gem"
+  opts.on('-i','--infile CLUSTERFILE', 'cluster file') do |infile|
+    options.infile = infile
+  end
 
-  o.on('--example_parameter [EXAMPLE_PARAMETER]', 'TODO: put a description for the PARAMETER') do |example_parameter|
-    # TODO: your logic here, below an example
-    options[:example_parameter] = 'this is a parameter'
+  opts.on('-o','--outfile OUTPUTFILE', 'output file') do |outfile|
+    options.outfile = outfile
   end
-  
-  o.separator ""
-  o.on("--switch-example", 'TODO: put a description for the SWITCH') do
-    # TODO: your logic here, below an example
-    self[:example_switch] = true
+
+  opts.on('-m','--members') do
+    options.members = true
   end
 
-  # Uncomment the following when using the bio-logger 
-  # o.separator ""
-  # o.on("--logger filename",String,"Log to file (default stderr)") do | name |
-  #   Bio::Log::CLI.logger(name)
-  # end
-  #
-  # o.on("--trace options",String,"Set log level (default INFO, see bio-logger)") do | s |
-  #   Bio::Log::CLI.trace(s)
-  # end
-  # 
-  # o.on("-q", "--quiet", "Run quietly") do |q|
-  #   Bio::Log::CLI.trace('error')
-  # end
-  # 
-  # o.on("-v", "--verbose", "Run verbosely") do |v|
-  #   Bio::Log::CLI.trace('info')
-  # end
-  # 
-  # o.on("--debug", "Show debug messages") do |v|
-  #   Bio::Log::CLI.trace('debug')
-  # end
-
-  o.separator ""
-  o.on_tail('-h', '--help', 'display this help and exit') do
-    options[:show_help] = true
+  opts.on('-c','--clusterid CLUSTERID',Integer,'cluster id') do |clusterid|
+    options.cluster_id = clusterid
   end
+
+end.parse!
+
+clusterfile = options.infile
+outfile     = options.outfile
+cluster_id  = options.cluster_id
+
+report = Bio::CdHitReport.new(clusterfile)
+
+def print_members(report)
+ report.clusters.map{|c| "#{c.cluster_id}:#{c.members}"}
 end
 
-begin
-  opts.parse!(ARGV)
+def find_members_for(report,cluster_id)
+  report.get_members(cluster_id)
+end
 
-  # Uncomment the following when using the bio-logger 
-  # Bio::Log::CLI.configure('bio-cd-hit-report')
 
-  # TODO: your code here
-  # use options for your logic
-rescue OptionParser::InvalidOption => e
-  options[:invalid_argument] = e.message
+begin
+  unless cluster_id.nil?
+    $stdout.puts find_members_for(report,cluster_id)
+  else
+    $stdout.puts print_members(report) if options.members
+  end
 end
+

data/lib/bio-cd-hit-report.rb

@@ -1 +1 @@
-require_relative "bio-cd-hit-report/cd-hit-report"
+require_relative 'bio-cd-hit-report/cd-hit-report'

data/lib/bio-cd-hit-report/cd-hit-report.rb

@@ -1,49 +1,34 @@
 module Bio
-
-  require_relative 'cluster.rb'
+  require_relative 'cluster'
+  require_relative 'parser'
 
   class CdHitReport
+    include Enumerable
 
     def initialize(file)
-      @file = file
-    end
-
-    def each_cluster(&block)
-      cluster_objs.each(&block)
+      @report  = CdHitParser.new
+      @report.report_file = file
     end
 
-    def total_clusters
-      cluster_objs.size
-    end
-
-    def get_cluster(name)
-      cluster_objs.select{|cluster| cluster.name == name.to_s}.pop.members
-    end
-
-    def max_members
-      cluster_objs.map{|c|c.size}.max
+    def clusters
+      cls = []
+      @report.each do |c|
+        cls << c
+      end
+      cls
     end
 
-    def min_members
-      cluster_objs.map{|c| c.size}.min
+    def each_cluster(&block)
+      clusters.each(&block)
     end
 
-    private
-    def cluster_objs
-      d = raw_data.map do |line|
-        cluster = line.split("\n").delete_if{|x| x == ">Cluster "}
-        id = cluster.first
-        cluster.shift
-        #puts id.inspect
-        Cluster.new(id,cluster)
-      end
-      d.delete_if {|obj| obj.id.nil?}
+    def total_clusters
+      clusters.size
     end
 
-
-    def raw_data
-      File.open(@file).readlines
+    def get_members(cluster_id)
+      clusters.select {|cluster| cluster.cluster_id == cluster_id.to_s}.map{|c|c.members}
     end
-
-  end #class
-end #module
+    alias :get_cluster :get_members
+  end
+end

data/lib/bio-cd-hit-report/cluster.rb

@@ -1,21 +1,30 @@
-class Cluster < Struct.new(:name,:data)
-  $/ = ">Cluster "
+class Cluster
+  attr_accessor :name, :data
 
-  def id
-    name
+  def initialize(arg={})
+    self.name = arg[:name]
+    self.data = arg[:data] 
   end
 
-  def size
-    entries.size
+  def cluster_id
+    name.scan(/Cluster\s(.*)/).join
   end
 
   def members
     entries.join(',')
   end
 
-  private
+  def representative
+    data.split("\n").map{|line|line.scan(/>(.+)\.{3}\s\*/)}.join
+  end
+  alias :rep_seq :representative
+
+  def size
+    entries.size
+  end
+  alias :length :size
+
   def entries
-    data.map {|entry| entry.scan(/>(.+)\.{3}/)}.flatten
+    data.split("\n").map{|line|line.scan(/>(.+)\.{3}/)}
   end
 end
-

data/lib/bio-cd-hit-report/parser.rb

@@ -0,0 +1,17 @@
+class CdHitParser
+  attr_accessor :report_file
+
+  def each
+    data,header = nil, nil
+    File.open(report_file).each do |line|
+      if line[0].chr == '>'
+        yield Cluster.new(:name => header,:data => data) if data
+        data = ''
+        header = line[1..-1].strip
+      else
+        data << line
+      end
+    end
+    yield Cluster.new(:name => header, :data => data)
+  end
+end

data/test/data/test.clstr

@@ -0,0 +1,27 @@
+>Cluster 0
+0       420nt, >B267-17_Contig1... at +/99.76%
+1       456nt, >B50-25_Contig1... *
+2       456nt, >B59-19_Contig1... at +/99.78%
+3       456nt, >B63-12_Contig1... at +/99.56%
+4       456nt, >B63-3_Contig1... at +/99.34%
+>Cluster 1
+0       450nt, >B189-10_Contig1... *
+1       414nt, >B189-24_Contig1... at +/99.28%
+2       414nt, >B189-27_Contig1... at +/99.52%
+3       414nt, >B189-3_Contig1... at +/98.79%
+>Cluster 2
+0       447nt, >B118-11_Contig1... *
+>Cluster 3
+0       447nt, >B160-13_Contig1... *
+1       408nt, >B160-8_Contig1... at +/99.02%
+>Cluster 4
+0       444nt, >B216-14_Contig1... *
+>Cluster 5
+0       444nt, >B41-13_Contig1... *
+>Cluster 6
+0       441nt, >B139-18_Contig1... *
+1       441nt, >B139-26_Contig1... at +/99.77%
+2       441nt, >B139-28_Contig1... at +/99.55%
+3       441nt, >B170-26_Contig1... at +/98.64%
+4       441nt, >B219-31_Contig1... at +/99.55%
+>Cluster 7

metadata

@@ -1,36 +1,32 @@
 --- !ruby/object:Gem::Specification
 name: bio-cd-hit-report
 version: !ruby/object:Gem::Version
-  version: 0.0.3
-  prerelease: 
+  version: 0.1.0
 platform: ruby
 authors:
 - George Githinji
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-09-16 00:00:00.000000000 Z
+date: 2013-04-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
-  name: minitest
+  name: bio
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '='
       - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
+        version: 1.4.3
+  type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - ! '>='
+    - - '='
       - !ruby/object:Gem::Version
-        version: '0'
+        version: 1.4.3
 - !ruby/object:Gem::Dependency
-  name: rdoc
+  name: minitest
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -38,15 +34,13 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: bundler
+  name: rdoc
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -54,47 +48,41 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: jeweler
+  name: bundler
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
-        version: '0'
+        version: 1.3.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
-        version: '0'
+        version: 1.3.1
 - !ruby/object:Gem::Dependency
-  name: bio
+  name: jeweler
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - '='
+    - - ! '>='
       - !ruby/object:Gem::Version
-        version: 1.4.2
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
-    - - '='
+    - - ! '>='
       - !ruby/object:Gem::Version
-        version: 1.4.2
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: rdoc
   requirement: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -102,7 +90,6 @@ dependencies:
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
-    none: false
     requirements:
     - - ! '>='
       - !ruby/object:Gem::Version
@@ -127,34 +114,33 @@ files:
 - lib/bio-cd-hit-report.rb
 - lib/bio-cd-hit-report/cd-hit-report.rb
 - lib/bio-cd-hit-report/cluster.rb
+- lib/bio-cd-hit-report/parser.rb
+- test/data/test.clstr
 - test/helper.rb
 - test/test_bio-cd-hit-report.rb
 homepage: http://github.com/georgeG/bioruby-cd-hit-report
 licenses:
 - MIT
+metadata: {}
 post_install_message: 
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
-      segments:
-      - 0
-      hash: 3441050478878586342
 required_rubygems_version: !ruby/object:Gem::Requirement
-  none: false
   requirements:
   - - ! '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 1.8.24
+rubygems_version: 2.0.3
 signing_key: 
-specification_version: 3
+specification_version: 4
 summary: Read and manipulate CD-HIT clusters
 test_files: []
+has_rdoc: