cagnut_gatk 0.3.0

data/lib/cagnut_gatk/functions/variant_filtration.rb

@@ -0,0 +1,88 @@
+module CagnutGatk
+  class VariantFiltration
+    extend Forwardable
+
+    def_delegators :'Cagnut::Configuration.base', :sample_name, :jobs_dir, :prefix_name,
+                   :ref_fasta, :dodebug, :java_path
+    def_delegators :'CagnutGatk.config', :variant_filtration_params
+
+    def initialize opts = {}
+      @order = sprintf '%02i', opts[:order]
+      @input = opts[:input].nil? ? "#{opts[:dirs][:input]}/#{sample_name}.vcf" : opts[:input]
+      @job_name = "#{prefix_name}_snpFiltr_#{sample_name}"
+      @output = "#{opts[:dirs][:output]}/#{sample_name}_filtered.vcf"
+    end
+
+    def run previous_job_id = nil
+      puts "Submitting #{sample_name} Jobs: variant (SNPs, INDELs) -filtration"
+      script_name = generate_script
+      ::Cagnut::JobManage.submit script_name, @job_name, cluster_options(previous_job_id)
+      [@job_name, @output]
+    end
+
+    def cluster_options previous_job_id = nil
+      {
+        previous_job_id: previous_job_id,
+        adjust_memory: ['h_stack=256M', 'h_vmem=10G'],
+        tools: ['gatk', 'variant_filtration']
+      }
+    end
+
+    def params_combination
+      @params_combination_hash ||= {
+        'java' => modified_java_array,
+        'params' => variant_filtration_options
+      }
+    end
+
+    def variant_filtration_options
+      array = variant_filtration_params['params'].dup
+      array << "-T VariantFiltration"
+      array << "-R #{ref_fasta}"
+      array << "--variant:VCF #{@input}"
+      array << "-o #{@output}"
+      array.uniq
+    end
+
+    def modified_java_array
+      array = variant_filtration_params['java'].dup
+      array.unshift(java_path).uniq
+    end
+
+    def generate_script
+      script_name = "#{@order}_gatk_variant_filtration"
+      file = File.join jobs_dir, "#{script_name}.sh"
+      File.open(file, 'w') do |f|
+        f.puts <<-BASH.strip_heredoc
+          #!/bin/bash
+
+          cd "#{jobs_dir}/../"
+          echo "#{script_name} is starting at $(date +%Y%m%d%H%M%S)" >> "#{jobs_dir}/finished_jobs"
+          if [ ! -s "#{@input}.idx" ]; then
+           echo "Incomplete VCF:" #{@input}
+           exit 100
+          fi
+
+          #{params_combination['java'].join("\s")} \\
+            #{params_combination['params'].join(" \\\n            ")} \\
+            #{::Cagnut::JobManage.run_local}
+
+          EXITSTATUS=$?
+
+          #if [ ! -s "#{@output}" ]; then exit 100;fi;
+
+          if [ ! -s "#{@output}.idx" ]
+          then
+            echo "vcf incomplete!"
+            exit 100;
+          fi
+          echo "#{script_name} is finished at $(date +%Y%m%d%H%M%S)" >> "#{jobs_dir}/finished_jobs"
+
+          exit $EXITSTATUS
+        BASH
+      end
+      File.chmod(0700, file)
+      script_name
+    end
+  end
+end

data/lib/cagnut_gatk/util.rb

@@ -0,0 +1,118 @@
+module CagnutGatk
+  class Util
+    attr_accessor :gatk, :config
+
+    def initialize config
+      @config = config
+      @gatk = CagnutGatk::Base.new
+    end
+
+    def count_read dirs, order=1, previous_job_id=nil, filename=nil
+      count_read_params(filename).each do |option|
+        gatk.count_read dirs, order, previous_job_id, option
+      end
+      order+1
+    end
+
+    def recal dirs, order=1, previous_job_id=nil, filename = nil
+      # if (@config['cagnut']['ref_fasta'].scan 'hg').empty?
+      #   # filename = "s_#{@config['line']}_merged_markdup.bam"
+      #   filename = "#{line}_realn.bam"
+      # else
+      #   previous_job_id, target_interval = realigner_target_creator previous_job_id, filename
+      #   previous_job_id, filename = indel_realigner previous_job_id, filename, target_interval
+      # end
+
+      previous_job_id, interval_list, order = realigner_target_creator dirs, order, previous_job_id, filename
+      previous_job_id, filename, order = indel_realigner dirs, order, previous_job_id, filename, interval_list
+      previous_job_id, order = base_recalibrator dirs, order, previous_job_id, filename
+      previous_job_id, filename = print_reads dirs, order, previous_job_id, filename
+      [previous_job_id, previous_job_id, order+1]
+    end
+
+    def realigner_target_creator dirs, order=1, previous_job_id=nil, filename=nil
+      job_name, interval_list = @gatk.realigner_target_creator dirs, order, previous_job_id, filename
+      [job_name, interval_list, order+1]
+    end
+
+    def indel_realigner dirs, order=1, previous_job_id=nil, filename=nil, interval_list=nil
+      job_name, filename = @gatk.indel_realigner dirs, order, previous_job_id, filename, interval_list
+      [job_name, filename, order+1]
+    end
+
+    def base_recalibrator dirs, order=1, previous_job_id=nil, filename=nil
+      before_and_after_generated_bqsr_file.each do |option|
+        previous_job_id = @gatk.base_recalibrator dirs, order, previous_job_id, filename, option
+      end
+      [previous_job_id, order+1]
+    end
+
+    def analyze_covariates dirs, order=1, previous_job_id=nil, filename=nil
+      job_name = @gatk.analyze_covariates dirs, order, previous_job_id, filename
+      [job_name, order+1]
+    end
+
+    def print_reads dirs, order=1, previous_job_id=nil, file_name=nil
+      job_name, filename = @gatk.print_reads dirs, order, previous_job_id, file_name
+      [job_name, filename, order+1]
+    end
+
+    def depth_of_coverage dirs, order=1, previous_job_id=nil, filename=nil
+      depth_of_coverage_params.each do |option|
+        @gatk.depth_of_coverage dirs, order, previous_job_id, filename, option
+      end
+      order+1
+    end
+
+    def haplotype_caller dirs, order=1, previous_job_id=nil, file_name=nil
+      job_name, filename = @gatk.haplotype_caller dirs, order, previous_job_id, file_name
+      [job_name, filename, order+1]
+    end
+
+    def unified_genotyper dirs, order=1, previous_job_id=nil, file_name=nil
+      job_name, filename = @gatk.unified_genotyper dirs, order, previous_job_id, file_name
+      [job_name, filename, order+1]
+    end
+
+    def snpcal dirs, order=1, previous_job_id=nil, filename = nil
+      previous_job_id, filename, order = variant_filtration dirs, order, previous_job_id, filename
+      variant_eval dirs, order, previous_job_id, filename
+    end
+
+    def variant_filtration dirs, order=1, previous_job_id=nil, filename=nil
+      job_name, filename = gatk.variant_filtration dirs, order, previous_job_id, filename
+      [job_name, filename, order+1]
+    end
+
+    def variant_eval dirs, order=1, previous_job_id=nil, filename=nil
+      job_name, filename = gatk.variant_eval dirs, order, previous_job_id, filename
+      [job_name, filename, order+1]
+    end
+
+    private
+
+    def before_and_after_generated_bqsr_file
+      [{ has_bqsr_file: false }, { has_bqsr_file: true }]
+    end
+
+    def count_read_params filename
+      ary = [{ input: filename }]
+      if @config['refs']['targets_file']
+        ary <<
+          { input: filename, target: @config['refs']['targets_file'] }
+      end
+      ary
+    end
+
+    def depth_of_coverage_params
+      ary = [{ suffix: 'genome' }]
+      if @config['refs']['targets_file']
+        ary << { suffix: 'target', target: @config['refs']['targets_file'] }
+      end
+      if @config['refs']['target_flanks_file']
+        ary << { suffix: 'flank', target: @config['refs']['target_flanks_file'] }
+      end
+      ary
+    end
+  end
+end

data/lib/cagnut_gatk/version.rb

@@ -0,0 +1,3 @@
+module CagnutGatk
+  VERSION = "0.3.0"
+end

data/lib/cagnut_gatk.rb

@@ -0,0 +1,16 @@
+require "cagnut_gatk/version"
+
+module CagnutGatk
+  class << self
+    def config
+      @config ||= begin
+        CagnutGatk::Configuration.load(Cagnut::Configuration.config, Cagnut::Configuration.params['gatk'])
+        CagnutGatk::Configuration.instance
+      end
+    end
+  end
+end
+
+require 'cagnut_gatk/configuration'
+require 'cagnut_gatk/base'
+require 'cagnut_gatk/util'

metadata

@@ -0,0 +1,135 @@
+--- !ruby/object:Gem::Specification
+name: cagnut_gatk
+version: !ruby/object:Gem::Version
+  version: 0.3.0
+platform: ruby
+authors:
+- Shi-Gang Wang
+- Tse-Ching Ho
+autorequire: 
+bindir: exe
+cert_chain: []
+date: 2016-11-01 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: cagnut_core
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.12'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.12'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+description: Cagnut Gatk tools
+email:
+- seanwang@goldenio.com
+- tsechingho@goldenio.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- ".rspec"
+- ".ruby-version"
+- ".travis.yml"
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- bin/console
+- bin/setup
+- cagnut_gatk.gemspec
+- lib/cagnut_gatk.rb
+- lib/cagnut_gatk/base.rb
+- lib/cagnut_gatk/check_tools.rb
+- lib/cagnut_gatk/configuration.rb
+- lib/cagnut_gatk/functions/analyze_covariates.rb
+- lib/cagnut_gatk/functions/base_recalibrator.rb
+- lib/cagnut_gatk/functions/count_read.rb
+- lib/cagnut_gatk/functions/depth_of_coverage.rb
+- lib/cagnut_gatk/functions/haplotype_caller.rb
+- lib/cagnut_gatk/functions/indel_realigner.rb
+- lib/cagnut_gatk/functions/print_reads.rb
+- lib/cagnut_gatk/functions/realigner_target_creator.rb
+- lib/cagnut_gatk/functions/templates/analyze_covariates.sh
+- lib/cagnut_gatk/functions/templates/base_recalibrator.sh
+- lib/cagnut_gatk/functions/templates/count_read.sh
+- lib/cagnut_gatk/functions/templates/depth_of_coverage.sh
+- lib/cagnut_gatk/functions/templates/print_reads.sh
+- lib/cagnut_gatk/functions/unified_genotyper.rb
+- lib/cagnut_gatk/functions/variant_eval.rb
+- lib/cagnut_gatk/functions/variant_filtration.rb
+- lib/cagnut_gatk/util.rb
+- lib/cagnut_gatk/version.rb
+homepage: https://github.com/CAGNUT/cagnut_gatk
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.5.1
+signing_key: 
+specification_version: 4
+summary: Cagnut Gatk tools
+test_files: []