rbbt-study 0.2.30 → 0.2.31

data/lib/rbbt/entity/study/cnv/samples.rb

@@ -1,54 +0,0 @@
-module Sample
-  property :cnvs => :array2single do
-    study.has_cnv? ? study.cnv_cohort : {}
-  end
-
-  property :has_cnv? => :array2single do
-    study.has_cnv? ? cnvs.values_at(*self).collect{|cnvs| not cnvs.nil?} : [false] * self.length
-  end
-
-  property :gene_CN => :single do
-    gene_CN = {}
-    cnvs.variation.zip(cnvs.genes).each do |var, genes|
-      next if genes.empty?
-      genes = Annotated.purge genes 
-      case var
-      when "loss"
-        genes.each{|gene| gene_CN[gene] = "Lost"}
-      when "gain"
-        genes.each{|gene| gene_CN[gene] = "Gained"}
-      end
-    end
-    gene_CN
-  end
-  persist :gene_CN
-  
-  property :gained_cnvs => :single do
-    return nil if cnvs.nil?
-    return [] if cnvs.empty?
-    cnvs.select_by(:gain?)
-  end
- 
-  property :lost_cnvs => :single do
-    return nil if cnvs.nil?
-    return [] if cnvs.empty?
-    cnvs.select_by(:loss?)
-  end
-  
-  property :gained_genes => :single do
-    Gene.setup(gene_CN.select{|g,v| v == "Gained"}.collect{|g,v| g}, "Ensembl Gene ID", self.study.organism)
-  end
-
-  property :lost_genes => :single do
-    Gene.setup(gene_CN.select{|g,v| v == "Lost"}.collect{|g,v| g}, "Ensembl Gene ID", self.study.organism)
-  end
-
-  property :cnv_genes => :single do
-    return nil if lost_genes.nil? or gained_genes.nil?
-    organism = study.metadata[:organism]
-    Gene.setup((lost_genes + gained_genes).uniq, "Ensembl Gene ID", lost_genes.organism)
-  end
-
-end
-
-

data/lib/rbbt/entity/study/enrichment.rb

@@ -1,418 +0,0 @@
-#{{{ SAMPLE ENRICHMENT
-dep do |jobname, inputs| job(inputs[:mutation_subset] || :relevant_mutations, jobname, inputs) end
-input :database, :string, "Database code"
-input :mutation_subset, :select, "Mutation subset to use", :relevant_mutations
-input :baseline, :select, "Type of baseline to use", :bases, :select_options => [:pathway_base_counts, :pathway_gene_counts]
-input :permutations, :integer, "Number of permutations in test", 10000
-input :fdr, :boolean, "BH FDR corrections", true
-input :masked_genes, :array, "Ensembl Gene ID list of genes to mask", []
-input :organism, :string, "Organism code", metadata[:organism]
-task :sample_pathway_enrichment => :tsv do |database,mutation_subset,baseline,permutations,fdr,masked_genes,organism|
-
-  mutations = step(mutation_subset).load
-
-  mutation_tsv = TSV.setup({}, :key_field => "Genomic Mutation", :fields => ["Sample"], :type => :flat)
-
-  study.cohort.each do |genotype|
-    sample = genotype.jobname
-    genotype.each do |mutation|
-      next unless mutations.include? mutation
-      mutation_tsv[mutation] ||= []
-      mutation_tsv[mutation] << sample
-    end
-  end
-
-  job = MutationEnrichment.job(:sample_pathway_enrichment, study, 
-                               :mutations => mutation_tsv, :database => database, :baseline => baseline, :fdr => fdr, 
-                               :masked_genes => masked_genes, :organism => organism, :permutations => permutations)
-
-  res = job.run
-  set_info :total_covered, job.info[:total_covered]
-  set_info :covered_mutations, job.info[:covered_mutations]
-  res
-end
-
-#{{{ METAGENOTYPE ENRICHMENT
-dep do |jobname, inputs| job(inputs[:mutation_subset] || :relevant_mutations, jobname, inputs) end
-input :database, :string, "Database code"
-input :mutation_subset, :select, "Mutation subset to use", :relevant_mutations
-input :baseline, :select, "Type of baseline to use", :pathway_base_counts, :select_options => [:pathway_base_counts, :pathway_gene_counts]
-input :fdr, :boolean, "BH FDR corrections", true
-input :masked_genes, :array, "Ensembl Gene ID list of genes to mask", []
-input :organism, :string, "Organism code", metadata[:organism]
-task :mutation_pathway_enrichment => :tsv do |database,mutation_subset,baseline,fdr,masked_genes,organism|
-
-  mutations = step(mutation_subset).load
-
-  job = MutationEnrichment.job(:mutation_pathway_enrichment, study, 
-                               :mutations => mutations, :database => database, :baseline => baseline, :fdr => fdr, 
-                               :masked_genes => masked_genes, :organism => organism)
-  res = job.run
-  set_info :total_covered, job.info[:total_covered]
-  set_info :covered_mutations, job.info[:covered_mutations]
-  res
-end
- 
-
-
-
-
-
-################################################################
-#{{{ OLD
-################################################################
-
-#{{{ BASE AND GENE COUNTS
-
-#input :masked_genes, :array, "Ensembl Gene ID list of genes to mask", []
-#input :organism, :string, "Organism code", metadata[:organism]
-#task :pathway_base_counts => :tsv do |masked_genes, organism|
-#  database = clean_name
-#  log :loading_genes, "Loading genes from #{ database } #{ organism }"
-#  case database
-#  when 'kegg'
-#    tsv = KEGG.gene_pathway.tsv :key_field => "KEGG Pathway ID", :fields => ["KEGG Gene ID"], :type => :flat, :persist => true, :merge => true
-#    total_genes = Gene.setup(tsv.values.compact.flatten.uniq, "KEGG Gene ID", organism).ensembl.compact
-#  when 'go', 'go_bp'
-#    tsv = Organism.gene_go_bp(organism).tsv :key_field => "GO ID", :fields => ["Ensembl Gene ID"], :type => :flat, :persist => true, :merge => true
-#    total_genes = Gene.setup(tsv.values.compact.flatten.uniq, "Ensembl Gene ID", organism).ensembl.compact
-#  when 'pfam'
-#    tsv = Organism.gene_pfam(organism).tsv :key_field => "Pfam Domain", :fields => ["Ensembl Gene ID"], :type => :flat, :persist => true, :merge => true
-#    total_genes = Gene.setup(tsv.values.compact.flatten.uniq, "Ensembl Gene ID", organism).ensembl.compact
-#  end
-#
-#  tsv.namespace = organism
-#
-#  counts = TSV.setup({}, :key_field => tsv.key_field, :fields => ["Bases"], :type => :single, :cast => :to_i, :namespace => organism)
-#
-#  log :processing_database, "Processing database #{database}"
-#  tsv.with_monitor do
-#    tsv.through do |pathway, genes|
-#      next if genes.nil? or genes.empty? 
-#      size = Gene.gene_list_exon_bases(genes.ensembl.compact.remove(masked_genes))
-#      counts[pathway] = size
-#    end
-#  end
-#
-#  log :computing_exome_size, "Computing number of exome bases covered by pathway annotations"
-#  total_size = Gene.gene_list_exon_bases(total_genes.remove(masked_genes))
-#
-#  set_info :total_size, total_size
-#  set_info :total_gene_list, total_genes.remove(masked_genes)
-#
-#  counts
-#end
-#
-#input :masked_genes, :array, "Ensembl Gene ID list of genes to mask", []
-#input :organism, :string, "Organism code", metadata[:organism]
-#task :pathway_gene_counts => :tsv do |masked_genes,organism|
-#  database = clean_name
-#  case database.to_s
-#  when 'kegg'
-#    tsv = KEGG.gene_pathway.tsv :key_field => "KEGG Pathway ID", :fields => ["KEGG Gene ID"], :type => :flat, :persist => true, :merge => true
-#    total_genes = Gene.setup(tsv.values.compact.flatten.uniq, "KEGG Gene ID", organism).ensembl.compact
-#  when 'go', 'go_bp'
-#    tsv = Organism.gene_go_bp(organism).tsv :key_field => "GO ID", :fields => ["Ensembl Gene ID"], :type => :flat, :persist => true, :merge => true
-#    total_genes = Gene.setup(tsv.values.compact.flatten.uniq, "Ensembl Gene ID", organism).ensembl.compact
-#  when 'pfam'
-#    tsv = Organism.gene_pfam(organism).tsv :key_field => "Pfam Domain", :fields => ["Ensembl Gene ID"], :type => :flat, :persist => true, :merge => true
-#    total_genes = Gene.setup(tsv.values.compact.flatten.uniq, "Ensembl Gene ID", organism).ensembl.compact
-#  end
-#
-#  counts = TSV.setup({}, :key_field => tsv.key_field, :fields => ["Genes"], :type => :single, :cast => :to_i, :namespace => organism)
-#
-#  tsv.through do |pathway, genes|
-#    next if genes.nil? or genes.empty? 
-#    genes = genes.ensembl.remove(masked_genes)
-#    num = genes.length
-#    counts[pathway] = num
-#  end
-#
-#  set_info :total_genes, total_genes.remove(masked_genes).length
-#  set_info :total_gene_list, total_genes.remove(masked_genes)
-#
-#  counts
-#end
-
-
-#
-#
-#dep do |jobname, inputs| job(inputs[:baseline], inputs[:database].to_s, inputs) end
-#dep do |jobname, inputs| job(inputs[:mutation_subset] || :relevant_mutations, jobname, inputs) end
-#dep :affected_samples_per_pathway
-#input :database, :string
-#input :mutation_subset, :select, "Mutation subset to use", :relevant_mutations
-#input :baseline, :select, "Type of baseline to use", :bases, :select_options => [:pathway_base_counts, :pathway_gene_counts]
-#input :permutations, :integer, "Number of permutations in test", 10000
-#input :fdr, :boolean, "BH FDR corrections", true
-#input :masked_genes, :array, "Ensembl Gene ID list of genes to mask", []
-#input :organism, :string, "Organism code", metadata[:organism]
-#task :sample_pathway_enrichment_old => :tsv do |database,mutation_subset,baseline,permutations,fdr,masked_genes,organism|
-#  pathway_counts                         = step(baseline).load
-#  pathway_counts.unnamed                 = true
-#  total_covered                          = step(baseline).info[:total_size] || step(baseline).info[:total_genes]
-#  total_pathway_genes_list               = step(baseline).info[:total_gene_list]
-#  mutations                              = step(mutation_subset).load
-#  affected_samples_per_pathway           = step(:affected_samples_per_pathway).load
-#  affected_samples_per_pathway.namespace = organism
-#
-#  affected_genes = mutations.genes.compact.flatten.uniq
-#
-#  case database.to_s
-#  when 'kegg'
-#    database_tsv = KEGG.gene_pathway.tsv :key_field => 'KEGG Pathway ID', :fields => ["KEGG Gene ID"], :type => :flat, :persist => true, :unnamed => false, :merge => true
-#  when 'go', 'go_bp'
-#    database_tsv = Organism.gene_go_bp(organism).tsv :key_field => "GO ID", :fields => ["Ensembl Gene ID"], :type => :flat, :persist => true, :unnamed => false, :merge => true
-#  when 'pfam'
-#    database_tsv = Organism.gene_pfam(organism).tsv :key_field => "Pfam Domain", :fields => ["Ensembl Gene ID"], :type => :flat, :persist => true, :unnamed => false, :merge => true
-#  end
-#
-#  covered_genes_per_samples = {}
-#  samples = []
-#  study.cohort.each do |genotype|
-#    samples << genotype.jobname
-#    covered_genes_per_samples[genotype.jobname] = genotype.subset(mutations).genes.compact.flatten.subset(total_pathway_genes_list)
-#  end
-#
-#  sample_mutation_tokens = []
-#  samples.collect{|sample| study.cohort[sample].subset(mutations).genes.select{|l| not l.nil? and (l & total_pathway_genes_list).any? }.length.times{ sample_mutation_tokens << sample } }
-#
-#  mutation_genes = Misc.process_to_hash(mutations){|list| list.genes}
-#  covered_mutations = mutations.select{|mutation|(mutation_genes[mutation] & total_pathway_genes_list).any? }.length
-#
-#  pathways = pathway_counts.keys
-#
-#  pathway_expected_counts = {}
-#  pathway_counts.with_monitor :desc => "Pathway gene counts" do
-#    pathway_counts.through do |pathway, count|
-#      next unless affected_samples_per_pathway.include?(pathway) and affected_samples_per_pathway[pathway].any?
-#      ratio = count.to_f / total_covered
-#      num_token_list = RSRuby.instance.rbinom(permutations, sample_mutation_tokens.length, ratio)
-#      pathway_expected_counts[pathway] = num_token_list.collect{|num_tokens|
-#        Misc.sample(sample_mutation_tokens, num_tokens.to_i).uniq.length
-#      }
-#    end
-#  end
-#
-#  tsv = TSV.setup({}, :key_field => affected_samples_per_pathway.key_field, :fields => ["Sample", "Matches", "Expected", "Ratio", "Pathway total", "p-value", "Ensembl Gene ID"], :namespace => organism, :type => :double)
-#  affected_samples_per_pathway.through do |pathway, samples|
-#    next unless samples.any?
-#    next unless pathway_expected_counts.include? pathway
-#    pathway_genes = database_tsv[pathway].ensembl
-#    samples = samples.uniq.select{|sample| (covered_genes_per_samples[sample] & pathway_genes).any?}
-#    count = samples.length
-#    expected = Misc.mean(pathway_expected_counts[pathway]).floor
-#    pvalue = pathway_expected_counts[pathway].select{|exp_c| exp_c > count}.length.to_f / permutations
-#    tsv[pathway] = [samples.sort, [count], [expected], [count.to_f / expected], [pathway_counts[pathway]], [pvalue], pathway_genes.subset(affected_genes)]
-#  end
-#
-#  FDR.adjust_hash! tsv, 5 if fdr
-#
-#  set_info :covered_mutations, covered_mutations
-#  set_info :total_covered, total_covered
-#
-#  tsv
-#end
-#
-#
-#
-#
-#dep do |jobname, inputs| job(inputs[:baseline], inputs[:database].to_s, inputs) end
-#dep do |jobname, inputs| job(inputs[:mutation_subset] || :relevant_mutations, jobname, inputs) end
-#input :database, :string
-#input :mutation_subset, :select, "Mutation subset to use", :relevant_mutations
-#input :baseline, :select, "Type of baseline to use", :bases, :select_options => [:pathway_base_counts, :pathway_gene_counts]
-#input :fdr, :boolean, "BH FDR corrections", true
-#input :masked_genes, :array, "Ensembl Gene ID list of genes to mask", []
-#input :organism, :string, "Organism code", metadata[:organism]
-#task :mutation_pathway_enrichment_old => :tsv do |database,mutation_subset,baseline,fdr,masked_genes,organism|
-#  counts        = step(baseline).load
-#  total_covered = step(baseline).info[:total_size] || step(baseline).info[:total_genes]
-#  mutations     = step(mutation_subset).load
-#
-#  affected_genes = mutations.genes.compact.flatten.uniq
-#
-#  # Get database tsv and native ids
-#
-#  case database
-#  when 'kegg'
-#    database_tsv = KEGG.gene_pathway.tsv :key_field => 'KEGG Gene ID', :fields => ["KEGG Pathway ID"], :type => :flat, :persist => true, :unnamed => true, :merge => true
-#    affected_genes_db = affected_genes.to_kegg
-#    all_db_genes = Gene.setup(database_tsv.keys, "KEGG Gene ID", organism).ensembl.uniq
-#  when 'go', 'go_bp'
-#    database_tsv = Organism.gene_go_bp(organism).tsv :key_field => "Ensembl Gene ID", :fields => ["GO ID"], :type => :flat, :persist => true, :unnamed => true, :merge => true
-#    affected_genes_db = affected_genes
-#    all_db_genes = Gene.setup(database_tsv.keys, "KEGG Gene ID", organism).uniq
-#  when 'pfam'
-#    database_tsv = Organism.gene_pfam(organism).tsv :key_field => "Ensembl Gene ID", :fields => ["Pfam Domain"], :type => :flat, :persist => true, :unnamed => true, :merge => true
-#    affected_genes_db = affected_genes
-#    all_db_genes = Gene.setup(database_tsv.keys, "KEGG Gene ID", organism).uniq
-#  end
-#
-#  affected_genes = affected_genes.remove(masked_genes)
-#  all_db_genes = all_db_genes.remove(masked_genes)
-#
-#  # Annotate each pathway with the affected genes that are involved in it
-#  
-#  affected_genes_per_pathway = {}
-#  affected_genes_db.zip(affected_genes).each do |gene_db,gene|
-#    next if gene_db.nil?
-#    pathways = database_tsv[gene_db]
-#    next if pathways.nil?
-#    pathways.uniq.each do |pathway|
-#      affected_genes_per_pathway[pathway] ||= []
-#      affected_genes_per_pathway[pathway] << gene
-#    end
-#  end
-#
-#  pvalues = TSV.setup({}, :key_field => database_tsv.fields.first, :fields => ["Matches", "Pathway total", "p-value", "Ensembl Gene ID"], :namespace => organism, :type => :double)
-#  mutation_genes = Misc.process_to_hash(mutations){|list| list.genes}
-#  covered_mutations = mutations.select{|mutation|(mutation_genes[mutation] & all_db_genes).any? }.length
-#
-#  affected_genes_per_pathway.each do |pathway, genes|
-#    pathway_total = counts[pathway]
-#    matches = mutations.select{|mutation| (mutation_genes[mutation] & genes).any? }.length
-#    pvalue = RSRuby.instance.binom_test(matches, covered_mutations, pathway_total.to_f / total_covered.to_f, "greater")["p.value"]
-#
-#    pvalues[pathway] = [[matches], [pathway_total], [pvalue], affected_genes.subset(genes).uniq.sort_by{|g| g.name || g}]
-#  end
-#
-#  FDR.adjust_hash! pvalues, 2 if fdr
-#
-#  set_info :covered_mutations, covered_mutations
-#  set_info :total_covered, total_covered
-#
-#  pvalues
-#end
-#
-#
-#dep do |jobname, inputs| job(:pathway_base_counts, inputs[:database].to_s, inputs) end
-#dep do |jobname, inputs| job(inputs[:mutation_subset] || :relevant_mutations, jobname, inputs) end
-#dep :affected_genes
-#input :database, :string
-#input :fdr, :boolean, "BH FDR corrections", true
-#input :mutation_subset, :select, "Mutation subset to use", :relevant_mutations
-#
-##{{{ RATIOS
-#
-#dep :affected_samples_per_pathway
-#dep :affected_genes
-#dep do |name, inputs| 
-#  database = inputs[:database]
-#  if inputs[:type] == :genes
-#    job(:pathway_base_counts, database, inputs)
-#  else
-#    job(:pathway_gene_counts, database, inputs)
-#  end
-#end
-#task :pathway_sample_ratios => :tsv do 
-#  num_samples = study.cohort.length
-#
-#  affected_samples_per_pathway = step(:affected_samples_per_pathway).load
-#  affected_genes = step(:affected_genes).load
-#
-#  ratios = TSV.setup({}, :key_field => sample_pathway_probability.key_field, :fields => ["Num Samples", "Expected", "Ratio", "Ensembl Gene ID"], :namespace => organism)
-#
-#  pathways.through do |pathway, probability|
-#    next unless affected_samples_per_pathway.include?(pathway) and affected_samples_per_pathway[pathway].length > 1
-#    affected_samples = (affected_samples_per_pathway[pathway] || []).length
-#    ratios[pathway] = [affected_samples, probability * num_samples, affected_samples.to_f / (probability * num_samples), pathway.genes.ensembl.subset(affected_genes)]
-#  end
-#
-#  ratios.namespace = organism
-#
-#  ratios
-#end
-#
-
-#input :database, :string
-#dep :damaging_mutations
-#dep :affected_genes
-#task :pathway_mutation_ratios => :tsv do |database|
-#  damaging_mutations = step(:damaging_mutations).load
-#
-#  affected_genes = step(:affected_genes).load
-#  affected_genes.organism = organism
-#
-#  pathways = case database
-#             when 'go'
-#               affected_genes.go_terms
-#             when 'go_bp'
-#               affected_genes.go_bp_terms
-#             when 'go_cc'
-#               affected_genes.go_cc_terms
-#             when 'go_mf'
-#               affected_genes.go_mf_terms
-#             when 'pfam'
-#               affected_genes.pfam_domains
-#             else
-#               affected_genes.send(database + '_pathways')
-#             end.compact.flatten.uniq
-#
-#  key_field = nil
-#  pathway_genes = Misc.process_to_hash(pathways) do |pathways|
-#    pathways.uniq.collect do |pathway| 
-#      case database
-#      when 'kegg'
-#        KeggPathway.setup(pathway, organism)
-#        key_field = "KEGG Pathway ID"
-#        Gene.setup(pathway.genes, "KEGG Gene ID", organism).ensembl
-#      when 'go', 'go_bp', 'go_mf', 'go_cc'
-#        GOTerm.setup(pathway, organism)
-#        key_field = "GO ID"
-#        Gene.setup(pathway.genes, "Ensembl Gene ID", organism).ensembl
-#      when 'pfam'
-#        PfamDomain.setup(pathway, organism)
-#        key_field = "Pfam Domain"
-#        Gene.setup(pathway.genes, "Ensembl Gene ID", organism).ensembl
-#      end
-#    end
-#  end
-#
-#  pathway_mutation_ratios = TSV.setup({}, :key_field => key_field, 
-#    :fields => ["Mutations per MB", "# Mutations", "# Damaging Mut.",  "# Genes", "# Bases", "Ensembl Gene ID"], :type => :double)
-#
-#
-#  pathway_sizes = Misc.process_to_hash(pathways) do |pathways| 
-#    pathways.collect{|pathway| Gene.gene_list_exon_bases(pathway_genes[pathway])}
-#  end
-#
-#  pathways_for_mutations = Misc.process_to_hash(study.cohort.flatten){|all_mutations| Gene.setup(study.cohort.collect{|genotype| genotype.genes }.flatten(1), "Ensembl Gene ID", organism).collect{|genes| 
-#    genes.nil? ? [] : genes.collect{|gene| 
-#      case database
-#      when 'go'
-#        gene.go_terms
-#      when 'go_bp'
-#        gene.go_bp_terms
-#      when 'go_cc'
-#        gene.go_cc_terms
-#      when 'go_mf'
-#        gene.go_mf_terms
-#      when 'pfam'
-#        gene.pfam_domains
-#      else
-#        affected_genes.send(database + '_pathways')
-#      end 
-#    }.flatten}
-#  }
-#
-#  mutations_in_pathway = Misc.process_to_hash(pathways) do |pathways|
-#    pathways.collect do |pathway|
-#      GenomicMutation.setup(pathways_for_mutations.select{|mut,pths| pths and pths.include? pathway}.collect{|mut, pths| mut}, "Pathway mutations #{pathway} in #{study}", organism, watson)
-#    end
-#  end
-#
-#  pathways.each do |pathway| 
-#    pathway_mutations = mutations_in_pathway[pathway]
-#    next if pathway_mutations.one?
-#    pathway_score = pathway_mutations.length.to_f / pathway_sizes[pathway]
-#    genes = pathway_mutations.genes.compact.flatten.subset(affected_genes).subset(pathway_genes[pathway])
-#
-#    pathway_mutation_ratios[pathway] = [["%.5g" % (pathway_score * 1_000_000)], [pathway_mutations.length], [pathway_mutations.subset(damaging_mutations).length], [pathway_genes[pathway].length], [pathway_sizes[pathway]],  genes]
-#  end
-#
-#  pathway_mutation_ratios.namespace = organism
-#
-#  pathway_mutation_ratios
-#end
-#