rbbt-study 0.2.30 → 0.2.31

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

@@ -1,140 +0,0 @@
-input :cutoff, :integer, "Pixels of image", 2
-input :size, :integer, "Pixels of image", 14
-task :gene_mutation_plot => :binary do |cutoff, size|
-  png_file = file(study + ".png")
-  FileUtils.mkdir_p File.dirname png_file unless File.exists? File.dirname png_file
-  study.R "
-library(ggplot2)
-library(plyr)
-library(reshape)
-
-layer.mutations = rbbt.SE.plot.mutations('#{study}', cutoff=#{cutoff});
-p <- ggplot() + layer.mutations 
-
-p <- p + opts(axis.text.x=theme_text(angle=90), panel.background = theme_rect(fill='white', colour='steelblue'))
-
-ggsave(p, filename='#{png_file}', height=#{size}, width=#{size});
-"
-  Open.read(png_file, :mode => 'rb')
-end
-
-
-input :database, :string, "Database code", :kegg
-input :size, :integer, "Pixels of image", 14
-task :pathway_mutation_plot => :binary do |database,size|
-  png_file = file(study + ".png")
-  FileUtils.mkdir_p File.dirname png_file unless File.exists? File.dirname png_file
-  study.R "
-library(ggplot2)
-library(plyr)
-library(reshape)
-
-
-study = '#{study}'
-# Sample mutations
-sample.mutated.genes = rbbt.SE.sample.mutated.genes(study)
-sample.mutated.genes$Sample = rownames(sample.mutated.genes)
-
-# Pathway enrichment
-pathway.enrichment = rbbt.ruby.substitutions(
-    \"
-    require 'rbbt/workflow'
-    require 'rbbt/entity'
-    require 'rbbt/entity/gene'
-    require 'rbbt/sources/pfam'
-    require 'rbbt/sources/kegg'
-    require 'rbbt/sources/go'
-
-    YAML::ENGINE.yamler = 'syck' if defined? YAML::ENGINE and YAML::ENGINE.respond_to? :yamler
-
-    Workflow.require_workflow 'StudyExplorer'
-
-    study = Study.setup('STUDY')
-
-    pathways = study.job(:mutation_pathway_enrichment, study, :baseline => :pathway_base_counts, :database => '#{database}', :fdr => false).run.select('p-value'){|pvalue| pvalue = pvalue.first.to_f if Array === pvalue; pvalue < 0.2}
-    pathways.add_field 'Name' do |pathway, values|
-        [pathway.name]
-    end
-
-    pathways.add_field 'Gene' do |pathway, values|
-        values['Ensembl Gene ID'].name
-    end
-
-    pathways = pathways.select('Name'){|name| name.first.to_s !~ /cancer|olfactory|glioma|melanoma|malaria|leukemia|carcinoma|sarcoma/i}
-
-    \", substitutions=list(STUDY=study));
-
-
-# Sample pathway mutations
-find.mutated.pathways.for.sample <- function(x, pathway.info){
-    all.genes = names(x);
-    genes = all.genes[x==TRUE];
-    ddply(pathway.info, 'Name', function(x){pathway.genes = unlist(strsplit(x$Gene, '\\\\|')); if (length(intersect(genes, pathway.genes)) > 0){TRUE}else{FALSE}})
-}
-sample.pathway.mutations = ddply(sample.mutated.genes, 'Sample', find.mutated.pathways.for.sample, pathway.info = pathway.enrichment)
-names(sample.pathway.mutations) = c('Sample', 'Pathway', 'Mutated')
-
-p <- ggplot(sample.pathway.mutations) + geom_tile(aes(x=Sample, y=Pathway, alpha=Mutated))
-
-p <- rbbt.SE.plot.sort.by.pathway.mutations(p)
-
-
-# Mark repeated genes
-
-
-d = p$data
-d$Exclusive = FALSE
-
-pathway.genes = list();
-for(pathway in levels(d$Pathway)){
-   pathway.genes[pathway] = strsplit(pathway.enrichment[pathway.enrichment[,'Name'] == pathway, 'Gene'], '\\\\|')
-}
-
-find.exclusive.pathway.genes <- function(data, pathways){
-  found.genes = c();
-  exclusive.pathway.genes = list();
-  sample = as.character(unique(data$Sample));
-  for(pathway in pathways){
-     current.pathway.genes = pathway.genes[[pathway]];
-     sample.genes = names(sample.mutated.genes)[sample.mutated.genes[sample,] == TRUE]
-     sample.pathway.genes = intersect(current.pathway.genes, sample.genes);
-     exclusive.genes = setdiff(sample.pathway.genes, found.genes);
-     found.genes = c(found.genes, exclusive.genes)
-     exclusive.pathway.genes[[pathway]] = exclusive.genes
-  }
-
-  return(exclusive.pathway.genes);
-}
-
-exclusive.pathway.genes = dlply(d, 'Sample', find.exclusive.pathway.genes, pathways = levels(d$Pathway)) 
-
-for( sample in names(exclusive.pathway.genes)){
-     pathway.exclusive.genes = exclusive.pathway.genes[[sample]];
-     for( pathway in names(pathway.exclusive.genes)){
-        if (length(pathway.exclusive.genes[[pathway]]) > 0){
-           print(sample)
-           print(pathway)
-           d[(d$Sample == sample & d$Pathway == pathway), 'Exclusive'] = TRUE
-        }
-     }
-}
-
-p$data = d
-
-
-p <- p + aes(fill=Exclusive)
-
-p <- p + opts(axis.text.x=theme_text(angle=90), panel.background = theme_rect(fill='white', colour='steelblue'))
-
-p
-
-
-
-
-ggsave(p, filename='#{png_file}', height=#{size}, width=#{size});
-"
-  Open.read(png_file, :mode => 'rb')
-end
-
-
-

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

@@ -1,78 +0,0 @@
-module Study
-end
-
-module Sample
-  extend Entity
-
-  annotation :study
-
-  self.format = ["Sample ID"]
-
-  def dir
-    return nil if study.nil?
-    return study.dir if study.respond_to? :dir
-    begin
-      Study.setup(study).dir
-    rescue
-      Log.warn "Error accessing sample dir from study: #{$!.message}"
-      nil
-    end
-  end
-
-  def organism
-    return nil if study.nil?
-    study.organism
-  end
-
-  def study
-    @study ||= begin
-                 study = info[:study] 
-                 if study.nil?
-                   study = Study.identify_study(self)
-                   self.study = study
-                 end
-                 study
-               end
-  end
-
-end
-
-module Study
-
-  def sample_info
-    return nil unless dir.samples.exists?
-    @sample_info ||= dir.samples.tsv.tap{|tsv| tsv.entity_options = {:study => self }}
-  end
-
-  def samples
-    @samples ||= begin
-                   samples = local_persist("Sample", :array) do
-                     if sample_info.nil?
-                       self.cohort.collect{|g| g.jobname }
-                     else
-                       sample_info.keys
-                     end
-                   end
-                   Sample.setup(samples, :study => self)
-                   samples.study = self
-                   samples
-                 end
-  end
-
-  def match_samples(list)
-    if donor_id_field = (sample_info = self.sample_info).fields.select{|f| f =~ /donor\s+id/i}.first
-      list_donors = sample_info.select(list).slice(donor_id_field).values.compact.flatten
-      list_donor_samples = sample_info.select(list_donors).keys
-      list = list_donor_samples.annotate((list + list_donor_samples).uniq)
-    end
-    list
-  end
-
-  def self.identify_study(samples)
-    samples = Array === samples ? samples.flatten : [samples]
-
-    studies = Study.studies.select{|study| Study.setup(study); (study.samples & samples).any? }
-
-    studies.first
-  end
-end

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

@@ -1,87 +0,0 @@
-require 'rbbt/entity/snp'
-
-#require 'rbbt/entity/study/snp/samples'
-
-module StudyWorkflow
-end
-
-module Study
-  def has_snp?
-    dir.snp.exists?
-  end
-
-  def snp_files
-    @snp_files ||= dir.snp.find.glob("*")
-  end
-
-  def snp_cohort
-    if @snp_cohort.nil?
-      @snp_cohort = {}
-      snp_files.each do |f| 
-        sample = File.basename(f)
-        Sample.setup(sample, self)
-        snps = Open.read(f).split("\n").sort
-        SNP.setup(snps)
-        @snp_cohort[sample] =  snps
-      end
-    end
-    @snp_cohort
-  end
-end
-
-module Study
-
-  def snp_index
-    local_persist_tsv("SNP2Samples", "SNP2Samples", {}, :persist => true, :serializer => :clean) do |data|
-
-      require 'progress-monitor'
-      Progress.monitor "SNP files", :stack_depth => 0
-      snp_files.each do |file|
-        file = file.to_s
-        sample = File.basename file
-        File.open(file.to_s) do |f|
-          while line = f.gets
-            snp = line.strip
-            snp, allele = snp.split ":"
-            snp_str = data[snp]
-
-            if snp_str.nil?
-              snp_str = ""
-            else
-              snp_str += "\t"
-            end
-
-            if allele
-              snp_str << sample << ":" << allele
-            else
-              snp_str << sample
-            end
-            data[snp] = snp_str
-          end
-        end
-      end
-
-      TSV.setup data
-      data.key_field = "RS ID"
-      data.fields = ["Sample"]
-      data.type = :flat
-      data.serializer = :list
-      data
-    end
-  end
-
-  property :samples_with_snp => :single2array do |snp|
-    Sample.setup((snp_index[snp] || []).collect{|s| s.split(":").first}, self)
-  end
-
-  property :samples_with_homozygous_snp => :single2array do |snp|
-    Sample.setup((snp_index[snp] || []).collect{|s| s.split(":")}.select{|s,g| g == "2"}.collect{|s,g| s}, self)
-  end
-
-  property :samples_with_heterozygous_snp => :single2array do |snp|
-    Sample.setup((snp_index[snp] || []).collect{|s| s.split(":")}.select{|s,g| g == "1"}.collect{|s,g| s}, self)
-  end
-
-
-
-end