DomFun 0.1.1 → 0.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ab8f445b1ecc8e416c559fb87b46c853e2cc74e1
-  data.tar.gz: 7d28a92576ffc5f4816c0062b4702f6195d2e178
+  metadata.gz: 8d744b8695b8d35ec3ec38ec0ea52f6f6c7b4bbc
+  data.tar.gz: ed2b3da0e0db38bcae175d477bcee641106684e9
 SHA512:
-  metadata.gz: 88cbe093ee831669f5bc6037bfafa4d94e1865ab032fc613ae62550b46811d502a43ec7f1c632a81bb622d36923cafc493c400d7ecf98c572434461ad8731dd0
-  data.tar.gz: 9820f00ac5e9d63b88449bc8a0e47b5c97c7e6541791d24522b0a97ba64363d61df1343f4acf04d65a7036ccd9cfc8529ac5c672cf69b7e3f2eec6994812f736
+  metadata.gz: 4be2da53056a56a7edae46c41c3ce84e525e93b602f2469084317ca1a9ee49872e684c3a6135a847f34fabc61133fdff7d767a410d976154eefd9a4389a96d24
+  data.tar.gz: 43c5717119062bee38dd29eff45fad55f120ac4158579c54d942e3fc688e16d23a2e9a5332608c4324e7196dbdac877397afaa5febbbad4e860e9a76dde98f5b

data/DomFun.gemspec

@@ -37,7 +37,7 @@ Gem::Specification.new do |spec|
   spec.require_paths = ["lib"]
 
   spec.add_development_dependency "bundler", "~> 2.0"
-  spec.add_development_dependency "rake", "~> 10.0"
+  spec.add_development_dependency "rake", "~> 12.3.1"
   spec.add_development_dependency "rspec", "~> 3.0"
 
   spec.add_dependency "NetAnalyzer", "~> 0.1.5"

data/bin/add_protein_functional_families.rb

@@ -9,14 +9,15 @@
 REPORT_FOLDER=File.expand_path(File.join(File.dirname(__FILE__), '..', 'templates'))
 ROOT_PATH = File.dirname(__FILE__)
 $: << File.expand_path(File.join(ROOT_PATH, '..', 'lib', 'DomFun'))
-require 'generalMethods.rb'
+require 'generalMethods'
 require 'csv'
 require 'optparse'
+require 'fileutils'
 
 ##########################
 #METHODS
 ##########################
-def build_tripartite_networks(nomenclature_annotations, cath_data, path, protein2gene)
+def build_tripartite_networks(nomenclature_annotations, cath_data, path, protein2gene, translate2gene)
 	records = Hash.new(0)
   nomenclature_annotations.each do |nomenclature, protein_annotations|
 		annots = []
@@ -24,16 +25,17 @@ def build_tripartite_networks(nomenclature_annotations, cath_data, path, protein
 		protein_annotations.each do |protID, annotations|
 			query_cath_data = cath_data[protID]
 			if !query_cath_data.nil?
-				#gene_ID = protein2gene[protID] unless protein2gene[protID].nil?
-        
-				#gene_ID = protID if gene_ID.nil?
+        if !translate2gene
+          recordID = protID
+        else
+  				recordID = protein2gene[protID]
+  				recordID = protID if recordID.nil?
+        end
 				annotations.each do |annotation|
-          #annots << [annotation, gene_ID]
-					annots << [annotation, protID]
+					annots << [annotation, recordID]
 				end
 				query_cath_data.each do |data|
-          #datas << [data, gene_ID]
-					datas << [data, protID]
+					datas << [data, recordID]
 				end
 			end
 		end
@@ -77,6 +79,11 @@ OptionParser.new do |opts|
     options[:search_domain] = false
   end
 
+  options[:output_file] = 'uniprot_translated.txt'
+  opts.on("-o", "--output_file PATH", "Output file with UniProt GeneName structure for prediction") do |data|
+    options[:output_file] = data
+  end
+
   options[:annotation_types] = %w[ kegg reactome go]
   opts.on("-p", "--annotation_types STRING", "List of annotation types separated by commas") do |data|
     options[:annotation_types] = data.split(",")
@@ -87,6 +94,11 @@ OptionParser.new do |opts|
     options[:output_stats] = data
   end
 
+  options[:translate2gene] = false
+  opts.on("-T", "--translate2gene", "Translate proteins to genes") do
+    options[:translate2gene] = true
+  end
+
   options[:category_type] = 'funfamID'
   opts.on("-t", "--category_type STRING", "Input category of domains. Options: funfamID, superfamilyID") do |data|
     options[:category_type] = data
@@ -109,25 +121,24 @@ end.parse!
 ##########################
 
 puts "Loading data..."
-cath_data, protein2gene, gene2proteins, cath_proteins_number = load_cath_data(options[:input_domains], options[:category_type])
+cath_data, protein2gene, cath_proteins_number = load_cath_data(options[:input_domains], options[:category_type])
 nomenclature_annotations, number_of_proteins, proteins_without_annotations = load_proteins_file(options[:input_annotations], options[:annotation_types])
+
 networks_path = nil
 if options[:category_type] == 'funfamID'
 	networks_path = 'networks/funfam_networks'
 else
 	networks_path = 'networks/superfamily_networks'
 end
+FileUtils.mkdir_p networks_path
 puts "Generating tripartite networks. This can take a while, please wait."
-protein_stats = build_tripartite_networks(nomenclature_annotations, cath_data, networks_path, protein2gene)
-handler = File.open(options[:output_stats], 'w')
-protein_stats.each do |annotation_type, number_of_proteins|
-	handler.puts "#{annotation_type}\t#{number_of_proteins}"
-end
-handler.puts "Total of Uniprot proteins\t#{number_of_proteins}"
-handler.puts "Total of Uniprot proteins without annotations\t#{proteins_without_annotations.length}"
-handler.puts "Total of CATH proteins\t#{cath_proteins_number}"
-handler = File.open(options[:unnanotated_proteins], 'w')
-proteins_without_annotations.each do |unnanotated_prot|
-	handler.puts unnanotated_prot
+protein_stats = build_tripartite_networks(nomenclature_annotations, cath_data, networks_path, protein2gene, options[:translate2gene])
+File.open(options[:output_stats], 'w') do |f|
+  protein_stats.each do |annotation_type, number_of_proteins|
+  	f.puts "#{annotation_type}\t#{number_of_proteins}"
+  end
+  f.puts "Total of Uniprot proteins\t#{number_of_proteins}"
+  f.puts "Total of Uniprot proteins without annotations\t#{proteins_without_annotations.length}"
+  f.puts "Total of CATH proteins\t#{cath_proteins_number}"
 end
-handler.close
+File.open(options[:unnanotated_proteins], 'w') {|f| f.puts proteins_without_annotations.join("\n")}

data/bin/association_metrics_average.rb

@@ -0,0 +1,94 @@
+#! /usr/bin/env ruby
+#Tool for calculating averages between different association values file.
+#File structure: prec rec cut meth
+#Load all files (7) stored in the same directory and calculate average;
+#of lines for each method. Return a file with the same structure;
+#giving name as "average" to the last column
+
+require 'optparse'
+
+##########################
+#METHODS
+##########################
+
+def load_association_file(filename)
+	fileInfo = []
+	header = ''
+	line_number = 0
+	File.open(filename).each do |line|
+		line.chomp!
+		if line_number == 0
+			header = line
+		else
+			cut, precision, recall, meth = line.split("\t")
+			fileInfo << [cut.to_f, precision.to_f, recall.to_f, meth]
+		end
+		line_number += 1
+	end
+	return fileInfo, header
+end
+
+def calculate_average(all_files, cols_for_average)
+	average = []
+	n_files = all_files.length.to_f
+	ref_file = all_files.shift
+	summatory_file = []
+	ref_file.each_with_index do |line, i|
+		all_files.each do |file|
+			line2 = file[i]
+			cols_for_average.each do |col|
+				line[col] = line[col] + line2[col]
+			end
+		end
+		summatory_file << line
+	end
+	summatory_file.each do |line|
+		cols_for_average.each do |col|
+			line[col] = line[col]/n_files
+		end
+		average << line
+	end
+	return average
+end
+
+
+##########################
+#OPT-PARSER
+##########################
+
+options = {}
+OptionParser.new do |opts|
+  opts.banner = "Usage: #{__FILE__} [options]"
+ 
+  options[:file_names] = nil
+  opts.on("-f", "--file_names STRING", "Input file names to calculate averages. Please separate names by commas") do |file_names|
+    options[:file_names] = file_names.split(',')
+  end
+
+  options[:which_cols] = nil
+  opts.on("-c", "--which_cols STRING", "Cols for performing average analysis") do |which_cols|
+    options[:which_cols] = which_cols.split(',').map{|i| i.to_i - 1}
+  end
+
+end.parse!
+
+##########################
+#MAIN
+##########################
+
+all_files = []
+header = nil
+options[:file_names].each do |filename|
+	file, header = load_association_file(filename)
+	all_files << file
+end
+
+average = calculate_average(all_files, options[:which_cols])
+
+puts header
+average.each do |line|
+	puts line.join("\t")
+end
+	
+
+

data/bin/domains_to_function_predictor.rb

@@ -11,7 +11,7 @@
 REPORT_FOLDER=File.expand_path(File.join(File.dirname(__FILE__), '..', 'templates'))
 ROOT_PATH = File.dirname(__FILE__)
 $: << File.expand_path(File.join(ROOT_PATH, '..', 'lib', 'DomFun'))
-require 'generalMethods.rb'
+require 'generalMethods'
 require 'csv'
 require 'optparse'
 require "statistics2"
@@ -23,15 +23,24 @@ require 'report_html'
 #METHODS
 ##########################
 
-def get_protein_domains(cath_data, protein)
-  domains_to_predict = nil
-  unless cath_data[protein].nil?
-    domains_to_predict = cath_data[protein].uniq
+def get_protein_domains(cath_data, protein, gene2protein, identifier_mode)
+  domains_to_predict = []
+  if identifier_mode == 'mixed'
+    proteins = gene2protein[protein]
+    if !proteins.nil?
+      proteins.each do |protein|
+        domains_to_predict.concat(cath_data[protein])
+      end
+    else
+      domains_to_predict = cath_data[protein]
+    end
+  else
+    domains_to_predict = cath_data[protein]
   end
-  return domains_to_predict
+  domains_to_predict = [] if domains_to_predict.nil?
+  return domains_to_predict.uniq
 end
 
-
 def load_domain_to_pathway_association(associations_file, threshold)
 	domain_to_pathway_associations = {}
 	File.open(associations_file).each do |line|
@@ -62,7 +71,6 @@ end
 def search4function(domains_to_predict, domain_to_pathway_associations)
   domain_to_function_and_association_value = {}
   domains_to_predict.each do |domain|
-    #puts domain
     associations = domain_to_pathway_associations[domain]
     if !associations.nil?
       domain_to_function_and_association_value[domain] = associations
@@ -132,7 +140,6 @@ def scoring_funsys(function_to_domains, domain_annotation_matrix, scoring_system
       combined_pvalue = Statistics2.chi2_x(sample_length *2, -2*sum)
       domains_array[i] << combined_pvalue
     elsif scoring_system == 'harmonic'
-      #STDERR.puts associations.inspect
       lns = associations.map{|a| 10 ** -a}
       inv = lns.map{|n| 1.fdiv(n)}
       sum = inv.inject(0){|s,x| s + x}
@@ -144,7 +151,6 @@ def scoring_funsys(function_to_domains, domain_annotation_matrix, scoring_system
       domains_array[i] << combined_z_score
     elsif scoring_system == 'average'
       sum = associations.inject(0){|s,x| s + x.abs}.fdiv(associations.length)
-      #STDERR.puts sum.inspect
       domains_array[i] << sum
     elsif scoring_system == 'sum'
       sum = associations.inject(0){|s,x| s + x.abs}
@@ -158,7 +164,6 @@ def scoring_funsys(function_to_domains, domain_annotation_matrix, scoring_system
   else
     function_to_domains.select!{|function, attributes| attributes.last >= pvalue_threshold}
   end
-  #STDERR.puts function_to_domains.inspect
 end
 
 
@@ -198,6 +203,11 @@ OptionParser.new do |opts|
     options[:integration_method] = data
   end
 
+  options[:identifier_mode] = 'normal'
+  opts.on("-I", "--identifier_mode STRING", "Identifier mode: normal or mixed") do |data|
+    options[:identifier_mode] = data
+  end
+
   options[:output_file] = 'predictions_file.txt'
   opts.on("-o", "--output_file PATH", "Predictions file") do |data|
     options[:output_file] = data
@@ -235,7 +245,7 @@ end.parse!
 ##########################
 
 # 1. Load protein domains classification to get domains from proteins to predict
-cath_data, protein2gene, gene2proteins, cath_proteins_number = load_cath_data(options[:protein_domains_file], options[:domain_category])
+cath_data, protein2gene, cath_proteins_number = load_cath_data(options[:protein_domains_file], options[:domain_category])
 # 2. Load protein(s) to predict
 if File.exist?(options[:proteins_2predict])
   if !options[:multiple_proteins]
@@ -256,17 +266,16 @@ else
   end
 end
 
-
 # 3. Load domain-FunSys associations
 domain_to_pathways_associations = load_domain_to_pathway_association(options[:input_associations], options[:association_threshold])
 # 4. Prediction
-handler = File.open(options[:output_file], 'w')
+#handler = File.open(options[:output_file], 'w')
+gene2protein = invert_hash(protein2gene) if options[:identifier_mode] == 'mixed'
 options[:proteins_2predict].each do |protein|
-  domains = get_protein_domains(cath_data, protein)
-  next if domains.nil?
+  domains = get_protein_domains(cath_data, protein, gene2protein, options[:identifier_mode])
+  next if domains.empty?
   null_value = 0
   domain_function_assocValue = search4function(domains, domain_to_pathways_associations)
-  
   function_to_domains, association_scores = group_by_function(domain_function_assocValue) 
   annotation_matrix = generate_domain_annotation_matrix(function_to_domains, association_scores, domains, 0) 
   
@@ -281,7 +290,8 @@ options[:proteins_2predict].each do |protein|
 
     function_to_domains.each do |funsys, domains_data|
       score = domains_data.pop
-      handler.puts "#{protein}\t#{domains_data.join(',')}\t#{funsys}\t#{score}"
+      #handler.puts "#{protein}\t#{domains_data.join(',')}\t#{funsys}\t#{score}"
+      puts "#{protein}\t#{domains_data.join(',')}\t#{funsys}\t#{score}"
     end
 end
-handler.close
+#handler.close

data/lib/DomFun/generalMethods.rb

@@ -4,6 +4,7 @@ def load_proteins_file(file, annotation_types)
 	annotation_types.each do |type| # initialize annotation hashes
 		protein_annotations[type] = {}
 	end
+	fields_to_split = annotation_types.length 
 	counter = 0
 	File.open(file).each do |line|
 		line.chomp!
@@ -12,7 +13,7 @@ def load_proteins_file(file, annotation_types)
 			next
 		end
 		line.gsub!(' ', '')
-		fields = line.split("\t", 4)
+		fields = line.split("\t", fields_to_split + 1)
 		protID = fields.shift
 		annotation_types.each_with_index do |type, i|
 			annotations = fields[i].split(/[;,]/)
@@ -28,7 +29,7 @@ def load_proteins_file(file, annotation_types)
 					protein_annotations[type][protID] = annotations
 				end
 			end
-			if fields.count("") == 3
+			if fields.count("") == fields_to_split
 				proteins_without_annotations << protID
 			end
 		end
@@ -37,23 +38,22 @@ def load_proteins_file(file, annotation_types)
 	return protein_annotations, counter, proteins_without_annotations.uniq
 end
 
-def load_cath_data(file, category, meth='protACC')
+def load_cath_data(file, category, dictionary_key='gene_name')
+	if dictionary_key == 'gene_name'
+		field = 3
+	elsif dictionary_key == 'geneID' # UNIPROT entry_name
+		field = 4
+	end
 	cath_data = {}
-	protein2gene = {}
-	gene2proteins = {}
+	protein2gene_dict = {}
 	csv_file = CSV.read(file, { :col_sep => "\t" })
 	csv_file.delete_at(0)
 	csv_file.each do |protein_domains_data|
 		next if protein_domains_data.empty?
 		protein_id = protein_domains_data[0]
-		if meth == 'protACC'
-			field = 3
-		elsif meth == 'geneID'
-			field = 4
-		end
-		gene_name = protein_domains_data[field]
-		next if gene_name.include?('fusion')
-		gene_name = gene_name.gsub(' ', '_') if gene_name.include?(' ')
+		protein_alternative_name = protein_domains_data[field]
+		next if protein_domains_data[3].include?('fusion') # Only can checked in cath gene name field
+		protein_alternative_name.gsub!(' ', '_') if protein_alternative_name.include?(' ')
 		superfamilyID = protein_domains_data[5]
 		funfamID = protein_domains_data[6]
 		term2save = nil
@@ -63,16 +63,23 @@ def load_cath_data(file, category, meth='protACC')
 			term2save = funfamID
 		end
 		add_term2dictionary(cath_data, protein_id, term2save)
-		protein2gene[protein_id] = gene_name if gene_name != 'NULL'
-		query = gene2proteins[gene_name]
-	    if query.nil?
-	      	gene2proteins[gene_name] = [protein_id] if protein_id != 'NULL'
-	    else
-	      query << protein_id if protein_id != 'NULL'
-		end
+		protein2gene_dict[protein_id] = protein_alternative_name if protein_alternative_name != 'NULL'
 	end
 	cath_proteins_number = cath_data.keys.length
-	return cath_data, protein2gene, gene2proteins, cath_proteins_number
+	return cath_data, protein2gene_dict, cath_proteins_number
+end
+
+def invert_hash(hash)
+	new_hash = {}
+	hash.each do |k, v|
+		query = new_hash[v]
+		if query.nil?
+			new_hash[v] = [k]
+		else
+			query << k
+		end
+	end
+	return new_hash
 end
 
 def add_term2dictionary(dict, key, term)

data/lib/DomFun/version.rb

@@ -1,3 +1,3 @@
 module DomFun
-  VERSION = "0.1.1"
+  VERSION = "0.1.2"
 end

data/templates/report_data.erb

@@ -0,0 +1,59 @@
+<div class="chunk">
+	<style type="text/css" scoped>	
+	.tooltip {
+    position: relative;
+    display: inline-block;
+    border-bottom: 1px dotted black;
+}
+
+.tooltip .tooltiptext {
+    visibility: hidden;
+    width: 120px;
+    background-color: #555;
+    color: #fff;
+    text-align: center;
+    border-radius: 6px;
+    padding: 5px 0;
+    position: absolute;
+    z-index: 1;
+    bottom: 125%;
+    left: 50%;
+    margin-left: -60px;
+    opacity: 0;
+    transition: opacity 1s;
+}
+
+.tooltip .tooltiptext::after {
+    content: "";
+    position: absolute;
+    top: 100%;
+    left: 50%;
+    margin-left: -5px;
+    border-width: 5px;
+    border-style: solid;
+    border-color: #555 transparent transparent transparent;
+}
+
+.tooltip:hover .tooltiptext {
+    visibility: visible;
+    opacity: 1;
+}
+	</style>
+</div>
+
+<div>
+<h2 style="text-align: center; background-color:#d6eaf8">Ranked FunSys predicted based on queried protein domains</h2>
+<p>The following table shows the protein domains, the FunSys associated, the association value calculated between the domain and the FunSys and the combined scored. Results are ranked from the highest to the lowest combined score.</p>
+	<%= 
+	if !@hash_vars[:predictions].nil?
+		# gene_var = @hash_vars[:genes_with_kegg_data]
+		table(id: :predictions, header: true, border: 2, 
+		cell_align: %w(center, center center center center)) do |data|
+		predictions = @hash_vars[:predictions]
+		# data.each_with_index do |row, row_number|
+		data.unshift(["ProteinID", "Domains", "FunSys", "Association values", "Combined score"])
+		end
+	end
+	%>
+
+</div>

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: DomFun
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.1.2
 platform: ruby
 authors:
 - Elena Rojano, Pedro Seoane
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2019-11-21 00:00:00.000000000 Z
+date: 2020-06-16 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -30,14 +30,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '10.0'
+        version: 12.3.1
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '10.0'
+        version: 12.3.1
 - !ruby/object:Gem::Dependency
   name: rspec
   requirement: !ruby/object:Gem::Requirement
@@ -83,6 +83,7 @@ files:
 - README.md
 - Rakefile
 - bin/add_protein_functional_families.rb
+- bin/association_metrics_average.rb
 - bin/console
 - bin/domains_to_function_predictor.rb
 - bin/generate_CAFA2_dataset.rb
@@ -100,6 +101,7 @@ files:
 - lib/DomFun.rb
 - lib/DomFun/generalMethods.rb
 - lib/DomFun/version.rb
+- templates/report_data.erb
 homepage: https://bitbucket.org/elenarojano/domfun
 licenses:
 - MIT