sylfy 0.0.2 → 0.1.0

data/lib/sylfy/service/keggrest/conv.rb

@@ -1,7 +1,3 @@
-# 
-# UniSysDB library in Ruby
-# Copyright (C) 2012
-#
 # @author Natapol Pornputtapong <natapol@chalmers.se>
 #
 
@@ -12,47 +8,75 @@ module Sylfy
 		module KEGGREST
 			
 			module_function
-			
-			def conv(id, targetdb = nil)
+			# interface to KEGG conv service
+			#
+			# @param targetdb [Symbol] please consult {http://www.kegg.jp/kegg/rest/keggapi.html}
+			#
+			# @param sourcedb [String] can be database name or specific ID(s) joined with '+'
+			#	
+			# @return [Hash] results with sourcedb ID as keys
+			# 
+			def conv(targetdb, *sourcedb)
 				
-				help = %Q(
-help:	conv(id, targetdb = nil)
-help:
-help:	
-help:	targetdb = <database>
-help:	For gene identifiers:
-help:	<kegg_db> = <org>
-help:	<org> = KEGG organism code or T number
-help:	<outside_db> = ncbi-gi | ncbi-geneid | uniprot
-help:
-help:	For chemical substance identifiers:
-help:	<kegg_db> = drug | compound | glycan
-help:	<outside_db> = pubchem | chebi
-help:	
-help:	return a Hash
-				)
-				if id == :help
-					puts help
-					return {}
-				else
+				begin
+					doc = URI.parse("#{Sylfy::Service::KEGGREST::BASEURI}/conv/#{targetdb.to_s}/#{sourcedb.join("+")}").read().strip()
+					result = {}
 					
-					begin
-						doc = URI.parse("#{@@baseuri}/conv/#{targetdb}/#{id}").read().strip()
-						result = {}
-						
-						doc.split(/\n/).each do |line|
-							dat = line.chomp.split(/\t/)
-							result[dat[0]] = [] if !result.has_key?(dat[0])
-							result[dat[0]].push(dat[1])
-						end
-						
-						return result
-						
-					rescue OpenURI::HTTPError
-						raise Sylfy::Service::DataNotFound, "Query not found."
+					doc.split(/\n/).each do |line|
+						dat = line.chomp.split(/\t/)
+						result[dat[0]] = [] if !result.has_key?(dat[0])
+						result[dat[0]].push(dat[1])
 					end
 					
+					return result
+					
+				rescue OpenURI::HTTPError
+					raise Sylfy::Service::DataNotFound, "Query not found."
 				end
+				
+			end
+			
+			def ncbi2ko(*sourcedb)
+				begin
+					ids = sourcedb.flatten.map { |x| x =~ /^\d+$/ ? "ncbi-geneid:" + x : x}
+					hash1 = conv(:genes, ids)
+					list1 = []
+					hash1.each_value do |v|
+						list1 += v
+					end
+					hash2 = link(:ko, list1)
+					results = {}
+					hash1.each_pair do |k, vs|
+						results[k] ||= []
+						vs.each {|v| results[k] = results[k] | hash2[v] if hash2.has_key?(v)}
+					end
+					
+					results.delete_if {|key, value| value.empty? }
+					
+					return results
+				rescue Sylfy::Service::DataNotFound
+					return {}
+				end
+			end
+			
+			def ncbi2ko_org(org)
+				
+				results = {}
+				
+				begin
+					hash1 = conv(org.to_s, 'ncbi-geneid')
+					hash2 = link(:ko, org.to_s)
+					hash1.each_pair do |k, vs|
+						results[k] ||= []
+						vs.each {|v| results[k] = results[k] | hash2[v] if hash2.has_key?(v)}
+					end
+					results.delete_if {|key, value| value.empty? }
+					return results
+				rescue
+					return results
+				end
+				
+				
 			end
 		
 		end

data/lib/sylfy/service/keggrest/find.rb

@@ -1,7 +1,3 @@
-# 
-# UniSysDB library in Ruby
-# Copyright (C) 2012
-#
 # @author Natapol Pornputtapong <natapol@chalmers.se>
 #
 
@@ -12,28 +8,17 @@ module Sylfy
 		module KEGGREST
 			
 			module_function
-			# Method for setting inchi data member
+			# interface to KEGG find service
 			#
-			# == Parameters:
-			# inchi::
-			#   input inchi string
-			def find(database, query=nil, option=nil)
-				
-				help = "
-help:	find(database, query, option=nil)
-help:
-help:	type#1
-help:	
-help:	database = <database>
-help:	<database> = pathway | module | disease | drug | environ | ko | genome |
-help:		<org> | compound | glycan | reaction | rpair | rclass | enzyme |
-help:		genes | ligand
-help:	<org> = KEGG organism code or T number
-help:	
-help:	<option> = formula | exact_mass | mol_weight only for <database> = compound | drug
-help:	
-help:	return a List
-				"
+			# @param database [Symbol] database name can be :pathway, :module, :disease, :drug, :environ, :ko, :genome, org, :compound, :glycan, :reaction, :rpair, :rclass, :enzyme, :genes, :ligand
+			#
+			# @param query [String] Query text keyword can be join with + sign or in double qoutw
+			#   
+			# @param option [Symbol] can be :formula, :exact_mass, :mol_weight with database = :compound, :drug
+			#	
+			# @return [Hash] results with sourcedb ID as keys
+			#
+			def find(database, query, option=nil)
 				
 				result_tag = {
 					:pathway => [:NAME],
@@ -52,31 +37,25 @@ help:	return a List
 				}
 				result_tag.default = [:NAME, :DEFINITION, :ORTHOLOGY]
 				result = {}
-				if database == :help
-					puts help
-					return result
-				else
-					begin
-						if ['compound', 'drug'].include?(database.to_s) && option != nil &&
-							['formula', 'exact_mass' 'mol_weight'].include?(option.to_s)
-							URI.parse("#{@@baseuri}/find/#{database}/#{query}/#{option}").read().strip().each do |line|
-								dat = line.chomp.split(/\t/)
-								result[dat[0]] = { option.to_sym => dat[1]}
-							end
-						else
-							URI.parse("#{@@baseuri}/find/#{database}/#{query}").read().strip().each do |line|
-								dat = line.chomp.split(/\t/)
-								result[dat[0]] = {}
-								info = dat[1].split('; ')
-								info.each_index {|ind| result[dat[0]][result_tag[database.to_sym][ind]] = info[ind]}
-							end
+				begin
+					if ['compound', 'drug'].include?(database.to_s) && option != nil &&
+						['formula', 'exact_mass' 'mol_weight'].include?(option.to_s)
+						URI.parse("#{Sylfy::Service::KEGGREST::BASEURI}/find/#{database}/#{URI.escape(query)}/#{option}").read.strip.split(/\n/).each do |line|
+							dat = line.chomp.split(/\t/)
+							result[dat[0]] = { option.to_sym => dat[1]}
+						end
+					else
+						URI.parse("#{Sylfy::Service::KEGGREST::BASEURI}/find/#{database}/#{URI.escape(query)}").read.strip.split(/\n/).each do |line|
+							dat = line.chomp.split(/\t/)
+							result[dat[0]] = {}
+							info = dat[1].split('; ')
+							info.each_index {|ind| result[dat[0]][result_tag[database.to_sym][ind]] = info[ind]}
 						end
-						
-						return result
-					rescue OpenURI::HTTPError
-						raise Unisys::ServiceException, "Query not found."
 					end
-				
+					
+					return result
+				rescue OpenURI::HTTPError
+					raise Unisys::ServiceException, "Query not found."
 				end
 				
 			end

data/lib/sylfy/service/keggrest/get.rb

@@ -0,0 +1,82 @@
+# @author Natapol Pornputtapong <natapol@chalmers.se>
+#
+
+module Sylfy
+	
+	module Service
+	
+		module KEGGREST
+			
+			module_function
+			# interface to KEGG get service
+			#
+			# @param dbentries [Symbol, String] can be :pathway, :brite, :module, :disease, :drug, :environ, :ko, :genome, :org, :compound, :glycan, :reaction, :rpair, :rclass, :enzyme
+			#
+			# @param option [Symbol] can be aaseq, :ntseq, :mol, :kcf, :image, :kgml
+			#	
+			# @return [Array] An array of Bio::KEGG object or string if option = :mol, :kcf, :kgml or Bio::FastaFormat if option =  :aaseq, :ntseq
+			#
+			def get(dbentries, option = nil)
+				
+				if option != nil && ['aaseq', 'ntseq', 'mol', 'kcf', 'kgml'].include?(option.to_s)
+					text = "#{dbentries}/#{option}"
+				else
+					text = "#{dbentries}"
+				end
+				
+				begin
+					doc = URI.parse("#{Sylfy::Service::KEGGREST::BASEURI}/get/#{text}").read
+					
+				rescue OpenURI::HTTPError
+					raise Sylfy::Service::DataNotFound, "Query not found."
+				end
+				
+				if ['aaseq', 'ntseq'].include?(option.to_s)
+					return Bio::FastaFormat.new(doc)
+				elsif option.to_s == 'mol'
+					return doc
+				elsif option.to_s == 'kcf'
+					return doc
+				elsif option.to_s == 'kgml'
+					return Bio::KEGG::KGML.new(doc.strip)
+				else
+					result = []
+				
+					doc.strip.split('///').each do |entry|
+						case entry
+						when /^ENTRY\s+[A-Z0-9]+\s+Glycan/
+							result.push(Bio::KEGG::GLYCAN.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+Compound/
+							result.push(Bio::KEGG::COMPOUND.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+CDS/
+							result.push(Bio::KEGG::GENES.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+Drug/
+							result.push(Bio::KEGG::DRUG.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+Reaction/
+							result.push(Bio::KEGG::REACTION.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+RPair/
+							result.push(entry)
+						when /^ENTRY\s+[A-Z0-9]+\s+Genome/
+							result.push(Bio::KEGG::GENOME.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+Enzyme/
+							result.push(Bio::KEGG::ENZYME.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+Module/
+							result.push(Bio::KEGG::MODULE.new(entry))
+						when /^ENTRY\s+[A-Z0-9]+\s+Pathway/
+							result.push(Bio::KEGG::PATH.new(entry))
+						else
+							result.push(entry)
+						end
+					end
+				end
+				
+				return result
+				
+			end
+		
+		end
+		
+	end
+	
+end
+

data/lib/sylfy/service/keggrest/link.rb

@@ -1,7 +1,3 @@
-# 
-# UniSysDB library in Ruby
-# Copyright (C) 2012
-#
 # @author Natapol Pornputtapong <natapol@chalmers.se>
 #
 
@@ -12,44 +8,28 @@ module Sylfy
 		module KEGGREST
 			
 			module_function
-			# Method for setting inchi data member
+			# interface to KEGG link service
+			#
+			# @param targetdb [Symbol] can be pathway, :brite, :module, :disease, :drug, :environ, :ko, :genome, org, :compound, :glycan, :reaction, :rpair, :rclass, :enzyme
+			#
+			# @param sourcedb [Symbol] :pathway, :brite, :module, :disease, :drug, :environ, :ko, :genome, org, :compound, :glycan, :reaction, :rpair, :rclass, :enzyme, :genes
 			#
-			# == Parameters:
-			# inchi::
-			#   input inchi string
-			def link(id, targetdb = nil)
+			# @return [Hash] results with ID and description
+			#
+			def link(targetdb, *sourcedb)
 				
-				help = "
-help:	link(id, targetdb = nil)
-help:
-help:	targetdb = pathway | brite | module | disease | drug | environ | ko | genome | <org> | compound | glycan | reaction | rpair | rclass | enzyme
-help:
-help:	id = KEGG database entries involving the following databases: pathway | brite | module | disease | drug | environ | ko | genome |
-help:		<org> | compound | glycan | reaction | rpair | rclass | enzyme
-help:	<org> = KEGG organism code or T number
-help:
-help:	<option> = aaseq | ntseq | mol | kcf | image
-				"
-				if id == :help
-					puts help
-					return {}
-				else
-					
-					begin
-						doc = URI.parse("#{@@baseuri}/link/#{targetdb}/#{id}").read().strip()
-						result = {}
-						
-						doc.each do |line|
-							dat = line.chomp.split(/\t/)
-							result[dat[0]] = result.has_key?(dat[0]) ? result[dat[0]].push(dat[1]) : [dat[1]]
-						end
+				begin
+					doc = URI.parse("#{Sylfy::Service::KEGGREST::BASEURI}/link/#{targetdb.to_s}/#{sourcedb.join("+")}").read().strip()
+					result = {}
 					
-						return result
-					rescue OpenURI::HTTPError
-						raise Sylfy::Service::DataNotFound, "Query not found."
+					doc.split(/\n/).each do |line|
+						dat = line.chomp.split(/\t/)
+						result[dat[0]] = result.has_key?(dat[0]) ? result[dat[0]].push(dat[1]) : [dat[1]]
 					end
 				
-					
+					return result
+				rescue OpenURI::HTTPError
+					raise Sylfy::Service::DataNotFound, "Query not found."
 				end
 				
 			end

data/lib/sylfy/service/keggrest/list.rb

@@ -1,7 +1,3 @@
-# 
-# UniSysDB library in Ruby
-# Copyright (C) 2012
-#
 # @author Natapol Pornputtapong <natapol@chalmers.se>
 #
 
@@ -12,55 +8,27 @@ module Sylfy
 		module KEGGREST
 			
 			module_function
-			
-			def list(param)
+			# interface to KEGG list service
+			#
+			# @param database [Symbol] :pathway, :module, :disease, :drug, :environ, :ko, :genome, org, :compound, :glycan, :reaction, :rpair, :rclass, :enzyme, :genes, :ligand
+			#
+			# @param org [Symbol] organism code can be :hsa, :eco, etc {http://www.kegg.jp/kegg/rest/keggapi.html}
+			#
+			# @return [Hash] results with ID and description
+			#
+			def list(database, org = nil)
 				
-				help = %Q(
-help:	list(param)
-help:
-help:	type#1
-help:	
-help:	param = \"<database>\"
-help:	<database> = pathway | brite | module | disease | drug | environ | ko | genome |
-help:					<org> | compound | glycan | reaction | rpair | rclass | enzyme | organism
-help:	
-help:	<org> = KEGG organism code or T number
-help:	
-help:	
-help:	type#2
-help:	
-help:	param = \"<database>/<org>\"
-help:	<database> = pathway | module
-help:	<org> = KEGG organism code
-help:	
-help:	type#3
-help:	
-help:	param = \"<dbentries>\"
-help:	<dbentries> = KEGG database entries involving the following databases: pathway | brite | module | disease | drug | environ | ko | genome |
-help:					<org> | compound | glycan | reaction | rpair | rclass | enzyme | organism
-help:	
-help:	return a List
-				)
-				
-				if param == :help
-					puts help
-					return {}
-				else
-					
-					begin
-						doc = URI.parse("#{@@baseuri}/list/#{param.to_s}").read().strip()
-						result = {}
-						
-						doc.each do |line|
-							dat = line.chomp.split(/\t/, 2)
-							result[dat[0]] = dat[1]
-						end
-						
-						return result
-					rescue OpenURI::HTTPError
-						raise Sylfy::Service::DataNotFound, "Data not found."
+				begin
+					doc = URI.parse("#{Sylfy::Service::KEGGREST::BASEURI}/list/#{database.to_s}#{org ? "/#{org}" : ""}").read.strip
+					result = {}
+					doc.split(/\n/).each do |line|
+						dat = line.chomp.split(/\t/, 2)
+						result[dat[0]] = dat[1]
 					end
 					
+					return result
+				rescue OpenURI::HTTPError
+					raise Sylfy::Service::DataNotFound, "Data not found."
 				end
 				
 			end

data/lib/sylfy/service/lipidmaprest.rb

@@ -0,0 +1,228 @@
+# 
+# UniSysDB library in Ruby
+# Copyright (C) 2012
+#
+# @author Natapol Pornputtapong <natapol@chalmers.se>
+#
+require 'open-uri'
+require 'net/http'
+
+module Sylfy
+	
+	module Service
+		
+		
+		module LipidMap
+			module REST
+				@@baseuri = "http://www.lipidmaps.org/data"
+				
+				module_function
+				# Method for setting inchi data member
+				#
+				# == Parameters:
+				# id::
+				#   LipidMap id to search
+				# outputtype::
+				#   TSV, CSV, SDF or MDLMOL [defualt: TSV]
+				
+				def lmsdRecord(id, outputtype = "TSV")
+				  
+					uri = "#{@@baseuri}/LMSDRecord.php?LMID=#{id}"
+					uri += "&Mode=File"
+					uri += "&OutputType=#{["TSV", "CSV", "SDF"].include?(outputtype) ? outputtype : 'TSV'}"
+					uri += "&OutputDelimiter=Comma&OutputQuote=Yes&OutputColumnHeader=Yes"
+					
+					begin
+						doc = URI.parse(uri).read().strip()
+						case outputtype
+						when 'TSV', nil
+							return CSV.parse(doc, :headers => true, :col_sep => "\t")
+						when 'CSV'
+							return CSV.parse(doc, :headers => true)
+						else
+							return doc
+						end
+					rescue OpenURI::HTTPError
+						return nil
+					end
+					
+				end
+				
+				# Method for setting inchi data member
+				#
+				# == Parameters:
+				# str::
+				#   search string
+				# mode::
+				#   Search mode ProcessStrSearch, ProcessTextSearch or ProcessTextOntologySearch
+				# outputtype::
+				#   TSV, CSV, SDF or MDLMOL [defualt: TSV]
+				
+				def lmsdSearch(str, mode, outputtype = "TSV")
+				  
+					uri = "#{@@baseuri}/structure/LMSDSearch.php?#{str}&OutputMode=File&Mode=#{mode}"
+					uri += "&OutputType=#{["TSV", "CSV", "SDF"].include?(outputtype) ? outputtype : 'TSV'}"
+					uri += "&OutputDelimiter=Comma&OutputQuote=Yes&OutputColumnHeader=Yes"
+					
+					begin
+						doc = URI.parse(uri).read().strip()
+						case outputtype
+						when 'TSV', nil
+							return CSV.parse(doc, :headers => true, :col_sep => "\t")
+						when 'CSV'
+							return CSV.parse(doc, :headers => true)
+						else
+							return doc
+						end
+					rescue OpenURI::HTTPError
+						return nil
+					end
+					
+				end
+				
+				private :lmsdSearch
+				
+				# Method for setting inchi data member
+				#
+				# == Parameters:
+				# string::
+				#   search string
+				# field::
+				#   field type: LMID, Name, Formula, SMILESString [defualt: LMID]
+				# exactMatch::
+				#   Boolean [defualt: yes]
+				# outputtype::
+				#   TSV, CSV, SDF or MDLMOL [defualt: TSV]
+				
+				def lmsdStrSearch(string, field = 'LMID', exactMatch = true, outputtype = "TSV")
+				  
+					searchType = exactMatch ? 'ExactMatch' : 'SubStructure'
+					
+					if ['LMID', 'Name', 'Formula', 'SMILESString'].include?(field)
+						return lmsdSearch("#{field}=#{URI.encode(string)}&SearchType=#{searchType}", "ProcessStrSearch", outputtype)
+					else
+						raise Sylfy::Service::ParameterError, "#{field} field does not in service."
+					end
+				end
+				
+				# Method for setting inchi data member
+				#
+				# == Parameters:
+				# hash::
+				#   searh hash
+				#   {
+				#     :LMID	=> LM_ID,
+				#     :Name	=> part of any common name,
+				#     :Formula => formula,
+				#     :ExactMass => positive number,
+				#     :ExactMassOffSet => positive number,
+				#     :CoreClass => integer [1 to 8],
+				#       1=Fatty Acyls [FA] 
+				#       2=Glycerolipids [GL] 
+				#       3=Glycerophospholipids [GP] 
+				#       4=Sphingolipids [SP] 
+				#       5=Sterol Lipids [ST] 
+				#       6=Prenol Lipids [PR] 
+				#       7=Saccharolipids [SL] 
+				#       8=Polyketides [PK]
+				#     :MainClass => integer,
+				#       e.g.  FA10=101
+				#             FA11=111
+				#             ST05=505
+				#     :SubClass => integer
+				#       e.g.  ST0402=50402
+				#   }
+				#   
+				# outputtype::
+				#   TSV, CSV, SDF or MDLMOL [defualt: TSV]
+				
+				def lmsdTxtSearch(hash, outputtype = "TSV")
+				  
+					string = ''
+					
+					hash.each_pair do |k, v|
+						string += '&' if string != ''
+						string += "#{k.to_s}=#{v}"
+					end
+					
+					return lmsdSearch(string, "ProcessTextSearch", outputtype)
+				end
+				
+				# Method for setting inchi data member
+				#
+				# == Parameters:
+				# onto::
+				#   string = "<ontoStr>[,<ontoStr>,<ontoStr>,<ontoStr>]"
+				#   <ontoStr> = <ParamName> <ParamModifier> 
+				#   <ParamName> = carbons, doublebonds, triplebonds, rings, OH, NH2, OOH, ketones, epoxides, COOH, methyls, SH, Br, Cl, F, Methylenes, CHO, OMe, OAc, COOMe, Ester, Ether
+				#   <ParamModifier> = eq, ge, le
+				#   <ParamValue> = integer
+				#   example: "carbons eq 20, triplebonds ge 2" or "carbons eq 10"
+				# option::
+				#   {
+				#     :OutputType => TSV, CSV, SDF [defualt: TSV],
+				#   }
+				def lmsdOntoSearch(onto, outputtype = "TSV")
+				  
+					paramName = ["carbons", "doublebonds", "triplebonds", "rings", "OH", "NH2", "OOH", "ketones", "epoxides"]
+					paramName = paramName + ["COOH", "methyls", "SH", "Br", "Cl", "F", "Methylenes", "CHO", "OMe", "OAc", "COOMe", "Ester", "Ether"]
+					
+					paramModifier = ["eq", "ge", "le"]
+					
+					i = 1
+					ontostr = ""
+					
+					onto.chomp.gsub(/, /, ',').split(/,/).each do |str|
+						dat = str.split(/ /)
+						if paramName.include?(dat[0]) && paramModifier.include?(dat[1]) && i < 5
+							ontostr += '&' if i > 1
+							ontostr += "OntologyParamName#{i}=#{dat[0]}&OntologyParamModifier#{i}=#{dat[1]}&OntologyParamValue#{i}=#{dat[2].to_i}"
+							i += 1
+						end
+					end
+					
+					if ontostr != ''
+						return lmsdSearch(ontostr, "ProcessTextOntologySearch", outputtype)
+					else
+						raise Sylfy::Service::ParameterError, "ontology string error."
+					end
+				  
+				  
+				end
+				
+				
+				
+			end
+			
+			module_function
+			
+			# Method for setting inchi data member
+			#
+			# == Parameters:
+			# id::
+			#   LipidMap id to search
+			# :id, :dataPrimarySource, :xrefs, :relations :inchi, :formula, :smiles, :inchiKey :names
+			def idbyinchi(inchi)
+				result = []
+				if inchi =~ Sylfy::Pattern::INCHI
+					smile = Rubabel[inchi, :inchi].to_s(:smi)
+					REST::lmsdStrSearch(smile, 'SMILESString').each {|e| result.push(e["LM_ID"])} if smile != nil
+				else
+					raise Sylfy::Service::ParameterError, "#{inchi} is not InChI."
+				end
+				return result
+			end
+		end
+	end
+end
+
+#test = File.open('tmp.mol', 'w')
+#test.write(Unisys::Service::RESTLipidMap.lmsdRecord('LMFA08040013', {:OutputType => 'MDLMOL'}))
+#test.close
+#output = `obabel -imol tmp.mol -oinchi`
+#p output
+#Unisys::Service::RESTLipidMap.getSmallMolecule('LMFA08040013')
+#puts Unisys::Service::RESTLipidMap.lmsdRecord('LMFA08040013', {:OutputType => 'MDLMOL'})
+#puts Unisys::Service::RESTLipidMap.lmsdSearchStr('LMSL02000001')
+#puts Unisys::Service::RESTLipidMap.lmsdSearchTxt({:CoreClass => 1, :MainClass => 101, :SubClass => 10101})
+#puts Unisys::Service::RESTLipidMap.lmsdSearchOnto('carbons eq 20')