exodb 0.1.2 → 0.1.3

data/lib/exodb/datamodel/generef.rb

@@ -0,0 +1,193 @@
+# 
+# Exodb
+# Copyright (C) 2014
+#
+# author: Natapol Pornputtapong <natapol.por@gmail.com>
+#
+# Documentation: Natapol Pornputtapong (RDoc'd and embellished by William Webber)
+#
+
+# raise "Please, use ruby 1.9.0 or later." if RUBY_VERSION < "1.9.0"
+
+module Exodb
+	
+	class Chrref < Reference
+		
+		include Exodb::GeneLocationField
+		
+		index({oid: 1}, background: true)
+		
+	end
+	
+	class Generef < Reference
+		
+		include Exodb::GeneLocationField
+		
+		@@expanding = 500
+		
+		field :psuedo,				type: Boolean
+		
+		index({sequence: 'text'}, background: true)
+		index({oid: 1, psudo: 1, chrrefseq: 1, assembly: 1}, background: true)
+		
+		has_many :genes
+		embeds_many :isoforms
+		embeds_many :occurrents
+		
+		validates_format_of :chrrefseq, with: /\A(urn:miriam:refseq)/
+		
+		#oid = "chrrefseq:start..stop"
+		
+		def self.expanding
+			return @@expanding
+		end
+		
+		# Download gene symbol from HGNC service
+		#
+		def dl_symbol!
+			
+			baseuri = "http://rest.genenames.org/search"
+			
+			
+			query = ""
+			
+			if self.get_xref('urn:miriam:refseq')
+				query = "#{baseuri}/refseq_accession/#{self.chrrefseq.id.split('.')[0]}"
+			elsif self.get_xref('urn:miriam:ncbigene')
+				query = ""
+			end
+			
+			if !query.empty?
+				response = JSON.parse(open(query, 'Accept' => 'application/json').read)['response']
+				if !response['docs'].empty?
+					
+					response['docs'].each do |e|
+						self.add_to_set(:xrefs, "urn:miriam:hgnc:#{e["hgnc_id"]}")
+						self.add_to_set(:xrefs, "urn:miriam:hgnc.symbol:#{e["symbol"]}")
+					end
+					
+					self.save!
+					
+				end
+				
+			end
+			
+		end
+		
+		# Download incident data from TCGA
+		#
+		def dl_occurrent!
+			
+			if self.get_xref('urn:miriam:hgnc.symbol')
+				
+				cancerstudies = []
+				
+				open("http://www.cbioportal.org/public-portal/webservice.do?cmd=getCancerStudies") {|f|
+					f.each_line {|line| cancerstudies.push(line.chomp.split("\t")[0])}
+				}
+				
+				occurrents = {}
+				totalcase = {}
+				
+				cancerstudies.each do |study|
+					
+					totalcase[study] = 0 if !totalcase.has_key?(study)
+					
+					open("http://www.cbioportal.org/public-portal/webservice.do?cmd=getCaseLists&cancer_study_id=#{study}") do |f|
+						f.each_line do |line|
+							totalcase[study] += line.chomp.split(/\t/)[4].split(' ').length if line =~ /\tSequenced Tumors\t/
+						end
+					end
+					
+					occurrents[study] = {} if !occurrents.has_key?(study)
+					
+					open("http://www.cbioportal.org/public-portal/webservice.do?cmd=getMutationData&genetic_profile_id=#{study}_mutations&gene_list=#{self.get_xref('urn:miriam:hgnc.symbol').id}") do |f|
+						f.each_line do |line|
+							dat = line.chomp.split(/\t/)
+							
+							if dat[5] == 'Missense_Mutation'
+								
+								occurrents[study][dat[7].split(/(\d+)/)[1]] = [] if !occurrents[study].has_key?(dat[7].split(/(\d+)/)[1])
+								occurrents[study][dat[7].split(/(\d+)/)[1]].push(dat[2])
+								
+							end
+							
+						end
+						
+						
+					end
+				end
+				
+				self.occurrents.clear if self.occurrents
+				
+				occurrents.each_pair do |cancertype, v|
+					v.each_pair do |position, occur|
+						self.occurrents << Occurrent.new({cancertype: cancertype, position: position, occur: occur.uniq.sort, casenumber: totalcase[cancertype]})
+					end
+				end
+				
+				self.save!
+				
+			end
+			
+		end
+		
+		# return longest splice of this gene
+		def longest_splice()
+			length = 0
+			longest = nil
+			self.isoforms.each do |e|
+				
+				if e.prot_len > length
+					length = e.prot_len
+					longest = e
+				end
+				
+			end
+			
+			return longest
+		end
+		
+		alias_method :longest_isoform, :longest_splice
+		
+		# Check that this gene has any splice variant
+		#
+		# @return [Boolean] true if has any splices
+		def has_splices?
+			return self.isoforms.exists?
+		end
+		
+		alias_method :has_isoforms?, :has_splices?
+		
+		# Check if Generef has sequence
+		#
+		# @return [Boolean] Return true if there is a sequence
+		def has_sequence?()
+			return self[:sequence] ? true : false
+		end
+		
+		# Check if Generef can translate
+		#
+		# @return [Boolean] Return true if this can be translate
+		def can_translated?()
+			return self.has_sequence? && self.has_splices? && self.longest_splice != nil ? true : false
+		end
+		
+		# Get gene symbol
+		#
+		# @return [String] Return gene symbol or any id from xrefs or 'nosymbol'
+		def symbol
+			
+			if self.get_xref('urn:miriam:hgnc.symbol')
+				return self.get_xref('urn:miriam:hgnc.symbol').id
+			elsif self.xrefs && !self.xrefs.empty?
+				return self.xrefs.sort[0].id
+			else
+				return 'nosymbol'
+			end
+			
+		end
+		
+	end
+	
+end

data/lib/exodb/datamodel/isoform.rb

@@ -0,0 +1,237 @@
+# 
+# Exodb
+# Copyright (C) 2014
+#
+# author: Natapol Pornputtapong <natapol.por@gmail.com>
+#
+# Documentation: Natapol Pornputtapong (RDoc'd and embellished by William Webber)
+#
+
+# raise "Please, use ruby 1.9.0 or later." if RUBY_VERSION < "1.9.0"
+
+module Exodb
+    
+	class Isoform
+		
+		include Mongoid::Document
+		
+		include Exodb::XrefsField
+		
+		field :exon,				type: Array # array of [start, stop]
+		field :cds,					type: Array # array of [start, stop]
+		
+		embedded_in :generef
+		
+		# generate genbank style location string
+		#
+		# @param [Symbol] field to generate [:exon, :cds]
+		# @param [Symbol] type to generate [:rel, :abs, :trela]
+		#
+		# @return [String] a location string
+		def genbank_loc(field, type)
+			
+			reducer = case type.to_sym
+			when :rel
+				self.generef[:start] - 1
+			when :trela
+				self.generef[:seqstart] - 1
+			else
+				0
+			end
+			
+			loc = []
+			
+			self[field.to_sym].each do |pos|
+				loc.push("#{pos[0] - reducer}..#{pos[1] - reducer}")
+			end
+			
+			return self.generef.strand == '+' ? "join(#{loc.join(',')})" : "complement(join(#{loc.join(',')}))"
+		end
+		
+		protected :genbank_loc
+		
+		# Get exon location
+		#
+		# @return [Bio::Locations] of exon
+		def exon_location
+			return Bio::Locations.new(genbank_loc(:exon, :abs))
+		end
+		
+		# Get exon location
+		#
+		# @return [Bio::Locations] of exon
+		def cds_location
+			return Bio::Locations.new(genbank_loc(:cds, :abs))
+		end
+		
+		# Get spliced RNA sequence
+		#
+		# @return [Bio::Sequence] an RNA sequence
+		def mrna_seq
+			return self.return self.generef.whole_seq.splice(genbank_loc(:exon, :trela)).rna
+		end
+		
+		# Get spliced coding region sequence
+		#
+		# @param [Integer] end position to get sequence
+		#
+		# @return [Bio::Sequence] an coding region sequence
+		def cds_seq
+			return self.generef.whole_seq.splice(genbank_loc(:cds, :trela))
+		end
+		
+		# Get spliced protein sequence
+		#
+		# @return [Bio::Sequence] an protein sequence
+		def prot_seq
+			return self.cds_seq().translate
+		end
+		
+		# get length of spliced RNA
+		#
+		# @return [Integer] length of spliced RNA
+		def mrna_len
+			return self.exon_location.length
+		end
+		
+		# get length of protein product
+		#
+		# @return [Integer] length of protein product
+		def prot_len
+			return self.cds_location.length
+		end
+		
+		# Get the codon sequence at the giving position base on position of amino acid
+		#
+		# @param [Integer] codon position
+		# @return [Bio::Sequence] the codon at given position
+		def codon_at(codon_pos)
+			return self.cds_seq.subseq(((codon_pos - 1) * 3) + 1 , ((codon_pos - 1) * 3) + 3)
+		end
+		
+		# convert genomic position to cds position
+		#
+		# @param [Integer] genomic position
+		# @return [Integer] Return all information of cds at given position
+		def genomic2cds_pos(pos)
+			cds_location.relative(pos)
+		end
+		
+		# convert genomic position to cds position
+		#
+		# @param [Integer] genomic position
+		# @return [Array] Return [codon and position in codon]
+		def genomic2prot_pos(pos)
+			cds_location.relative(pos, :aa)
+		end
+		
+		# get splice related position 1-3 base into exon and 3-8 base into intron
+		#
+		# @return [Array] Return position of splice related site in array of [start, stop]
+		# SPL-D		splice_donor_variant		SO:0001575
+		# SPL-A		splice_acceptor_variant		SO:0001574
+		# SPL-R		splice_region_variant		SO:0001630
+		def splice_rel_location
+			
+			if @splice_location == nil
+				
+				results = {}
+				right = nil
+				
+				self[:exon].each do |exon|
+					if !right.blank?
+						if self.generef.strand == '+'
+							results[right + 1] = 'SPL-D'
+							results[right + 2] = 'SPL-D'
+							results[exon[0] - 1] = 'SPL-A'
+							results[exon[0] - 2] = 'SPL-A'
+						else
+							results[right + 1] = 'SPL-A'
+							results[right + 2] = 'SPL-A'
+							results[exon[0] - 1] = 'SPL-D'
+							results[exon[0] - 2] = 'SPL-D'
+						end
+						
+						Range.new(right - 2, right).each {|e| results[e] = 'SPL-R'}
+						Range.new(right + 3, right + 8).each {|e| results[e] = 'SPL-R'}
+						Range.new(exon[0], exon[0] + 2).each {|e| results[e] = 'SPL-R'}
+						Range.new(exon[0] - 3, exon[0] - 8).each {|e| results[e] = 'SPL-R'}
+						
+						right = exon[1]
+					else
+						right = exon[1]
+					end
+				end
+				
+				@splice_location = results
+				return results
+			else
+				return @splice_location
+			end
+			
+		end
+		
+		def utr5location
+			return Bio::Locations.new(self.generef.strand == '+' ? "#{self[:exon][0][0]}..#{self[:cds][0][0] - 1}" : "complement(#{self[:cds][-1][1] + 1}..#{self[:exon][-1][1]})")
+		end
+		
+		def utr3location
+			return Bio::Locations.new(self.generef.strand == '+' ? "#{self[:cds][-1][1] + 1}..#{self[:exon][-1][1]}" : "complement(#{self[:exon][0][0]}..#{self[:cds][0][0] - 1})")
+		end
+		
+		# predict efect of variants
+		#
+		# @param [array] list of variants
+		#
+		# @return [array] list of predicted effect
+		# SPL-D		splice_donor_variant		SO:0001575
+		# SPL-A		splice_acceptor_variant		SO:0001574
+		# SPL-R		splice_region_variant		SO:0001630
+		# COD-SS
+		# COD-SN
+		# COD-DS
+		# COD-DN
+		# COD-IS
+		# COD-IN
+		# UTR-3
+		# UTR-5
+		# INT
+		# INI
+		# STO-L
+		# STO-G
+		# PRO
+		# TFB
+		def effect_pred(position, alt)
+			
+			results = []
+			
+			prot_pos = genomic2prot_pos(position)
+			
+			if prot_pos
+				case alt
+				when /^-([ATCG])+/
+					results.push({vartype: $1.length % 3 == 0 ? 'COD-DS' : 'COD-DN'})
+				when /^\+([ATCG])+/
+					results.push({vartype: ($1.length - 1) % 3 == 0 ? 'COD-IS' : 'COD-IN'})
+				else
+					results.push({vartype: 'COD-SS'})
+				end
+			else
+				#case
+				#when test
+				#	#code
+				#when test
+				#	#code
+				#else
+				#	results.push({vartype: 'INT'})
+				#end
+			end
+			
+			results.push({vartype: splice_rel_location[position]}) if splice_rel_location.has_key?(position)
+			
+			return results
+			
+		end
+	end
+	
+end

data/lib/exodb/datamodel/reference.rb

@@ -28,7 +28,7 @@ module Exodb
 		PATTERN = /(?<gene>[A-Z0-9]+)-?(?<position>[0-9,]*|[is]?)(?<to>[A-Z=]*)/
 		SILENTSIGN = '='
 		
-		include Exodb::GenomeLocationField
+		include Exodb::VarLocationField
 		
 		field :reference,			type: String
 		field :alternate,			type: String
@@ -48,339 +48,35 @@ module Exodb
 		
 	end
 	
-	class Generef < Reference
-		
-		include Exodb::GenomeLocationField
-		
-		field :sequence,			type: String
-		field :chrrefseq,			type: String # refseq id of chromomose
-		field :psuedo,				type: Boolean
-		field :genomeref,			type: String
-		
-		index({sequence: 'text'}, background: true)
-		
-		has_many :genes
-		embeds_many :isoforms
-		embeds_many :occurrents
-		
-		validates_format_of :chrrefseq, with: /\A(urn:miriam:refseq)/
-		
-		
-		# Download sequence from web service please use by caution. NCBI will block scamming sequest
-		#
-		def dl_seq!
-			
-			case self.chrrefseq
-			when /\Aurn:miriam:refseq:/
-				self.sequence = Bio::FastaFormat.new(Bio::NCBI::REST.efetch(self.chrrefseq.split(':', 4), {"db"=>"nucleotide", "rettype"=>"fasta", "retmode"=>"text", "seq_start"=>self.start, "seq_stop"=>self.end})).seq
-			end
-			
-			self.save!
-			
-		end
-		
-		# Download gene symbol from HGNC service
-		#
-		def dl_symbol!
-			
-			baseuri = "http://rest.genenames.org/search"
-			
-			
-			query = ""
-			
-			if self.get_xref('urn:miriam:refseq')
-				query = "#{baseuri}/refseq_accession/#{self.chrrefseq.id.split('.')[0]}"
-			elsif self.get_xref('urn:miriam:ncbigene')
-				query = ""
-			end
-			
-			if !query.empty?
-				response = JSON.parse(open(query, 'Accept' => 'application/json').read)['response']
-				if !response['docs'].empty?
-					
-					response['docs'].each do |e|
-						self.add_to_set(:xrefs, "urn:miriam:hgnc:#{e["hgnc_id"]}")
-						self.add_to_set(:xrefs, "urn:miriam:hgnc.symbol:#{e["symbol"]}")
-					end
-					
-					self.save!
-					
-				end
-				
-			end
-			
-		end
-		
-		# Download incident data from TCGA
-		#
-		def dl_occurrent!
-			
-			if self.get_xref('urn:miriam:hgnc.symbol')
-				
-				cancerstudies = []
-				
-				open("http://www.cbioportal.org/public-portal/webservice.do?cmd=getCancerStudies") {|f|
-					f.each_line {|line| cancerstudies.push(line.chomp.split("\t")[0])}
-				}
-				
-				occurrents = {}
-				totalcase = {}
-				
-				cancerstudies.each do |study|
-					
-					totalcase[study] = 0 if !totalcase.has_key?(study)
-					
-					open("http://www.cbioportal.org/public-portal/webservice.do?cmd=getCaseLists&cancer_study_id=#{study}") do |f|
-						f.each_line do |line|
-							totalcase[study] += line.chomp.split(/\t/)[4].split(' ').length if line =~ /\tSequenced Tumors\t/
-						end
-					end
-					
-					occurrents[study] = {} if !occurrents.has_key?(study)
-					
-					open("http://www.cbioportal.org/public-portal/webservice.do?cmd=getMutationData&genetic_profile_id=#{study}_mutations&gene_list=#{self.get_xref('urn:miriam:hgnc.symbol').id}") do |f|
-						f.each_line do |line|
-							dat = line.chomp.split(/\t/)
-							
-							if dat[5] == 'Missense_Mutation'
-								
-								occurrents[study][dat[7].split(/(\d+)/)[1]] = [] if !occurrents[study].has_key?(dat[7].split(/(\d+)/)[1])
-								occurrents[study][dat[7].split(/(\d+)/)[1]].push(dat[2])
-								
-							end
-							
-						end
-						
-						
-					end
-				end
-				
-				self.occurrents.clear if self.occurrents
-				
-				occurrents.each_pair do |cancertype, v|
-					v.each_pair do |position, occur|
-						self.occurrents << Occurrent.new({cancertype: cancertype, position: position, occur: occur.uniq.sort, casenumber: totalcase[cancertype]})
-					end
-				end
-				
-				self.save!
-				
-			end
-			
-		end
-		
-		# return sequence as Bio::Sequence object
-		#
-		# @return [Bio::Sequence] the contents reversed lexically
-		def to_seq
-			return self.sequence ? Bio::Sequence.auto(self.sequence) : Bio::Sequence.auto("")
-		end
-		
-		# return longest splice of this gene
-		def longest_splice()
-			length = 0
-			longest = nil
-			self.isoforms.each do |e|
-				
-				if e.prot_len > length
-					length = e.prot_len
-					longest = e
-				end
-				
-			end
-			
-			return longest
-		end
-		
-		# Check that this gene has any splice variant
-		#
-		# @return [Boolean] true if has any splices
-		def has_splices?
-			return self.isoforms.exists?
-		end
-		
-		# Check if Generef has sequence
-		#
-		# @return [Boolean] Return true if there is a sequence
-		def has_sequence?()
-			return self[:sequence] ? true : false
-		end
-		
-		# Check if Generef can translate
-		#
-		# @return [Boolean] Return true if this can be translate
-		def can_translated?()
-			return self.has_sequence? && self.has_splices? && self.longest_splice != nil ? true : false
-		end
-		
-		# Get gene symbol
-		#
-		# @return [String] Return gene symbol or any id from xrefs or 'nosymbol'
-		def symbol
-			
-			if self.get_xref('urn:miriam:hgnc.symbol')
-				return self.get_xref('urn:miriam:hgnc.symbol').id
-			elsif self.xrefs && !self.xrefs.empty?
-				return self.xrefs.sort[0].id
-			else
-				return 'nosymbol'
-			end
-			
-		end
-		
-	end
-	
-	class Isoform
+	class Mapping
 		
 		include Mongoid::Document
 		
-		include Exodb::XrefsField
-		
-		field :exon,				type: Array
-		field :cds,					type: Array
-		
-		embedded_in :generef
-		
-		# join exon or cds position into a string
-		#
-		# @param [Array] input array exon or cds
-		# @param [Interger] Position to stop positive value for forward read negative value for complement
-		#
-		# @return [String] a string in start..end,start..end,...
-		def get_join_str(arr, position = 0)
-			
-			reducer = self.generef.start - 1
-			tmparr = []
-			found = false
-			
-			if position > 0
-				add = true
-				arr.each do |e|
-					
-					if e[0] <= position && position <= e[1]
-						tmparr.push([e[0], position])
-						add = false
-						found = true
-					else
-						tmparr.push(e) if add
-					end
-					
-				end
-			elsif position < 0
-				position = position.abs
-				add = false
-				arr.each do |e|
-					
-					if e[0] <= position && position <= e[1]
-						tmparr.push([position, e[1]])
-						add = true
-						found = true
-					else
-						tmparr.push(e) if add
-					end
-					
-				end
-			else
-				tmparr = arr
-			end
-			
-			tmparr = [] if !found && position != 0
-			str = []
-			
-			tmparr.each do |e|
-				str.push("#{e[0] - reducer}..#{e[1] - reducer}")
-			end
-			return str.join(',')
-			
-		end
-		
-		def get_exon_join(position = 0)
-			get_join_str(self[:exon], position)
-		end
-		
-		def get_cds_join(position = 0)
-			get_join_str(self[:cds], position)
-		end
-		
-		# Get spliced DNA sequence
-		#
-		# @return [Bio::Sequence] an DNA sequence
-		def get_dna_seq
-			parent =  self.generef
-			return parent.strand == '+' ? parent.to_seq.splicing("join(#{self.get_exon_join})") : parent.to_seq.splicing("complement(join(#{self.get_exon_join}))")
-		end
-		
-		# Get spliced RNA sequence
-		#
-		# @return [Bio::Sequence] an RNA sequence
-		def get_mrna_seq
-			parent =  self.generef
-			return parent.strand == '+' ? parent.to_seq.splicing("join(#{self.get_exon_join})").rna : parent.to_seq.splicing("complement(join(#{self.get_exon_join}))").rna
-		end
-		
-		# Get spliced coding region sequence
-		#
-		# @param [Integer] end position to get sequence
-		#
-		# @return [Bio::Sequence] an coding region sequence
-		def get_cds_seq(position = 0)
-			
-			parent =  self.generef
-			if parent.strand == '+'
-				join = self.get_cds_join(position)
-				return !join.empty? ? parent.to_seq.splicing("join(#{join})") : ""
-			else
-				join = self.get_cds_join(-position)
-				return !join.empty? ? parent.to_seq.splicing("join(#{join})") : ""
+		field :chr,					type: String
+		field :start,				type: Integer
+		field :stop,				type: Integer
+		field :tchr,				type: String
+		field :tstart,				type: Integer
+		field :tstop,				type: Integer
+		field :coeff,				type: Integer # coefficient for conversion
+		field :from,				type: String # from assembly version
+		field :to,					type: String # to assembly version
+		
+		index({oid: 1, chr: 1, start: 1, stop: 1, from: 1, to: 1}, background: true)
+		
+		def self.convert(locstr, target = Exodb::LATESTASSEMBLY)
+			begin
+				query = parse_locstr(locstr)
+				return self.where({from: query['assembly'], to: target, :start.lte => query[:pos], :stop.gte => query[:pos]}).first.convert(query[:pos])
+			rescue
+				
 			end
-			
 		end
 		
-		# Get spliced protein sequence
-		#
-		# @return [Bio::Sequence] an protein sequence
-		def get_prot_seq
-			parent =  self.generef
-			return parent.strand == '+' ? parent.to_seq.splicing("join(#{self.get_cds_join})").translate : parent.to_seq.splicing("complement(join(#{self.get_cds_join}))").translate
-		end
-		
-		# get length of spliced RNA
-		#
-		# @return [Integer] length of spliced RNA
-		def rna_len
-			return self.get_mrna_seq.length
-		end
-		
-		# get length of protein product
-		#
-		# @return [Integer] length of protein product
-		def prot_len
-			return self.get_prot_seq.length
-		end
-		
-		# Get the codon sequence at the giving position base on position of amino acid
-		#
-		# @param [Integer] codon position
-		# @return [Bio::Sequence] the codon at given position
-		def get_codon(codon_pos)
-			return self.get_cds_seq().subseq(((codon_pos - 1) * 3) + 1 , ((codon_pos - 1) * 3) + 3)
-		end
-		
-		# convert genomic position to codon position
-		#
-		# @param [Integer] genomic position
-		# @return [Array] Return all information of codon at given position
-		def get_prot_pos(pos)
-			
-			seqlen = self.get_cds_seq(pos).length
-			if seqlen != 0
-				return [((seqlen - 1) / 3) + 1, ((seqlen - 1) % 3) + 1]
-			else
-				return []
-			end
-			
+		def convert(pos)
+			return  "#{self[:tchr]}:#{(coeff > 0 ? tstart : tstop) + ((pos - self[:start]) * coeff)}:#{to}"
 		end
 		
 	end
 	
-end
+end