full_lengther_next 0.0.1

data/bin/make_user_db.rb

@@ -0,0 +1,144 @@
+#!/usr/bin/env ruby
+
+# 15-2-2011 Noe Fernandez-Pozo
+# Script to create your own Full-LengtherNext User database.
+
+require 'net/ftp'
+
+#receive one argument or fail
+if (ARGV.size != 2)
+	
+	puts "incorrect number of arguments, you need a taxonomic group like 'Coniferopsida', you can search it in 'http://www.ncbi.nlm.nih.gov/Taxonomy/'
+and a UniProt taxonomic group from this list:
+
+	fungi
+	human
+	invertebrates
+	mammals
+	plants
+	rodents
+	vertebrates
+
+mode of use: ruby make_user_db.rb coniferopsida plants\n\n"
+
+  Process.exit(-1);
+end
+
+(my_group,uniprot_group)=ARGV
+
+################################################### Functions
+
+def filter_incomplete_seqs(output_file,file_name, my_group)
+	
+	puts " filtering sequences"
+	
+	# UniProtKB fragments with FT NON_CONS and FT NON_TER features.
+	# 
+	#     * FT NON_TER: The residue at an extremity of the sequence is not the terminal residue. If applied to position 1, this signifies that the first position is not the N-terminus of the complete molecule. If applied to the last position, it means that this position is not the C-terminus of the complete molecule. There is no description field for this key. Examples of NON_TER key feature lines:
+	#       FT NON_TER 1 1
+	#       FT NON_TER 29 29
+	#     * FT NON_CONS: Non-consecutive residues. Indicates that two residues in a sequence are not consecutive and that there are a number of unreported or missing residues between them. Example of a NON_CONS key feature line:
+	#       FT NON_CONS 1683 1684
+	# 
+	# NON_CONS fragments are not indicated as non-consecutive in InterPro and being non-consecutive the match to methods may be incorrect if the method spans the 'break'.
+	
+	newseq=false
+	print_seq=false
+	incomplete=false
+	id=''
+	description = ''
+	organism_name = ''
+	seq = ''
+	organelle = ''
+	
+	File.open(file_name).each_line do |line|
+		if (newseq == false)
+			if (line =~ /^AC\s+(\w+);/)
+				id=$1
+				newseq = true
+				description = ''
+				organism_name = ''
+				seq = ''
+				print_seq = false
+				incomplete = false
+				organelle = ''
+			end
+		else
+			if (line =~ /^DE\s+(.+)\;*/)
+				if (description == '')
+					description = $1
+					description.sub!(/RecName: Full=/,'sp=')
+					description.sub!(/SubName: Full=/,'tr=')
+				end
+				if (line =~ /Flags: Fragment/)
+					# puts "#{id} #{line}"
+					incomplete = true
+				end
+			elsif (line =~ /^OS\s+(.+)/)
+				organism_name = $1
+			elsif (line =~ /^OG\s+(.+)/)
+				organelle = $1
+			elsif (line =~ /^OC\s+[\w\s\;]*#{my_group}/i) && (!incomplete)
+				print_seq=true
+				# puts "#{id}   #{organism_name} print_seq?: #{print_seq}"
+			elsif (line =~ /^FT\s+NON_TER\s+/)
+				print_seq=false
+				# puts "#{id}   NON_TER"
+			elsif (line =~ /^FT\s+NON_CONS\s+(\d+)\s+/)
+				print_seq=false
+				# puts "#{id}   NON_CONS"
+			elsif (line =~ /^\s+([\w\s]+)/)
+				seq += $1
+			elsif (line =~ /^\/\//)
+				seq.gsub!(/\s*/,'')
+				if (seq !~ /^M/i)
+					print_seq=false
+				end
+				newseq = false
+				
+				if (print_seq)
+					output_file.puts ">#{id} #{description} #{organism_name} #{organelle}\n#{seq}"
+				end
+			end
+		end
+	end
+end
+
+########################################################
+##  MAIN
+########################################################
+
+ROOT_PATH=File.dirname(__FILE__)
+
+# $: << File.expand_path(File.join(ROOT_PATH, "classes"))
+
+# load gem path, only to test locally
+# $: << File.expand_path('~/progs/ruby/gems/full_lengther_next/lib')
+
+require 'full_lengther_next'
+
+if ENV['BLASTDB'] && File.exists?(ENV['BLASTDB'])
+  formatted_db_path = ENV['BLASTDB']
+else # otherwise use ROOTPATH + DB
+  formatted_db_path = File.expand_path(File.join(ROOT_PATH, "blast_dbs"))
+end
+
+ENV['BLASTDB']=formatted_db_path
+
+
+if !File.exists?(File.join(ENV['BLASTDB'], my_group))
+	Dir.mkdir("blast_dbs/#{my_group}")
+end
+
+output_file_path=File.join(ENV['BLASTDB'],my_group,my_group+".fasta")
+
+output_file = File.new(output_file_path, "w")
+
+filter_incomplete_seqs(output_file, File.join(ENV['BLASTDB'], "uniprot_sprot_#{uniprot_group}.dat"), my_group)
+filter_incomplete_seqs(output_file, File.join(ENV['BLASTDB'], "uniprot_trembl_#{uniprot_group}.dat"), my_group)
+
+output_file.close
+
+`makeblastdb -in #{output_file_path} -dbtype 'prot' -parse_seqids`
+
+puts "make_user_db.rb has finished"

data/lib/full_lengther_next.rb

@@ -0,0 +1,13 @@
+$:.unshift(File.dirname(__FILE__)) unless
+  $:.include?(File.dirname(__FILE__)) || $:.include?(File.expand_path(File.dirname(__FILE__)))
+
+ROOT_PATH=File.join(File.dirname(__FILE__),'full_lengther_next')
+
+$: << File.expand_path(File.join(ROOT_PATH, 'classes'))
+
+
+module FullLengtherNext
+  VERSION = '0.0.1'
+  
+  FULLLENGHTER_VERSION = VERSION
+end

data/lib/full_lengther_next/classes/common_functions.rb

@@ -0,0 +1,94 @@
+
+module CommonFunctions
+	
+	def contenidos_en_prot(hit, full_prot, q)
+		
+		is_ok = false
+		q_index_start = 9999
+		fr_index_start = 0
+		min_index_start = 9999
+		aas_parecidos = 0
+		masked_x = 0
+		suma_fragments = 0
+
+		masked_x = hit.q_seq.count('X')
+		masked_x = masked_x + hit.q_seq.count('-')
+		
+		full_prot = full_prot.gsub(/[\-Xx]+/,'')
+		compare_prot = hit.q_seq.gsub(/[\-Xx]+/,'-')
+		fragments_array = compare_prot.split(/\-+/)
+		
+		fragments_array.each do |seq|
+			# puts "seq: #{seq}\nfull_prot: #{full_prot}"
+			simliar_fragment = full_prot.lcs(seq)
+			suma_fragments += simliar_fragment.length
+			
+			fr_index_start = full_prot.index(simliar_fragment)
+
+			if (q_index_start == 9999)
+				q_index_start = fr_index_start
+			end
+			full_prot = full_prot[(fr_index_start + simliar_fragment.length)..full_prot.length]
+		end
+		
+		simliar_fragment = full_prot.lcs(compare_prot)
+		
+		# if ($verbose)
+			# puts "#{q.query_def}-------------------------------------#{suma_fragments} de #{compare_prot.length}"
+			# puts "#{q.query_def}-------------------------------------#{suma_fragments + masked_x} >= #{compare_prot.length * 0.7}"
+			# puts "\nfull: #{full_prot}\ncomp: #{compare_prot}\nsimliar_fragment: #{simliar_fragment}"
+		# end
+		
+		if (suma_fragments + masked_x >= compare_prot.length * 0.7)
+			is_ok = true
+			# puts "OK -- encontramos suficiente similitud entre query y subject -- OK"
+		else
+			is_ok = false
+			# puts "\nfull: #{full_prot}\ncomp: #{compare_prot}"
+			# puts "Warning!: no match comparing proteins"
+		end
+
+		min_index_start = [min_index_start, q_index_start].min
+		
+		if (min_index_start == 9999)
+			min_index_start = 0
+		end
+		
+		return [is_ok, min_index_start]
+	end
+
+
+
+
+	def reverse_seq(query_fasta, h_qframe, h_qstart, h_qend)
+
+		q_frame = -h_qframe.to_i
+
+		q_beg = query_fasta.length - h_qend - 1
+		q_end = query_fasta.length - h_qstart - 1
+
+		query_fasta = query_fasta.complementary_dna
+
+		# el qend y el qstart estan al reves porque cuando la seq tiene frame negativo el blast los pone al reves
+		return [query_fasta, q_frame, q_beg, q_end]
+	end
+
+
+
+	def corrige_frame(ref_frame,ref_start,ref_end)
+
+		if (ref_frame.abs == 2)
+			ref_start = ref_start + 1
+			ref_end = ref_end + 1
+		elsif (ref_frame.abs == 3)
+			ref_start = ref_start + 2
+			ref_end = ref_end + 2
+		end
+		
+		return [ref_start,ref_end]
+
+	end
+
+
+
+end

data/lib/full_lengther_next/classes/fl2_stats.rb

@@ -0,0 +1,222 @@
+
+module Fl2Stats
+	
+	# --------------------------------------------------------------------------------       Main
+	def summary_stats
+		stats_file = File.open('fl2_results/summary_stats.txt', 'w')
+
+		total_seqs = 0
+
+		num1 = annotation_stats(stats_file)
+		num2 = testcode_stats(stats_file)
+
+		total_seqs = num1 + num2
+
+		stats_file.puts "\nInput sequences in your fasta: #{total_seqs}\n\n"
+	end
+
+	# ----------------------------------------------------------------------------------      Functions
+	def stats_my_db(db_name, array)
+
+		if (db_name !~ /^sp_/) && (db_name !~ /^tr_/)
+			array[1] += 1
+		elsif (db_name =~ /^sp_/)
+			array[2] += 1
+		elsif (db_name =~ /^tr_/)
+			array[3] += 1
+		end
+
+		return array
+	end
+
+
+	def annotation_stats(stats_file)
+
+		seqs_number = 0
+		array_of_all_accs = []
+		array_of_complete_accs = []
+		error_1_num = 0
+		
+		seqs_longer_200 = 0
+		seqs_shorter_200 = 0
+		complete_longer_200 = 0
+		complete_shorter_200 = 0
+		
+		seqs_longer_500 = 0
+		seqs_shorter_500 = 0
+		complete_longer_500 = 0
+		complete_shorter_500 = 0
+		
+		complete = [0,0,0,0]
+		putative_complete = [0,0,0,0]
+		c_terminus = [0,0,0,0]
+		putative_c_terminus = [0,0,0,0]
+		n_terminus = [0,0,0,0]
+		putative_n_terminus = [0,0,0,0]
+		internal = [0,0,0,0]
+		cod_seq = [0,0,0,0]
+
+
+		File.open('fl2_results/annotations.txt').each do |line|
+			line.chomp!
+			(name,fasta_length,acc,db_name,status,kk1,kk2,kk3,kk4,kk5,msgs) = line.split("\t")
+
+			if (line !~ /^Query_id\t/)
+				seqs_number += 1
+				array_of_all_accs.push acc
+				# -------------------------------------------------------------------------
+				if (fasta_length.to_i >= 200)
+					seqs_longer_200 += 1
+				else
+					seqs_shorter_200 += 1
+				end
+				if (fasta_length.to_i >= 500)
+					seqs_longer_500 += 1
+				else
+					seqs_shorter_500 += 1
+				end
+				# -------------------------------------------------------------------------
+				if (msgs =~ /ERROR#1/)
+					error_1_num += 1
+				end
+				# -------------------------------------------------------------------------
+				if (status == 'Complete')
+					complete[0] += 1
+					array_of_complete_accs.push acc
+					complete = stats_my_db(db_name, complete)
+					
+					if (fasta_length.to_i >= 200)
+						complete_longer_200 += 1
+					else
+						complete_shorter_200 += 1
+					end
+					
+					if (fasta_length.to_i >= 500)
+						complete_longer_500 += 1
+					else
+						complete_shorter_500 += 1
+					end
+					
+				elsif (status == 'Putative Complete')
+					putative_complete[0] += 1
+					putative_complete = stats_my_db(db_name, putative_complete)
+				elsif (status == 'C-terminus')
+					c_terminus[0] += 1
+					c_terminus = stats_my_db(db_name, c_terminus)
+				elsif (status == 'N-terminus')
+					n_terminus[0] += 1
+					n_terminus = stats_my_db(db_name, n_terminus)
+				elsif (status == 'Putative C-terminus')
+					putative_c_terminus[0] += 1
+					putative_c_terminus = stats_my_db(db_name, putative_c_terminus)
+				elsif (status == 'Putative N-terminus')
+					putative_n_terminus[0] += 1
+					putative_n_terminus = stats_my_db(db_name, putative_n_terminus)
+				elsif (status == 'Internal')
+					internal[0] += 1
+					internal = stats_my_db(db_name, internal)
+				elsif (status == 'Coding Seq')
+					cod_seq[0] += 1
+					cod_seq = stats_my_db(db_name, cod_seq)
+				end
+				# -------------------------------------------------------------------------
+			end
+
+		end
+
+		stats_file.puts "--- Annotation Summary ---"
+		stats_file.puts "\n------------------------------ Summary of sequences found by similarity -----"
+
+		stats_file.puts "\n\tSequences found: #{seqs_number}\t\t(>200: #{seqs_longer_200}, <200: #{seqs_shorter_200})\t(>500: #{seqs_longer_500}, <500: #{seqs_shorter_500})"
+		stats_file.puts "\tDifferent IDs:   #{array_of_all_accs.uniq.count}"
+
+		stats_file.puts "\n\tsequences with sense and antisense hits error: #{error_1_num}"
+		stats_file.puts "\n------------------------------------------------- Full-Length Sequences -----"
+		stats_file.puts "\tComplete Seqs: #{complete[0]} ("+ '%.3f' % (complete[0].to_f/seqs_number.to_f*100) +" %)\t\t(>200: #{complete_longer_200}, <200: #{complete_shorter_200})\t(>500: #{complete_longer_500}, <500: #{complete_shorter_500})"
+		stats_file.puts "\tDifferent IDs: #{array_of_complete_accs.uniq.count} ("+ '%.3f' % (array_of_complete_accs.uniq.count.to_f/seqs_number.to_f*100) +" %)"
+		stats_file.puts "\n\t\tuser_db: #{complete[1]}\n\t\tsp: #{complete[2]}\n\t\ttr: #{complete[3]}"
+		stats_file.puts "-----------------------------------------------------------------------------"
+
+		stats_file.puts "\n\tputative completes: #{putative_complete[0]}\n\t\tuser_db: #{putative_complete[1]}\n\t\tsp: #{putative_complete[2]}\n\t\ttr: #{putative_complete[3]}"
+		stats_file.puts "\n\tn-terminus: #{n_terminus[0]}\n\t\tuser_db: #{n_terminus[1]}\n\t\tsp: #{n_terminus[2]}\n\t\ttr: #{n_terminus[3]}"
+		stats_file.puts "\n\tputative_n_terminus: #{putative_n_terminus[0]}\n\t\tuser_db: #{putative_n_terminus[1]}\n\t\tsp: #{putative_n_terminus[2]}\n\t\ttr: #{putative_n_terminus[3]}"
+		stats_file.puts "\n\tc-terminus: #{c_terminus[0]}\n\t\tuser_db: #{c_terminus[1]}\n\t\tsp: #{c_terminus[2]}\n\t\ttr: #{c_terminus[3]}"
+		stats_file.puts "\n\tputative_c_terminus: #{putative_c_terminus[0]}\n\t\tuser_db: #{putative_c_terminus[1]}\n\t\tsp: #{putative_c_terminus[2]}\n\t\ttr: #{putative_c_terminus[3]}"
+		stats_file.puts "\n\tinternal: #{internal[0]}\n\t\tuser_db: #{internal[1]}\n\t\tsp: #{internal[2]}\n\t\ttr: #{internal[3]}"
+		stats_file.puts "\n\tcoding sequences with unknown status: #{cod_seq[0]}\n\t\tuser_db: #{cod_seq[1]}\n\t\tsp: #{cod_seq[2]}\n\t\ttr: #{cod_seq[3]}"
+
+		return seqs_number
+	end
+
+
+	def testcode_stats(stats_file)
+
+		seqs_number = 0
+		coding = 0
+		putative_coding = 0
+		unknown = 0
+		
+		coding_longer_200 = 0
+		coding_shorter_200 = 0
+		unknown_longer_200 = 0
+		unknown_shorter_200 = 0
+
+		coding_longer_500 = 0
+		coding_shorter_500 = 0
+		unknown_longer_500 = 0
+		unknown_shorter_500 = 0
+
+		File.open('fl2_results/tcode_result.txt').each do |line|
+			line.chomp!
+			(name,fasta_length,acc,db_name,status) = line.split("\t")
+
+			if (line !~ /^Query_id\t/)
+				seqs_number += 1
+
+				if (status == 'coding')
+					coding += 1
+					if (fasta_length.to_i >= 200)
+						coding_longer_200 += 1
+						coding_longer_500 += 1
+					else
+						coding_shorter_200 += 1
+						coding_shorter_500 += 1
+					end
+				elsif (status == 'putative_coding')
+					putative_coding += 1
+				elsif (status == 'unknown')
+					unknown += 1
+					if (fasta_length.to_i >= 200)
+						unknown_longer_200 += 1
+						unknown_longer_500 += 1
+					else
+						unknown_shorter_200 += 1
+						unknown_shorter_500 += 1
+					end
+					
+				end
+
+			end
+
+		end
+
+
+		stats_file.puts "\n--------------------------- Test Code Summary\n\n\ttotal seqs: #{seqs_number}"
+		stats_file.puts "\n\tcoding sequences: #{coding}"
+		stats_file.puts "\t\tlonger than 200 bp: #{coding_longer_200}"
+		stats_file.puts "\t\tshorter than 200 bp: #{coding_shorter_200}"
+		stats_file.puts "\t\tlonger than 500 bp: #{coding_longer_500}"
+		stats_file.puts "\t\tshorter than 500 bp: #{coding_shorter_500}"
+		stats_file.puts "\n\tputative coding sequences: #{putative_coding}\n"
+		stats_file.puts "\n\tunknown: #{unknown} ("+ '%.3f' % (unknown.to_f/seqs_number.to_f*100) +" %)"
+		stats_file.puts "\t\tlonger than 200 bp: #{unknown_longer_200}"
+		stats_file.puts "\t\tshorter than 200 bp: #{unknown_shorter_200}"
+		stats_file.puts "\t\tlonger than 500 bp: #{unknown_longer_500}"
+		stats_file.puts "\t\tshorter than 500 bp: #{unknown_shorter_500}"
+		stats_file.puts "\n\tUnknown sequences have a bad test code score or haven't got an ORF longer than 200 nt"
+		stats_file.puts "---------------------------------------------"
+
+		return seqs_number
+	end
+	
+end