bacterial-annotator 0.4.1 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 12f3850a1e1eb738530d8cdb31c002b3ec34419c
-  data.tar.gz: d765574dd7200b59aecbfe1f7c589d1a1ea15527
+  metadata.gz: befd57ce78f0c186da1553c7372c3aa6faeb9d90
+  data.tar.gz: 5e37d6a7e579a1e9e428deb9864e4a9d5ea9f057
 SHA512:
-  metadata.gz: 1f8f63582170adc0d2efb47dd655364c46822ba600778638226278d07d7b06ec2ae838058a7ee3ceb3b8d9b2668587711995b94c1fcdd7fdb6816f8384283bb1
-  data.tar.gz: 3a210bef0d5c0112add5867997894ce893f20ccdd867b72fa34b4d5a55ec9821e6e878e3e6168917705bdfbdeec12c4c70cc1cdb93e570b650c6eca28b18ee61
+  metadata.gz: a9a9766113cef56ae7ed35749cd5fbc10d746aa82e403596dccd0c5e7946786b136a69e19c74a4cece73549b0f1a8de077a8c106fc0e2310f7b000dc6cbad962
+  data.tar.gz: 00a0c5cf815252fa45ffae194318f730cf69e27dd53643659fb06d2dac131a3de881cbad2595b43df6ba5014be75cef8b337e6710afb46e42b420fdd1cf9b178

data/bin/ba_prodigal

@@ -31,7 +31,7 @@ def installProdigal
   rescue
     abort "Problem installing Prodigal, aborting"
   end
-  
+
 end
 
 

data/bin/bacterial-annotator

@@ -46,22 +46,21 @@ annotate [OPTIONS]
     --force/-f		Force to overwrite the output directory
 
   // Dataset
-    --refgenome/-g        <GBK_ID> 	Provide a Genbank file or a Gbk Accession ID.
-    --guessref		Will guess the best reference genome to use for the annotation.
+    --refgenome/-g	<GBK_ID> 	Provide a Genbank file or a Gbk Accession ID.
+    --guessref	 	Will guess the best reference genome to use for the annotation.
 
-    --remotedb		<remote_database> [nr|refseq|swissprot]
-			  Complete the annotation of remaining CDS with a remote NCBI BLAST
-			  Can be very slow, better to use an external database !
-
-    --externaldb		<proteins fasta_file>
-			  Complete or do the annotation of remaining CDS with this database (a protein fasta file).
+    --externaldb	<proteins fasta_file>
+			  Finish or do a complete annotation with this sequence database (a protein fasta file).
 			  Fasta headers need to look similar to NCBI or EBI fasta headers, ex.:
 			  >gi|385721352|gb|AFI72857.1| NDM-1 [Escherichia coli]
 			  >sp|C7C422|BLAN1_KLEPN Beta-lactamase NDM-1 OS=Klebsiella pneumoniae..
 
   // Other options
-    --pidentity		Minimum percentage identity to incorporate a CDS annotation [default=0.7]
-    --minlength		Minimum contig length for annotation [default=500]
+    --pidentity		<% identity> Minimum percentage identity to incorporate a CDS annotation [default=0.7]
+    --pcoverage		<% identity> Minimum percentage of coverage over protein alignment to incorporate a CDS annotation [default=0.7]
+			             .. otherwise hint for a non-functional protein
+
+    --minlength		<length> Minimum contig length for annotation [default=500]
 
     --meta		Better for metagenome and plasmid annotations because of disparate codon usage [default=off]
 
@@ -77,6 +76,7 @@ def parseOptions_annotate
   # default options
   options[:outdir] = "BAnnotation"
   options[:pidentity] = 70
+  options[:pcoverage] = 70
   options[:minlength] = 500
   options[:meta] = 0
 
@@ -95,10 +95,10 @@ def parseOptions_annotate
       options[:minlength] = ARGV.shift
     when "--pidentity"
       options[:pidentity] = ARGV.shift
+    when "--pcoverage"
+      options[:pcoverage] = ARGV.shift
     when "--meta"
       options[:meta] = 1
-    when "--remotedb"
-      options[:remote_db] = ARGV.shift
     when "--externaldb"
       options[:external_db] = ARGV.shift
     when "--help", "-h"
@@ -204,7 +204,7 @@ if ARGV.size > 1
   system("ba_raxml")
 
   options = {}
-  genomes_list = []
+  genomes_list = []             # TODO multiple input genomes
 
   if ARGV[0] == "annotate"
 
@@ -217,7 +217,6 @@ if ARGV.size > 1
 
     # Check Options
     if ! options.has_key? :refgenome and
-       ! options.has_key? :remote_db and
        ! options.has_key? :external_db
       puts "You didn't provide a reference genome or a database for the annotation !"
     elsif ! options.has_key? :input

data/lib/bacterial-annotator/{genbank-manip.rb → sequence-annotation.rb}

@@ -1,14 +1,13 @@
 # -*- coding: utf-8 -*-
 # author:  	maxime déraspe
 # email:	maximilien1er@gmail.com
-# review:  	
 # date:    	15-02-24
 # version: 	0.0.1
 # licence:  	
 
 
 
-class GenbankManip
+class SequenceAnnotation
 
   attr_accessor :gbk, :coding_seq, :cds_file, :rna_file
 
@@ -67,13 +66,16 @@ class GenbankManip
           protId = locustag
         end
 
-        @coding_seq[protId] = {protId: protId,
-                               location: loc,
-                               locustag: locustag,
-                               gene: gene[0],
-                               product: product[0],
-                               bioseq: pepBioSeq,
-                               bioseq_gene: dnaBioSeq}
+        @coding_seq[protId] = {
+          protId: protId,
+          location: loc,
+          locustag: locustag,
+          gene: gene[0],
+          product: product[0],
+          bioseq: pepBioSeq,
+          bioseq_gene: dnaBioSeq,
+          bioseq_len: pepBioSeq.length
+        }
       end
 
     end
@@ -110,11 +112,13 @@ class GenbankManip
         dna = get_DNA(ft,@bioseq)
         dnaBioSeq = Bio::Sequence.auto(dna)
 
-        @rna_seq[locustag] = {type: ft.feature.to_s,
-                              location: loc,
-                              locustag: locustag,
-                              product: product,
-                              bioseq_gene: dnaBioSeq}
+        @rna_seq[locustag] = {
+          type: ft.feature.to_s,
+          location: loc,
+          locustag: locustag,
+          product: product,
+          bioseq_gene: dnaBioSeq
+        }
 
       end
 
@@ -125,7 +129,6 @@ class GenbankManip
   end
 
 
-
   # Print CDS to files
   # RETURN : cds_file path
   def write_cds_to_file outdir
@@ -174,12 +177,108 @@ class GenbankManip
   end
 
 
+  # add annotation from reference prot synteny
+  def add_annotation_ref_synteny_prot synteny_prot, annotations, ref_genome=nil
+
+    contig = @gbk.definition
+
+    prot_iterator = 0
+    @gbk.features.each_with_index do |cds, ft_index|
+
+      next if cds.feature != "CDS"
+
+      prot_iterator+=1
+      prot_id = contig+"_"+prot_iterator.to_s
+
+      ftArray = []
+      cds.qualifiers = []
+
+      hit = nil
+
+      next if ! synteny_prot.has_key? prot_id or
+        ! synteny_prot[prot_id].has_key? :homology
+
+      # puts "#{annotations.keys}"
+      if annotations.has_key? synteny_prot[prot_id][:homology][:hits][0]
+        hit = annotations[synteny_prot[prot_id][:homology][:hits][0]]
+        # puts hit
+      else
+        puts "no hit for #{prot_id}"
+        next
+      end
+
+      # hit = annotations[synteny_prot[prot_id][:homology][:hits][0]]
+
+      if synteny_prot.has_key? prot_id
+
+        locus, gene, product, note, inference = nil
+        locus = hit[:locustag]
+        gene = hit[:gene]
+        product = hit[:product]
+        note = hit[:note]
+        inference = hit[:inference]
+        pId = synteny_prot[prot_id][:homology][:pId]
+        cov_query = (synteny_prot[prot_id][:homology][:cov_query]*100).round(2)
+        cov_subject = (synteny_prot[prot_id][:homology][:cov_subject]*100).round(2)
+        reference_prot_id = synteny_prot[prot_id][:homology][:hits][0]
+
+        qLocusTag = Bio::Feature::Qualifier.new('locus_tag', "#{prot_id}")
+        ftArray.push(qLocusTag)
+
+        if gene != nil
+          qGene = Bio::Feature::Qualifier.new('gene', gene)
+          ftArray.push(qGene)
+        end
+
+        if product != nil
+          qProd = Bio::Feature::Qualifier.new('product', product)
+          ftArray.push(qProd)
+        end
+
+        # check if there is a reference genome.. reference_locus shouldn't be nil in that case
+        if locus != nil
+          qNote = Bio::Feature::Qualifier.new('note', "corresponds to #{locus} locus (AA identity: #{pId}%; coverage(q,s): #{cov_query}%,#{cov_subject}%) from #{ref_genome}")
+          ftArray.push(qNote)
+
+          db_source = "[DBSource]"
+          if reference_prot_id.include? "_"
+            db_source = "RefSeq"
+          else
+            db_source = "INSD"
+          end
+          qInference = Bio::Feature::Qualifier.new('inference', "similar to AA sequence:#{db_source}:#{reference_prot_id}")
+          ftArray.push(qInference)
+
+        end
+
+        if note != nil
+          qNote = Bio::Feature::Qualifier.new('note', note)
+          ftArray.push(qNote)
+        end
+
+        if inference != nil
+          qInference = Bio::Feature::Qualifier.new('inference', inference)
+          ftArray.push(qInference)
+        end
+
+      end
+
+      cds.qualifiers = ftArray
+
+    end
+
+
+  end
+
+
   # add annotation to a genbank file produced by prodigal
   def add_annotations annotations, mode, reference_locus=nil
 
     # nb_of_added_ft = 0
     i = 0
 
+    fdebug = File.open("debug-add-annotation.txt","w")
+
     contig = @gbk.definition
 
     if mode == "inplace"
@@ -195,9 +294,19 @@ class GenbankManip
         i += 1
         prot_id = contig+"_"+i.to_s
         hit = nil
-        hit = annotations[prot_id] if annotations.has_key? prot_id
+
+        if annotations.has_key? prot_id
+          hit = annotations[prot_id]
+        else
+          puts "no hit for #{prot_id}"
+          next
+        end
 
         if hit != nil
+
+          fdebug.write(hit)
+          fdebug.write("\n")
+
           locus, gene, product, note = nil
           locus = hit[:locustag]
           gene = hit[:gene]
@@ -271,6 +380,8 @@ class GenbankManip
 
     end
 
+    fdebug.close
+
   end
 
 
@@ -315,3 +426,4 @@ class GenbankManip
 
 
 end                             # end of Class
+

data/lib/bacterial-annotator/{fasta-manip.rb → sequence-fasta.rb}

@@ -8,29 +8,35 @@
 
 
 
-class FastaManip
+class SequenceFasta
 
-  attr_reader :fasta_flat, :fasta_file, :prodigal_files
+  attr_reader :fasta_flat, :fasta_file, :annotation_files
 
   # Initialize fasta holder
   def initialize fasta_file, meta
 
     @fasta_file = fasta_file
     @fasta_flat = Bio::FlatFile.auto(@fasta_file)
-    @meta = meta
-    @prodigal_files = nil
-    @single_fasta = nil
-    @seq_info = nil
 
     if @fasta_flat.dbclass != Bio::FastaFormat
       abort "Aborting : The input sequence is not a fasta file !"
     end
 
+    # @contigs = extract_contigs(@fasta_flat)
+
+    @meta = meta
+
+    @annotation_files = nil
+    @single_fasta = nil
+    @seq_info = nil
+
   end
 
+
   # Run prodigal on the genome to annotate
   def run_prodigal root, outdir
-    @prodigal_files = {}
+
+    @annotation_files = {}
     Dir.mkdir "#{outdir}" if ! Dir.exists? "#{outdir}"
     if @meta
       system("#{root}/prodigal.linux -p meta -i #{@fasta_file} -a #{outdir}/Proteins.fa -d #{outdir}/Genes.fa -o #{outdir}/Genbanks.gbk -q")
@@ -38,30 +44,34 @@ class FastaManip
       system("#{root}/prodigal.linux -i #{@fasta_file} -a #{outdir}/Proteins.fa -d #{outdir}/Genes.fa -o #{outdir}/Genbanks.gbk -q")
     end
 
-    @prodigal_files = {multiGBK: "#{outdir}/Genbanks.gbk",
-                       contigs: [],
-                       contigs_length: [],
-                       genes: "#{outdir}/Genes.fa",
-                       proteins: "#{outdir}/Proteins.fa",
-                       prot_ids_by_contig: {},
-                       fasta_path: "#{outdir}/single-fasta/",
-                       gbk_path: "#{outdir}/single-genbank/"}
+    @annotation_files = {
+      multiGBK: "#{outdir}/Genbanks.gbk",
+      contigs: [],
+      contigs_length: [],
+      genes: "#{outdir}/Genes.fa",
+      proteins: "#{outdir}/Proteins.fa",
+      prot_ids_by_contig: {},
+      fasta_path: "#{outdir}/single-fasta/",
+      gbk_path: "#{outdir}/single-genbank/"
+    }
+
     split_fasta outdir
     split_genbank outdir, "#{outdir}/Genbanks.gbk"
     extract_cds_names
-    @prodigal_files
+    @annotation_files
+
   end
 
 
-  # Split Multi Genbanks file
+  # Split Multi Fasta file
   # RETURN : array of fasta files
   def split_fasta outdir
     @single_fasta = {}
     Dir.mkdir("#{outdir}/single-fasta") if ! Dir.exists?("#{outdir}/single-fasta")
     @fasta_flat.each_entry do |seq|
       file_name = seq.definition.chomp.split(" ")[0]
-      @prodigal_files[:contigs] << "#{file_name}"
-      @prodigal_files[:contigs_length] << seq.seq.length
+      @annotation_files[:contigs] << "#{file_name}"
+      @annotation_files[:contigs_length] << seq.seq.length
       File.open("#{outdir}/single-fasta/#{file_name}.fasta", "w") do |fwrite|
         fwrite.write(seq)
       end
@@ -108,7 +118,6 @@ class FastaManip
     outseq = "ORIGIN\n"
     # puts "ORIGIN"
 
-    ntNum = 0
     sequence = seq.seq.downcase
 
     nt_left = true
@@ -144,7 +153,7 @@ class FastaManip
 
     prot_ids = {}
     prot_length = {}
-    flatfile = Bio::FlatFile.auto(@prodigal_files[:proteins])
+    flatfile = Bio::FlatFile.auto(@annotation_files[:proteins])
 
     flatfile.each_entry do |entry|
       prot_id = entry.definition.split(" ")[0]
@@ -163,8 +172,8 @@ class FastaManip
       prot_array.sort! { |a,b| a.split("_")[-1].to_i <=> b.split("_")[-1].to_i }
     end
 
-    @prodigal_files[:prot_ids_by_contig] = prot_ids
-    @prodigal_files[:prot_ids_length] = prot_length
+    @annotation_files[:prot_ids_by_contig] = prot_ids
+    @annotation_files[:prot_ids_length] = prot_length
 
   end
 

data/lib/bacterial-annotator/{synteny-manip.rb → sequence-synteny.rb}

@@ -7,20 +7,43 @@
 # licence:  	
 
 
+class SequenceSynteny
 
-class SyntenyManip
+  attr_reader :query_file, :subject_file, :aln_hits, :query_sequences, :subject_sequences
 
-  attr_reader :query_file, :subject_file, :aln_hits
-
-  def initialize query_file, subject_file, name, pidentity, type
+  def initialize query_file, subject_file, name, pidentity, min_coverage, type
     @query_file = query_file
     @subject_file = subject_file
+
+    @query_sequences = get_sequences(query_file)
+    @subject_sequences = get_sequences(subject_file)
+
     @name = name
     @pidentity = pidentity
+    @min_coverage = min_coverage
     @aln_file = nil
     @type = type
+
   end                           # end of initialize
 
+
+  # get sequences name with length in hash
+  def get_sequences seq_file
+
+    sequences = {}
+    flat = Bio::FlatFile.auto("#{seq_file}")
+    flat.each_entry do |s|
+      s_name = s.definition.chomp.split(" ")[0]
+      sequences[s_name] = {}
+      sequences[s_name][:length] = s.seq.length
+      sequences[s_name][:conserved] = false
+      sequences[s_name][:contig] = s_name.split("_")[0..-2].join("_") if s_name.include? "_"
+    end
+
+    sequences
+
+  end
+
   # run blat on proteins
   def run_blat root, outdir
     base_cmd = "#{root}/blat.linux -out=blast8 -minIdentity=#{@pidentity}"
@@ -32,9 +55,98 @@ class SyntenyManip
     # extract_hits
   end                           # end of method
 
+
+  # Extract Hit from blast8 file and save it in hash
+  # contig-0_1      ABJ71957.1      96.92   65      2       0       1       65      1       65      9.2e-31 131.0
+  def extract_hits mode
+
+    feature = ""
+    File.open(@aln_file,"r") do |fread|
+      while l = fread.gets
+
+        lA = l.chomp!.split("\t")
+        key = lA[0]
+
+        # extraction of hit id depends on mode ..
+        if mode == :refgenome
+          hit = lA[1]
+          feature = "cds"
+        elsif mode == :externaldb
+          # hit = lA[1].chomp.split("|")[3]
+          hit = lA[1]
+          feature = "cds"
+        end
+
+        # compute coverage based on sequences length
+        cov_query = (lA[3].to_f/@query_sequences[key][:length]).round(2)
+        cov_subject = (lA[3].to_f/@subject_sequences[hit][:length]).round(2)
+
+        # assert cutoff on identity and coverage
+        # 1 -> pass cutoff, 0 under cutoff
+        assert_cutoff = [1,1,1]
+        assert_cutoff[0] = 0 if lA[2].to_f < @pidentity
+        assert_cutoff[1] = 0 if cov_query < @min_coverage
+        assert_cutoff[2] = 0 if cov_subject < @min_coverage
+
+        # first hit for query
+        if ! @query_sequences[key].has_key? :homology
+          @query_sequences[key][:conserved] = true
+          @subject_sequences[key][:conserved] = true
+          @query_sequences[key][:homology] = {
+            pId: lA[2].to_f.round(2),
+            cov_query: cov_query,
+            cov_subject: cov_subject,
+            evalue: lA[10],
+            score: lA[11].to_f,
+            hits: [hit],
+            length: [lA[3].to_i],
+            query_location: [[lA[6].to_i,lA[7].to_i]],
+            subject_location: [[lA[8].to_i,lA[9].to_i]],
+            feature: feature,
+            assert_cutoff: assert_cutoff
+          }
+          @subject_sequences[hit][:hits] = [key]
+
+        # query already got at least 1 hit and new_score > last_score
+        elsif lA[11].to_f > @query_sequences[key][:homology][:score]
+          @query_sequences[key][:conserved] = true
+          @subject_sequences[key][:conserved] = true
+          @query_sequences[key][:homology] = {
+            pId: lA[2].to_f.round(2),
+            cov_query: cov_query,
+            cov_subject: cov_subject,
+            evalue: lA[10],
+            score: lA[11].to_f,
+            hits: [hit],
+            length: [lA[3].to_i],
+            query_location: [[lA[6].to_i,lA[7].to_i]],
+            subject_location: [[lA[8].to_i,lA[9].to_i]],
+            feature: feature,
+            assert_cutoff: assert_cutoff
+          }
+          @subject_sequences[hit][:hits] =  [key]
+
+        # query already got at least 1 hit and score == last_score
+        elsif lA[11].to_f == @query_sequences[key][:homology][:score]
+          @query_sequences[key][:homology][:hits] << hit
+          @query_sequences[key][:homology][:length] << lA[3].to_i
+          @query_sequences[key][:homology][:query_location] << [lA[6].to_i,lA[7].to_i]
+          @query_sequences[key][:homology][:subject_location] << [lA[8].to_i,lA[9].to_i]
+          if @subject_sequences[hit].has_key? :hits
+            @subject_sequences[hit][:hits] << key
+          else
+            @subject_sequences[hit][:hits] = [key]
+          end
+        end
+      end
+    end
+
+  end                           # end of method
+
+
   # Extract Hit from blast8 file and save it in hash
   # contig-0_1      ABJ71957.1      96.92   65      2       0       1       65      1       65      9.2e-31 131.0
-  def extract_hits_prodigal mode, ref_cds=nil
+  def extract_hits_prodigal mode
 
     @aln_hits = {}
     feature = ""
@@ -49,8 +161,8 @@ class SyntenyManip
           hit = lA[1].chomp.split("|")[3]
           feature = "cds"
         end
+        next if lA[2].to_f < @pidentity
         if ! @aln_hits.has_key? key
-          next if lA[2].to_f < @pidentity
           @aln_hits[key] = {
             pId: lA[2].to_f.round(2),
             evalue: lA[10],
@@ -99,10 +211,12 @@ class SyntenyManip
           feature = hit_split[1]
           product = hit_split[2]
         end
+        next if lA[2].to_f < @pidentity
         if ! @aln_hits.has_key? key
-          next if lA[2].to_f < @pidentity
           @aln_hits[key] = {
             pId: lA[2].to_f.round(2),
+            # cov_query: (@query_sequences[key][:length]/lA[3].to_f).round(2),
+            # cov_subject: (@subject_sequences[hit][:length]/lA[3].to_f).round(2),
             evalue: lA[10],
             score: lA[11].to_f,
             hits: [hit],
@@ -115,6 +229,8 @@ class SyntenyManip
         elsif lA[11].to_f > @aln_hits[key][:score]
           @aln_hits[key] = {
             pId: lA[2].to_f.round(2),
+            # cov_query: (@query_sequences[key][:length]/lA[3].to_f).round(2),
+            # cov_subject: (@subject_sequences[hit][:length]/lA[3].to_f).round(2),
             evalue: lA[10],
             score: lA[11].to_f,
             hits: [hit],
@@ -135,7 +251,7 @@ class SyntenyManip
       end
     end
 
-    prune_aln_hits @aln_hits
+    # prune_aln_hits @aln_hits
 
   end                           # end of method
 
@@ -178,6 +294,10 @@ class SyntenyManip
           annotations[p][:length] = @aln_hits[p][:length][hit_index]
           i+=1
 
+          File.open("debug-annotation-by-contig.txt","a") do |fout|
+            fout.write("#{p} #{@aln_hits[p][:pId]} #{@aln_hits[p][:cov_query]} #{@aln_hits[p][:cov_subject]} #{ref_cds[h][:product]}\n")
+          end
+
         else
 
           annotations[p] = nil