mapp2g 0.1.4 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 80ed37b4687cdd64f5a292593e2b15addadbabe49223c79835b530e94ac23be6
-  data.tar.gz: 37a038d19321a88e10f44261a3027bda85c40c29d529b710bb93ed760ada585e
+  metadata.gz: caad7f8e3dc7a4894e8cf5d2de8bb8a706cf654896decfe09e58566bec2d4b74
+  data.tar.gz: 67dffdb34daf77d28ffce2c18d152344ad78b9cba6bd66f74b9c30f870e3e9ea
 SHA512:
-  metadata.gz: 79f3e7022e8532c1bb9fc0f333927042a4eead1a40d1ecbcb9408d9391a1f21e162830132c395411aac7065076fa14eed694f6ee5b6824a3edc1c193a99ff9ab
-  data.tar.gz: b6c93c34a6576ea1e57fdc2eecdb04e767a595b6b6a037885ab3a89ebf46f36272b6017f55441788ca1cb89dd80cbe273d6a7049f067445b48d8464a486402b2
+  metadata.gz: 47a0a17673e4aebb4b8d83dbbf7239e783aa926ad698ca967fcd718f825ca8f5883f63b6cdd6df41543aaa7ff301364c7e155d64ce0693569735696f2bbf4fd3
+  data.tar.gz: 4f31480439323991e2adffd8bbdd75fa05a68ff50ee683eb7b196de5087f169af63caad68538c1f976b662e6bf09d2ac5c4eea544f9f4aae11fa60498efc9f79

data/README.md

@@ -26,6 +26,9 @@ Usage: mapp2g [options]
     -h, --help                       show this help message and exit
 ```
 
+Query sequences should be in FASTA format. Multiple sequences can be included in one file.
+
+
 (example)
 ```
 mapp2g -q human_genome.fasta -q p53.protein.fasta
@@ -37,6 +40,18 @@ Reference genomes should be formated in blastdb before running mapp2g. blastdb c
 makeblastdb -in human_genome.fasta -dbtype nucl -parse_seqids
 ```
 
+## Outputs
+
+For each query, the following files are generated.
+
+- query sequence in fasta
+- blast output in tab-delmited format (format 6)
+- exonerate full output
+- exonerate alignment in gff3 format
+- report.json
+
+report.json contains all of the information above in json line format.
+
 
 ## License
 

data/exe/mapp2g

@@ -1,9 +1,11 @@
 #!/usr/bin/env ruby
 
-require "mapp2g"
+#require_relative '../lib/mapp2g' # for development
+require 'mapp2g'
 require 'bio'
 require 'tempfile'
 require 'optparse'
+require 'json'
 
 ### Parse options
 
@@ -62,6 +64,9 @@ rescue => e
 end
 
 begin
+  unless File.exist?(genome)
+    raise "genome file (#{genome}) not found"
+  end
   unless File.exist?("#{genome}.nsq") && File.exist?("#{genome}.nos")
     raise "genome is not indexed. Please run 'makeblastdb -in #{genome} -dbtype nucl -parse_seqids'"
   end
@@ -72,6 +77,8 @@ end
 
 ### Main
 
+report_json_lines = []
+
 Dir.mkdir(outdir)
 
 Bio::FlatFile.open(Bio::FastaFormat, query).each_with_index do |fas, i|
@@ -80,11 +87,42 @@ Bio::FlatFile.open(Bio::FastaFormat, query).each_with_index do |fas, i|
   tf.close
   id = (i + 1).to_s
   query_file_path = "#{outdir}/#{id}.fasta"
-  out_file_path = "#{outdir}/#{id}.exonerate.txt"
   File.open(query_file_path, "w"){|o| o.puts fas}
 
   mapper = Mapp2g::Mapper.new()
   res = mapper.run(query_file_path, genome)
-  File.open(out_file_path, "w"){|o| o.puts res}
+  if res
+    out_file_path = "#{outdir}/#{id}.exonerate.txt"
+    File.open(out_file_path, "w"){|o| o.puts res[:exonerate_result]}
+    out_file_path = "#{outdir}/#{id}.blast.txt"
+    File.open(out_file_path, "w"){|o| o.puts res[:blast_result]}
+    gff3 = Mapp2g::ExonerateOutput.new(res[:exonerate_result]).to_gff3()
+    out_file_path = "#{outdir}/#{id}.exonerate.gff3"
+    File.open(out_file_path, "w"){|o| o.puts gff3}
+
+    report = {
+      "runtime_id" => id,
+      "query_id" => fas.entry_id, 
+      "query_fasta" => fas.to_s,
+      "exonerate" => res[:exonerate_result], 
+      "blast" => res[:blast_result], 
+      "gff3" => gff3
+    }
+    report_json_lines << report.to_json
+
+  else
+    report = {
+      "runtime_id" => id,
+      "query_id" => fas.entry_id,
+      "query_fasta" => fas.to_s,
+      "exonerate" => nil,
+      "blast" => nil,
+      "gff3" => nil
+    }
+    STDERR.puts "No hit for #{fas.entry_id}"
+  end
 
 end
+
+report_json = report_json_lines.join("\n")
+File.open("#{outdir}/report.json", "w"){|o| o.puts report_json}

data/lib/mapp2g/mapper.rb

@@ -4,9 +4,9 @@ module Mapp2g
 
   class Mapper
 
-    EVALUE_DEFAULT = 1.0e-8
+    EVALUE_DEFAULT = 1.0e-5
     NCPU_DEFAULT = 4
-    MAX_HSP_INTERVAL = 50000
+    MAX_HSP_INTERVAL = 400000
     EXTENSION = 50000
     TMPDIR_DEFAULT = Dir.tmpdir
 
@@ -15,11 +15,11 @@ module Mapp2g
 
     ## step 1: tblastn
     def exec_tblastn_to_know_rough_target_region(query, genome, evalue=EVALUE_DEFAULT, ncpu=NCPU_DEFAULT)
-      cmd = "tblastn -db #{genome} -query #{query} -max_target_seqs 40 -soft_masking yes -seg yes -outfmt 6 -evalue #{evalue} -num_threads #{ncpu} "
+      cmd = "tblastn -db #{genome} -query #{query} -max_target_seqs 40 -soft_masking yes -seg yes -outfmt 6 -evalue #{evalue} -num_threads #{ncpu} -culling_limit 2"
       # puts cmd
       res = nil
-      IO.popen(cmd){|io| res = io.read}
-      # puts res
+      IO.popen(cmd){|io| res = io.read} 
+#     STDERR.puts res
       if res == ""
       ## no hit
         return nil
@@ -27,7 +27,7 @@ module Mapp2g
         lines = []
         prev_chr = nil
         res.split(/\n/).each do |l|
-      #    p prev_chr
+#          p prev_chr
           a = l.chomp.split(/\t/)
           unless prev_chr
             lines << l
@@ -42,7 +42,7 @@ module Mapp2g
         a = lines.shift.chomp.split(/\t/)
         left, right = [a[8].to_i, a[9].to_i].sort
         
-        STDERR.puts [left, right].inspect
+#        STDERR.puts [left, right].inspect
     
         lines.each do |l|
           a = l.chomp.split(/\t/)
@@ -57,14 +57,16 @@ module Mapp2g
             break
           end
         end
-      #  p [left, right]
+#        STDERR.puts [left, right].inspect
         
         top_chromosome = res.split(/\n/)[0].split(/\t/)[1]
         h = {
           :top_chromosome => top_chromosome, 
           :left => left,
-          :right => right
+          :right => right,
+          :blast_result => res
         }
+#        STDERR.puts h.inspect
         return h
       end
     end
@@ -117,8 +119,10 @@ module Mapp2g
         tf.close
       
         exonerate_result = exec_exonerate(query, tf.path)
-        return exonerate_result
-      
+        return {
+          :blast_result => hit[:blast_result], 
+          :exonerate_result => exonerate_result
+        }
       else
         return nil
       end

data/lib/mapp2g/report.rb

@@ -0,0 +1,75 @@
+module Mapp2g
+
+  class ExonerateOutput
+
+    def self.load(file)
+      self.new(File.read(file))
+    end
+
+    # @param exonerate_out [String] exonerate output text, not file path
+    def initialize(exonerate_out)
+      @exonerate_out = exonerate_out
+      @query_name = nil
+      @target = nil
+      @cigar = nil
+      @gff2_lines = []
+      #vulgar = nil
+      parse()
+    end
+
+    attr_reader :query_name, :target, :cigar, :gff2_lines
+
+    def parse(opt={})
+      @exonerate_out.each_line do |l|
+        if m = /\s+Query:\s/.match(l)
+          @query_name = m.post_match.chomp.split[0]
+        elsif m = /\s+Target:\s/.match(l)
+          @target = m.post_match.split[0]
+        elsif m = /^cigar:\s/.match(l)
+          @cigar = m.post_match.chomp
+        elsif /^#{@target}/ =~ l &&
+            (/\texonerate:est2genome\t/.match(l) || /\texonerate:protein2genome:local\t/.match(l)) &&
+            (/\texon\t/.match(l) || /\tgene\t/.match(l))
+          @gff2_lines << l.chomp
+        end
+      end
+    end
+
+    def to_gff3(opt={})
+      gff3_lines = []
+      @gff2_lines.each do |l|
+        a = l.chomp.split(/\t/)
+        b = Array.new(9)
+        a.each_with_index{|x, i| b[i] = x}
+        if b[2] == "gene"
+          b[2] = "match"
+          orig_attribute = b[8].split(";").map{|x| x.strip.split(/\s+/)}.to_h
+      #    p orig_attribute
+          c = @cigar.split(/\s+/)
+          cigar_pairs = c[9..-1].join.scan(/[MDI]\d+/)
+          attribute = {'ID' => @query_name,
+                       'Target' => [@query_name, c[1].to_i + 1, c[2]].join(" "),
+                       'Gap'=> cigar_pairs.join(" "),
+                       'identity' => orig_attribute['identity'],
+                       'similarity' => orig_attribute['similarity']}
+          b[8] = attribute.map{|k, v| "#{k}=#{v}"}.join(";")
+        elsif b[2] == "exon"
+          b[2] = "match_part"
+          orig_attribute = b[8].split(";").map{|x| x.strip.split(/\s+/)}.to_h
+          attribute = {'Parent' => @query_name,
+          'identity' => orig_attribute['identity'],
+          'similarity' => orig_attribute['similarity']}
+          b[8] = attribute.map{|k, v| "#{k}=#{v}"}.join(";")
+
+        else
+          raise
+        end
+        gff3_lines << b.join("\t")
+      end
+      return gff3_lines.join("\n")
+    end
+
+  end
+
+end
+

data/lib/mapp2g/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Mapp2g
-  VERSION = "0.1.4"
+  VERSION = "0.2.0"
 end

data/lib/mapp2g.rb

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require_relative "mapp2g/mapper"
+require_relative "mapp2g/report"
 require_relative "mapp2g/version"
 
 module Mapp2g

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: mapp2g
 version: !ruby/object:Gem::Version
-  version: 0.1.4
+  version: 0.2.0
 platform: ruby
 authors:
 - Shuji Shigenobu
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2023-07-09 00:00:00.000000000 Z
+date: 2023-07-17 00:00:00.000000000 Z
 dependencies: []
 description: mapp2g is a bioinformatics software, which map and align protein sequences
   (amino acid sequences) to genome references in a splicing-aware way. mapp2g alignment
@@ -28,6 +28,7 @@ files:
 - exe/mapp2g
 - lib/mapp2g.rb
 - lib/mapp2g/mapper.rb
+- lib/mapp2g/report.rb
 - lib/mapp2g/version.rb
 - mapp2g.gemspec
 - scripts/add_annotation_from_uniprot_fasta_to_gff.rb
@@ -57,7 +58,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.10
+rubygems_version: 3.4.15
 signing_key: 
 specification_version: 4
 summary: mapp2g is the tool to map protein sequences to genome references in a splicing-aware