transfuse 0.1.4 → 0.4.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a5b17813bba553fa78a7e3cf24366981e368ebfe
-  data.tar.gz: f57a5011b3296862518ce54a7f3d4bb928e05aab
+  metadata.gz: 03b84ea2236b7c43ce321f64352bb6cb7b7035d4
+  data.tar.gz: ed70d9a9c204a06b550a7b95c01c74afdcb98b51
 SHA512:
-  metadata.gz: 52af21dbf0c5b49aa746d515a214229bccd78a55d5998ddc929b2594d8abe5bafcb9cd1bbc98ec6ea1a41a5cb3240666616837d3d06c5cbacd25acca3fe392ca
-  data.tar.gz: 147cc022f22ad319ff0651a6092ced5d3f1c7e99ab3989a8091970ab5c25a97e42d9b98f7384e28712ef802c4ced8090c3942ba66105963013eb30e46c3b58cb
+  metadata.gz: 6074f06d8afd9e33fce7b77e0221ffc8a300da966cc7916f86eb245d02c1e2b5519a831e1c99962fcfd5b085b5cd3d94765297ec94469270e4246c6c9ab71830
+  data.tar.gz: d992bac7cb1a98cd523e4df28d5b684f42bf8d8d7deb7882d42ecd8bb50bb137a190200006c7528f834d9e30a214b1f304a46b8502176993e22b019658c19061

data/README.md

@@ -1 +1,42 @@
-Transfuse
+## Transfuse
+
+**Transfuse is currently in development and is not yet ready for use**
+
+Transfuse intelligently merges your multiple de novo transcriptome assemblies. Run multiple assemblies with different de novo assemblers, or different settings in the same assembler and have them combined into a single high quality transcriptome.
+
+Transfuse takes in the reads you used to do the assembly and a list of fasta files and produces a single output fasta file.
+
+### Installation and Running
+
+To install Transfuse, clone this repo:
+
+`git clone https://github.com/cboursnell/transfuse.git`
+
+Then build and install the ruby gem
+
+`gem build *spec; gem install *gem`
+
+### Usage
+
+Transfuse is run on the command line. The options are:
+
+```
+  -a, --assembly=<s>    assembly files in FASTA format, comma-separated
+  -l, --left=<s>        left reads file in FASTQ format
+  -r, --right=<s>       right reads file in FASTQ format
+  -o, --output=<s>      write merged assembly to file
+  -t, --threads=<i>     number of threads (default: 1)
+  -v, --verbose         be verbose
+  -e, --version         Print version and exit
+  -h, --help            Show this message
+```
+
+An example command:
+
+`transfuse --assembly soap-k31.fa,soap-k41.fa,soap-k51.fa --left reads_1.fq --right reads_2.fq --output soap-merged.fa --threads 12`
+
+### License
+
+This is adademic software - please cite us if you use it in your work.
+
+Transfuse is released under the MIT license.

data/Rakefile

@@ -5,15 +5,21 @@ Rake::TestTask.new do |t|
 end
 
 Rake::TestTask.new do |t|
-  t.name = :corset
+  t.name = :cluster
   t.libs << 'test'
-  t.test_files = ['test/test_corset.rb']
+  t.test_files = ['test/test_cluster.rb']
 end
 
 Rake::TestTask.new do |t|
-  t.name = :cluster
+  t.name = :fuse
   t.libs << 'test'
-  t.test_files = ['test/test_cluster.rb']
+  t.test_files = ['test/test_transfuse.rb']
+end
+
+Rake::TestTask.new do |t|
+  t.name = :cons
+  t.libs << 'test'
+  t.test_files = ['test/test_consensus.rb']
 end
 
 desc "Run tests"

data/bin/transfuse

@@ -22,23 +22,24 @@ opts = Trollop::options do
   OPTIONS:
 
   EOS
-  opt :assembly, "assembly files in FASTA format, comma-separated",
+  opt :assemblies, "assembly files in FASTA format, comma-separated",
       :type => String, :required => true
-  opt :scores, "transrate contig score output files, comma-separated",
-      :type => String
   opt :left, "left reads file in FASTQ format",
       :type => String
   opt :right, "right reads file in FASTQ format",
       :type => String
+  opt :scores, "transrate contig score output files, comma-separated. Ignored if reads are provided",
+      :type => String
   opt :output, "write merged assembly to file",
       :type => String, :required => :true
   opt :threads, "number of threads", :type => :int, :default => 1
+  opt :id, "sequence identity to cluster at", :type => :float, :default => 1.0
   opt :verbose, "be verbose"
 end
 
 transfuse = Transfuse::Transfuse.new opts.threads, opts.verbose
 
-assembly_files = transfuse.check_files opts.assembly
+assembly_files = transfuse.check_files opts.assemblies
 score_files = transfuse.check_files opts.score if opts.score
 left = transfuse.check_files opts.left if opts.left
 right = transfuse.check_files opts.right if opts.right
@@ -54,6 +55,7 @@ else
   abort msg
 end
 
+# filter out assemblies with low score
 assembly_files = transfuse.filter assembly_files, scores
 
 # concatenate assemblies into one fasta file
@@ -62,12 +64,11 @@ cat = transfuse.concatenate assembly_files
 # load fasta sequences from concatenated file into hash
 transfuse.load_fasta cat
 
-# cluster using vsearch or maybe cd-hit-est
-clusters = transfuse.cluster cat
-
-transfuse.sequence_alignment clusters
-# pull out contigs from each cluster based on the scores
-# best = transfuse.select_contigs clusters, scores
+# cluster using vsearch
+msa = transfuse.cluster cat, opts.id
 
-# transfuse.output_contigs best, cat, opts.output
+# read the msa from vsearch and produce a consensus fasta
+cons = transfuse.consensus msa, scores, opts.output
 
+# transrate the consensus output to remove low scoring contigs
+transfuse.transrate_consensus cons, opts.output, left, right

data/deps/deps.yaml

@@ -0,0 +1,11 @@
+vsearch:
+  binaries:
+    - vsearch
+  version:
+    number: '1.1.3'
+    command: 'vsearch --version'
+  url:
+    64bit:
+      linux: https://github.com/torognes/vsearch/releases/download/v1.1.3/vsearch-1.1.3-linux-x86_64
+      macosx: https://github.com/torognes/vsearch/releases/download/v1.1.3/vsearch-1.1.3-osx-x86_64
+  unpack: false

data/lib/transfuse/cluster.rb

@@ -5,61 +5,31 @@ module Transfuse
 
   class Cluster
 
-    def initialize threads, verbose
-      @cdhit = Which::which('cd-hit-est').first
-      raise "cd-hit-est was not in the PATH - please install it" unless @cdhit
+    def initialize threads, verbose, id
       @vsearch = Which::which('vsearch').first
       raise "vsearch was not in the PATH - please install it" unless @vsearch
-      @id = "1.00"
+      @id = id.to_s
       @threads = threads
       @verbose = verbose
     end
 
     def run fasta
-      use_cd_hit = false
-      if use_cd_hit
-        output = cd_hit fasta
-        return parse_output output
-      else
-        output = vsearch fasta
-        return parse_vsearch_output output
-      end
-    end
-
-    def cd_hit fasta
-      puts "running cd-hit-est" if @verbose
-      output = "#{File.basename(fasta, File.extname(fasta))}_cdhit.fa"
-      cdhit_cmd = generate_cdhit_command fasta, output
-      puts cdhit_cmd if @verbose
-      cluster = Cmd.new cdhit_cmd
-      cluster.run output
-      return "#{output}.clstr"
+      cluster_output, msa_output = vsearch fasta
+      return parse_vsearch_output(cluster_output, msa_output)
     end
 
     def vsearch fasta
-      puts "running vsearch" if @verbose
-      cluster_output = "#{fasta}.clust"
-      vsearch_cmd = generate_vsearch_command fasta, cluster_output
+      print "running vsearch..." if @verbose
+      cluster_output = "#{File.basename(fasta)}-#{@id}.clust"
+      msa_output = "#{File.basename(fasta)}-#{@id}.aln"
+      vsearch_cmd = generate_vsearch_command fasta, cluster_output, msa_output
       cluster = Cmd.new vsearch_cmd
       cluster.run cluster_output
-      return cluster_output
-    end
-
-    def generate_cdhit_command fasta, out
-      #cd-hit-est -i all.fa  -o cd-hit-clusters.txt -c 0.99999 -T 24 -d 100
-      cmd = "#{@cdhit}"
-      cmd << " -i #{fasta}"
-      cmd << " -o #{out}"
-      cmd << " -c #{@id}" # similarity = number of identical bases /
-                          #              length of shorter sequences
-      cmd << " -T #{@threads}"
-      cmd << " -n 10" # word length - maybe increase??
-      cmd << " -d 100" # output name width
-      cmd << " -g 1" # slower but more accurate mode
-      cmd << " -M 8000" # increase memory
+      puts " Done. Created #{cluster_output}" if @verbose
+      return [cluster_output, msa_output]
     end
 
-    def generate_vsearch_command fasta, out
+    def generate_vsearch_command fasta, out, msa
       vsearch = "#{@vsearch}"
       vsearch << " --cluster_fast #{fasta}"
       vsearch << " --id #{@id}"
@@ -67,45 +37,60 @@ module Transfuse
       vsearch << " --qmask none" # no masking
       vsearch << " --strand both"
       vsearch << " --uc #{out}"
+      vsearch << " --msaout #{msa}"
       vsearch << " --threads #{@threads}"
       return vsearch
     end
 
-    def parse_output cluster_output
-      puts "parsing cd-hit output #{cluster_output}" if @verbose
-      cluster_id = 0
-      clusters = {}
-      File.open(cluster_output).each_line do |line|
-        if line =~ />Cluster\ ([0-9]+)/
-          cluster_id = $1.to_i
-        elsif line =~ /[0-9]+\s+.+nt,\ >(.+)\.\.\.\sat\s([+\-])\/([0-9\.]+)\%/
-          contig_name = $1
-          strand = $2
-          id = $3.to_f
-          clusters[cluster_id] ||= []
-          clusters[cluster_id] << { :name => contig_name, :strand => strand }
-        elsif line =~ /[0-9]+\s+[0-9]+nt,\s>(.+)\.\.\.\s\*/
-          contig_name = $1
-          strand = "+"
-          clusters[cluster_id] ||= []
-          clusters[cluster_id] << { :name => contig_name, :strand => strand }
-        end
-      end
-      return clusters
-    end
-
-    def parse_vsearch_output cluster_output
+    def parse_vsearch_output cluster_output, msa_output
+      print "parsing vsearch output" if @verbose
       clusters = {}
+      lookup = {}
+      second = 0
+      count = 0
       File.open(cluster_output).each_line do |line|
+        count+=1
         if line.start_with?("S") or line.start_with?("H")
           cols = line.chomp.split("\t")
-          cluster = cols[1].to_i
+          cluster = cols[1]
+          len = cols[2].to_i
+          cigar = cols[7]
+          strand = cols[4]
+          strand = "+" if strand == "*"
           contig_name = cols[8]
+
           clusters[cluster] ||= []
-          clusters[cluster] << contig_name
+          clusters[cluster] << { :name => contig_name, :strand => strand }
+          lookup[contig_name] = cluster
+        end
+        if count%10_000==0 and @verbose
+          print "."
         end
       end
-      return clusters
+      puts " Done" if @verbose
+      print "parsing msa output    " if @verbose
+      count = 0
+      msa = {}
+      Bio::FastaFormat.open(msa_output).each do |entry|
+        count += 1
+        name = entry.entry_id
+        if name != "consensus"
+          # name = name[1..-1]
+          if name[0]=="*"
+            name = name[1..-1]
+          end
+          # what cluster is name in?
+          cluster = lookup[name]
+          msa[cluster] ||= []
+          msa[cluster] << { :name => name, :seq => entry.seq.seq }
+        end
+        if count%10_000==0 and @verbose
+          print "."
+        end
+
+      end
+      puts " Done" if @verbose
+      return msa
     end
 
   end

data/lib/transfuse/cmd.rb

@@ -18,7 +18,7 @@ module Transfuse
 
     def run file=nil
       unless file.nil?
-        if File.exist?(file)
+        if File.exist?(file) and File.stat(file).size > 0
           @stdout = ""
           @stderr = ""
           @status = Status.new

data/lib/transfuse/consensus.rb

@@ -0,0 +1,105 @@
+
+require 'bio'
+require 'set'
+
+module Transfuse
+
+  class Consensus
+
+    attr_reader :contigs
+
+    def initialize verbose
+      @verbose = verbose
+    end
+
+    def run msa, scores, output
+      return 1 if File.exist?(output)
+      print "writing consensus " if @verbose
+      # msa is a hash
+      #   key = cluster id
+      #   value = list
+      #     list of sequences in cluster aligned with gaps
+      preoutput = "#{File.basename(output, File.extname(output))}_cons.fa"
+      count = 0
+      File.open("#{output}.data", "w") do |out2|
+        File.open(preoutput, "w") do |out|
+          msa.each do |id, list|
+            count+=1
+            print "." if count%5_000==0 and @verbose
+            exons={}
+            cons = []
+            length = list[0][:seq].length
+              list.each_with_index do |hash, index|
+                seq = hash[:seq]
+                name = hash[:name]
+                out2.write "#{id}\t#{scores[name][:score]}\t#{name}\n"
+                prev = ""
+                gap = 0
+                exon = 0
+                seq.each_char do |c|
+                  if c=="-"
+                    base="-"
+                  else
+                    base="*"
+                  end
+                  if base!=prev
+                    if c=="-"
+                      gap+=1
+                    else
+                      exon+=1
+                    end
+                  end
+                  if c=="-"
+                    prev = "-"
+                  else
+                    prev = "*"
+                  end
+                end
+                exons[index] = exon
+              end
+
+            consensus = ""
+            0.upto(length-1) do |i|
+              base="N"
+              counts = {}
+              list.each_with_index do |hash, index|
+                seq = hash[:seq]
+                if seq[i] != "-" and seq[i] != "N"
+                  counts[seq[i]]||=0
+                  counts[seq[i]] += 1
+                  if exons[index]==1
+                    base = seq[i]
+                  end
+                end
+              end
+              if counts.size>0
+                base = counts.sort.last.first
+              end
+              consensus << base
+            end
+
+            if consensus.count("N") < consensus.length.to_f*0.5
+              cons << consensus
+            end
+
+            list.each_with_index do |hash, index|
+              if exons[index] > 1
+                cons << hash[:seq].delete("-")
+              end
+            end
+
+            cons.each_with_index do |s,index|
+              out.write ">contig#{id}.#{index+1}\n"
+              out.write "#{s}\n"
+            end
+
+          end # msa.each
+        end # file
+      end # file open
+      puts " Done" if @verbose
+      return preoutput
+    end # def
+
+  end
+
+end

data/lib/transfuse/transfuse.rb

@@ -6,27 +6,28 @@ end
 
 module Transfuse
 
+  require 'bio'
   require 'csv'
   require 'transrate'
+  require 'threach'
 
   class Transfuse
 
     def initialize threads, verbose
       @threads = threads
       @verbose = verbose
-      @clustalo = Which::which('clustalo').first
-      raise "clustalo was not in the PATH - please install it" unless @clustalo
     end
 
     def check_files string
+      # puts "check file string: #{string}" if @verbose
       list = []
       string.split(",").each do |file|
         file = File.expand_path(file)
         if File.exist?(file)
-          puts "#{file} exists" if @verbose
+          puts "#{File.basename(file)} exists" if @verbose
           list << file
         else
-          abort "#{file} not found"
+          abort "#{File.basename(file)} not found"
         end
       end
       return list
@@ -51,42 +52,27 @@ module Transfuse
       return File.expand_path(catted_fasta)
     end
 
-    def cluster file
-      puts "clustering #{file}" if @verbose
-      cluster = Cluster.new @threads, @verbose
-      return cluster.run file
-    end
-
     def load_fasta fasta
+      print "loading fasta sequence #{fasta}..." if @verbose
       @sequences = {}
+      count = 1
       Bio::FastaFormat.open(fasta).each do |entry|
         @sequences[entry.entry_id] = entry.seq.to_s
+        print "." if count%10_000==0 and @verbose
+        count +=1
       end
+      puts " Done" if @verbose
     end
 
-    def sequence_alignment clusters
-      clusters.each do |id, list| # threach
-        if list.size > 5
-          seq = ""
-          list.each do |hash|
-            seq << ">#{hash[:name]}\n"
-            if hash[:strand] == "+"
-              seq << "#{@sequences[hash[:name]]}\n"
-            elsif hash[:strand] == "-"
-              seq << "#{@sequences[hash[:name]].revcomp}\n"
-            else
-              abort "Unknown strand #{hash[:strand]}"
-            end
-          end
-          cmd = "echo -e \"#{seq}\" | #{@clustalo} -i - --outfmt fa "
-          cmd << "--output-order tree-order"
-          align = Cmd.new cmd
-          align.run
-          File.open("cluster#{id}.fa", "wb") do |out|
-            out.write align.stdout
-          end
-        end
-      end
+    def cluster file, id
+      puts "clustering #{file}" if @verbose
+      cluster = Cluster.new @threads, @verbose, id
+      return cluster.run file
+    end
+
+    def consensus msa, scores, output
+      cons = Consensus.new(@verbose)
+      return cons.run(msa, scores, output)
     end
 
     def load_scores files
@@ -96,8 +82,10 @@ module Transfuse
                           :header_converters => :symbol,
                           :converters => :all) do |row|
           name = row[:contig_name]
-          score = row[:score]
-          scores[name] = score
+          scores[name] = { :score => row[:score].to_f,
+                           :p_good => row[:p_good].to_f,
+                           :p_bases_covered => row[:p_bases_covered].to_f,
+                           :coverage => row[:coverage].to_f }
         end
       end
       return scores
@@ -107,13 +95,15 @@ module Transfuse
       filtered_files = []
       files.each_with_index do |file, index|
         new_filename = "#{File.basename(file, File.extname(file))}_filtered.fa"
-        unless File.exist?(new_filename)
+        if !File.exist?(new_filename) or File.stat(new_filename).size < 1
           File.open(new_filename, "wb") do |out|
-            puts "opening #{file}..."
+            puts "filtering #{file}..." if @verbose
             Bio::FastaFormat.open(file).each do |entry|
               contig_name = entry.entry_id
               contig_name = "contig#{index}_#{contig_name}"
-              if scores.key?(contig_name) and scores[contig_name] > 0.01
+              if scores.key?(contig_name) and
+                 scores[contig_name][:score] > 0.01 and
+                 scores[contig_name][:coverage] >= 1
                 out.write ">#{contig_name}\n"
                 out.write "#{entry.seq}\n"
               elsif !scores.key?(contig_name)
@@ -127,75 +117,105 @@ module Transfuse
       return filtered_files
     end
 
+    def transrate_consensus file, output, left, right
+      output = File.expand_path(output)
+      puts "transrate on #{file}" if @verbose
+      file = File.expand_path(file)
+      name = File.basename(file, File.extname(file))
+      dir = "transrate_#{name}"
+      Dir.mkdir(dir) unless Dir.exist?(dir)
+      Dir.chdir(dir) do
+        assembly = Transrate::Assembly.new(file)
+        transrater = Transrate::Transrater.new(assembly, nil, threads:@threads)
+        rename = "assembly_#{name}_score_optimisation.csv"
+        rm = transrater.read_metrics(left.join(','), right.join(','))
+        stats = rm.read_stats
+        File.rename("assembly_score_optimisation.csv", rename)
+        scores={}
+        assembly.each do |name, contig|
+          scores[name] = { :score => contig.score.to_f,
+                           :p_good => contig.p_good.to_f,
+                           :p_bases_covered => contig.p_bases_covered.to_f,
+                           :coverage => contig.coverage.to_f }
+        end
+        scores_file = "#{name}_scores.csv"
+        stats_file = "../#{name}_stats.txt"
+        puts "  writing scores" if @verbose
+        File.open(scores_file, "wb") do |out|
+          scores.each do |name, hash|
+            out.write "#{name}\t#{hash[:score]}\t#{hash[:p_good]}\t"
+            out.write "#{hash[:p_bases_covered]}\t#{hash[:coverage]}\n"
+          end
+        end
+        puts "  writing filtered fasta file" if @verbose
+        File.open(output, "wb") do |out|
+          assembly.each do |name, contig|
+            if contig.score.to_f > 0.01 and contig.coverage.to_f >= 1
+              out.write ">#{name}\n"
+              out.write "#{contig.seq.seq}\n"
+            end
+          end
+        end
+        puts "  writing stats" if @verbose
+        File.open(stats_file, "wb") do |out|
+          stats.each do |key, value|
+            out.write "#{key}\t#{value}\n"
+          end
+          out.write "assembly score:\t#{transrater.assembly_score}\n"
+          optimal = transrater.assembly_optimal_score("prefix")
+          out.write "optimal score :\t#{optimal[0]}\n"
+          out.write "cutoff        :\t#{optimal[1]}\n"
+        end
+      end
+    end
 
     def transrate files, left, right
       scores = {}
-      scores_file = "scores.csv"
+      shortname = ""
+      files.each do |n|
+        shortname << File.basename(n, File.extname(n))[0..4]
+      end
+      scores_file = "#{shortname}_scores.csv"
       if File.exist?(scores_file)
         puts "loading scores from file" if @verbose
         File.open(scores_file).each do |line|
-          name, score = line.chomp.split("\t")
-          scores[name] = score.to_f
+          name, score, p_good, p_bases_covered, coverage = line.chomp.split("\t")
+          scores[name] = { :score => score.to_f,
+                           :p_good => p_good.to_f,
+                           :p_bases_covered => p_bases_covered.to_f,
+                           :coverage => coverage.to_f }
         end
       else
         files.each_with_index do |fasta, index|
           puts "transrate on #{fasta}" if @verbose
-          assembly = Transrate::Assembly.new(fasta)
-          transrater = Transrate::Transrater.new(assembly, nil, threads:@threads)
-          transrater.read_metrics(left.join(','), right.join(','))
-          assembly.each do |name, contig|
-            name = "contig#{index}_#{name}"
-            scores[name] = contig.score
+          dir = "transrate_#{File.basename(fasta, File.extname(fasta))}"
+          Dir.mkdir(dir) unless Dir.exist?(dir)
+          Dir.chdir(dir) do
+            assembly = Transrate::Assembly.new(fasta)
+            transrater = Transrate::Transrater.new(assembly, nil, threads:@threads)
+            rename = "assembly#{index}_score_optimisation.csv"
+            transrater.read_metrics(left.join(','), right.join(','))
+            File.rename("assembly_score_optimisation.csv", rename)
+            assembly.each do |name, contig|
+              name = "contig#{index}_#{name}"
+              scores[name] = { :score => contig.score.to_f,
+                               :p_good => contig.p_good.to_f,
+                               :p_bases_covered => contig.p_bases_covered.to_f,
+                               :coverage => contig.coverage.to_f }
+
+            end
           end
         end
         File.open(scores_file, "wb") do |out|
-          scores.each do |name, score|
-            out.write "#{name}\t#{score}\n"
+          scores.each do |name, hash|
+            out.write "#{name}\t#{hash[:score]}\t#{hash[:p_good]}\t"
+            out.write "#{hash[:p_bases_covered]}\t#{hash[:coverage]}\n"
           end
         end
       end
       return scores
     end
 
-    def select_contigs clusters, scores
-      puts "selecting contigs" if @verbose
-      best = []
-      clusters.each do |cluster_id, list|
-        best_score = 0
-        best_contig = ""
-        list.each do |contig_name|
-          unless scores[contig_name]
-            abort "can't find #{contig_name} in scores hash\n"
-          end
-          if scores[contig_name] > best_score
-            best_score = scores[contig_name]
-            best_contig = contig_name
-          end
-        end
-        best << best_contig
-      end
-      return best
-    end
-
-    def output_contigs best, fasta, output
-      puts "writing contigs" if @verbose
-      # read in catted fasta sequences
-      sequences = {}
-      Bio::FastaFormat.open(fasta).each do |entry|
-        sequences[entry.entry_id] = entry.seq
-      end
-      File.open(output, "wb") do |out|
-        best.each do |contig_name|
-          if sequences.key?(contig_name)
-            out.write ">#{contig_name}\n"
-            out.write "#{sequences[contig_name]}\n"
-          else
-            puts "can't find #{contig_name} in #{fasta}"
-          end
-        end
-      end
-    end
-
   end
 
 end

data/lib/transfuse/version.rb

@@ -7,8 +7,8 @@ module Transfuse
   # Semantic Versioning 2.0 (http://semver.org/).
   module VERSION
     MAJOR = 0
-    MINOR = 1
-    PATCH = 4
+    MINOR = 4
+    PATCH = 2
     BUILD = nil
 
     STRING = [MAJOR, MINOR, PATCH, BUILD].compact.join('.')

data/lib/transfuse.rb

@@ -1,4 +1,5 @@
-require 'transfuse/cluster.rb'
-require 'transfuse/cmd.rb'
-require 'transfuse/transfuse.rb'
-require 'transfuse/version.rb'
+require 'transfuse/cluster'
+require 'transfuse/cmd'
+require 'transfuse/consensus'
+require 'transfuse/transfuse'
+require 'transfuse/version'

data/test/test_transfuse.rb

@@ -8,13 +8,13 @@ class TestTransfuse < Test::Unit::TestCase
   context 'transfuse' do
 
     setup do
-      @fuser = Transfuse::Transfuse.new 4
+      @fuser = Transfuse::Transfuse.new 4, true
     end
 
     teardown do
     end
 
-    should 'check for existence of files' do
+    should '1 check for existence of files' do
       list = []
       list << File.join(File.dirname(__FILE__), 'data', 'assembly1.fasta')
       list << File.join(File.dirname(__FILE__), 'data', 'assembly2.fasta')
@@ -22,7 +22,7 @@ class TestTransfuse < Test::Unit::TestCase
       assert_equal 2, files.length, "length"
     end
 
-    should "concatenate two files" do
+    should "2 concatenate two files" do
       list = []
       list << File.join(File.dirname(__FILE__), 'data', 'assembly1.fasta')
       list << File.join(File.dirname(__FILE__), 'data', 'assembly2.fasta')
@@ -36,71 +36,69 @@ class TestTransfuse < Test::Unit::TestCase
       end
     end
 
-    should "cluster fasta file" do
-      Dir.mktmpdir do |tmpdir|
+    should "3 cluster fasta file" do
+      # Dir.mktmpdir do |tmpdir|
+      tmpdir = Dir.mktmpdir
         Dir.chdir(tmpdir) do
           file = File.join(File.dirname(__FILE__), 'data', 'assembly1.fasta')
           hash = @fuser.cluster file
           assert_equal 250, hash.size, "output size"
         end
-      end
+      # end
     end
 
-    should "load scores from transrate output" do
+    should "4 load scores from transrate output" do
       files = []
       files << File.join(File.dirname(__FILE__), 'data', 'contig_scores1.csv')
       hash = @fuser.load_scores files
       assert_equal 99, hash.size
     end
 
-    should "filter contigs" do
+    should "5 run transrate on assembly files with reads" do
       files = []
-      scores = {}
-      files << File.join(File.dirname(__FILE__), 'data', 'assembly1.fasta')
-      scores["soap_contig173359"] = 1
-      scores["soap_contig38533"] = 0.5
-      scores["idba_contig44716"] = 0
-      new_list = @fuser.filter files, scores
-      assert_equal 1, new_list.length
-      cmd = "grep -c \">\" #{new_list.first}"
-      assert_equal 2, `#{cmd}`.chomp.split.first.to_i, "contigs"
+      left = []
+      right = []
+      files << File.join(File.dirname(__FILE__), 'data', 'assembly3.fasta')
+      left << File.join(File.dirname(__FILE__), 'data', 'left.fq')
+      right << File.join(File.dirname(__FILE__), 'data', 'right.fq')
+      # Dir.mktmpdir do |tmpdir|
+      tmpdir = Dir.mktmpdir
+        Dir.chdir(tmpdir) do
+          scores = @fuser.transrate files, left, right
+          assert_equal 100, scores.size, "scores size"
+        end
+      # end
     end
 
-    should "run transrate on assembly files with reads" do
+    should "6 filter contigs" do
       files = []
       left = []
       right = []
-      files << File.join(File.dirname(__FILE__), 'data', 'assembly3.fasta')
+      files << File.join(File.dirname(__FILE__), 'data', 'assembly1.fasta')
       left << File.join(File.dirname(__FILE__), 'data', 'left.fq')
       right << File.join(File.dirname(__FILE__), 'data', 'right.fq')
-      Dir.mktmpdir do |tmpdir|
+      # Dir.mktmpdir do |tmpdir|
+      tmpdir = Dir.mktmpdir
         Dir.chdir(tmpdir) do
           scores = @fuser.transrate files, left, right
-          assert_equal 100, scores.size, "scores size"
+          scores.each do |contig, score|
+            # puts "#{contig}\t#{score}"
+          end
+          new_list = @fuser.filter files, scores
+          assert_equal 1, new_list.length
+          cmd = "grep -c \">\" #{new_list.first}"
+          assert_equal 1, `#{cmd}`.chomp.split.first.to_i, "number of contigs"
         end
-      end
+      # end
+
     end
 
-    should "select contigs" do
-      clusters = {"0" => ["contig1", "contig2"], "1" => ["contig3", "contig4"]}
-      scores = { "contig1" => 0.2,
-                 "contig2" => 0.3,
-                 "contig3" => 0.4,
-                 "contig4" => 0.2 }
-      best = @fuser.select_contigs clusters, scores
-      assert_equal 2, best.size
-      assert_equal "contig2", best[0]
-      assert_equal "contig3", best[1]
+    should "7 get consensus of clusters" do
+
     end
 
-    should "output contigs" do
-      best = ["soap_contig173359", "oases_contig80246"]
-      file = File.join(File.dirname(__FILE__), 'data', 'assembly1.fasta')
-      Dir.mktmpdir do |tmpdir|
-        Dir.chdir(tmpdir) do
-          @fuser.output_contigs best, file, "out"
-        end
-      end
+    should "8 not fail when there are duplicated kmers in the input sequences" do
+
     end
 
   end

data/transfuse.gemspec

@@ -19,7 +19,7 @@ Gem::Specification.new do |gem|
   gem.add_dependency 'bio', '~> 1.4', '>= 1.4.3'
   gem.add_dependency 'fixwhich', '~> 1.0', '>= 1.0.2'
   gem.add_dependency 'bindeps', '~> 1.0', '>= 1.0.1'
-  gem.add_dependency 'transrate', '~> 1.0', '>= 1.0.0'
+  gem.add_dependency 'transrate', '~> 1.0', '>= 1.0.1'
 
   gem.add_development_dependency 'rake', '~> 10.3', '>= 10.3.2'
   gem.add_development_dependency 'turn', '~> 0.9', '>= 0.9.7'

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: transfuse
 version: !ruby/object:Gem::Version
-  version: 0.1.4
+  version: 0.4.2
 platform: ruby
 authors:
 - Richard Smith-Unna
@@ -94,7 +94,7 @@ dependencies:
         version: '1.0'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 1.0.0
+        version: 1.0.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -104,7 +104,7 @@ dependencies:
         version: '1.0'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 1.0.0
+        version: 1.0.1
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -210,7 +210,6 @@ extra_rdoc_files: []
 files:
 - ".gitignore"
 - Gemfile
-- Gemfile.lock
 - README.md
 - Rakefile
 - bin/transfuse
@@ -218,6 +217,7 @@ files:
 - lib/transfuse.rb
 - lib/transfuse/cluster.rb
 - lib/transfuse/cmd.rb
+- lib/transfuse/consensus.rb
 - lib/transfuse/transfuse.rb
 - lib/transfuse/version.rb
 - notes.md
@@ -248,7 +248,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.6
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: Merge assemblies

data/Gemfile.lock

@@ -1,87 +0,0 @@
-PATH
-  remote: .
-  specs:
-    transfuse (0.1.1)
-      bindeps (~> 1.0, >= 1.0.1)
-      bio (~> 1.4, >= 1.4.3)
-      fixwhich (~> 1.0, >= 1.0.2)
-      transrate (= 1.0.0.beta3)
-      trollop (~> 2.0)
-
-GEM
-  remote: https://rubygems.org/
-  specs:
-    ansi (1.5.0)
-    bindeps (1.1.2)
-      fixwhich (~> 1.0, >= 1.0.2)
-    bio (1.4.3.0001)
-    coveralls (0.8.1)
-      json (~> 1.8)
-      rest-client (>= 1.6.8, < 2)
-      simplecov (~> 0.10.0)
-      term-ansicolor (~> 1.3)
-      thor (~> 0.19.1)
-    crb-blast (0.6.4)
-      bindeps (~> 1.0, >= 1.0.3)
-      bio (~> 1.4, >= 1.4.3)
-      fixwhich (~> 1.0, >= 1.0.2)
-      threach (~> 0.2, >= 0.2.0)
-      trollop (~> 2.0)
-    docile (1.1.5)
-    domain_name (0.5.24)
-      unf (>= 0.0.5, < 1.0.0)
-    facade (1.0.6)
-    fix-trinity-output (1.0.0)
-      trollop (~> 2.0)
-    fixwhich (1.0.2)
-      pathname2 (~> 1.4, >= 1.4.4)
-    http-cookie (1.0.2)
-      domain_name (~> 0.5)
-    json (1.8.3)
-    mime-types (2.6.1)
-    minitest (4.7.5)
-    netrc (0.10.3)
-    pathname2 (1.7.3)
-      facade
-    rake (10.4.2)
-    rest-client (1.8.0)
-      http-cookie (>= 1.0.2, < 2.0)
-      mime-types (>= 1.16, < 3.0)
-      netrc (~> 0.7)
-    shoulda-context (1.2.1)
-    simplecov (0.10.0)
-      docile (~> 1.1.0)
-      json (~> 1.8)
-      simplecov-html (~> 0.10.0)
-    simplecov-html (0.10.0)
-    term-ansicolor (1.3.0)
-      tins (~> 1.0)
-    thor (0.19.1)
-    threach (0.2.0)
-    tins (1.5.2)
-    transrate (1.0.0.beta3)
-      bindeps (~> 1.1, >= 1.1.2)
-      bio (~> 1.4, >= 1.4.3)
-      crb-blast (~> 0.5, >= 0.5.0)
-      fix-trinity-output (~> 1.0, >= 1.0)
-      trollop (~> 2.0, >= 2.0.0)
-      yell (~> 2.0, >= 2.0.4)
-    trollop (2.1.1)
-    turn (0.9.7)
-      ansi
-      minitest (~> 4)
-    unf (0.1.4)
-      unf_ext
-    unf_ext (0.0.7.1)
-    yell (2.0.5)
-
-PLATFORMS
-  ruby
-
-DEPENDENCIES
-  coveralls (~> 0.7)
-  rake (~> 10.3, >= 10.3.2)
-  shoulda-context (~> 1.2, >= 1.2.1)
-  simplecov (~> 0.8, >= 0.8.2)
-  transfuse!
-  turn (~> 0.9, >= 0.9.7)