finishm 0.0.6 → 0.0.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: fd108980dbaa801a2989d17f35c59249cc7b16e2
-  data.tar.gz: 0533d456de22bcda37b7578f0860b50e96ebf6f6
+  metadata.gz: 97ecc9ab04ccb6e227d33d59a4219edfc0ba4614
+  data.tar.gz: 931671e63b1b82f20337dac7af39970d87f4c25a
 SHA512:
-  metadata.gz: c7c29e7353c0cc55a745a803fbeea5d87196a5e922c8551a0fb04b44bd928cd922d664d3e1cf1ad86539a5f090dfdfc77fad23667b2eda3bab822724c3da9b00
-  data.tar.gz: 56785e4404f2da18c76e399fc3747f71bb82fb259fcd42851b7559d7902f40b33d19b070f2c605b3b360de0fe182f2d2c4f5b4417a6430b3b2fd526cc561a0d9
+  metadata.gz: 48492828900c54d3fcbfd2e4ffae365476861a0cada3ed5b3c31917e9bc7bfc47a06055b1f1891003fad2896c5ce8b6576602a3c5e93c2ffab865001f3126065
+  data.tar.gz: 08d57f83a950941b8d202df5df572ee9735ba61a842e55d2c494169d24a248e7930c7034128981a1f8a09e4e09c17c161b6a662d5aa3090b09fc39720892dfe2

data/README.md

@@ -50,10 +50,10 @@ cd ../..
 
 ## Citation
 
-A manuscript describing the tools described here is currently in preparation. However, FinishM reuses code from velvet and BioRuby, so these tools may be worth citing.
+A manuscript describing the tools described here is currently in preparation. However, FinishM reuses code from velvet, clustalo and BioRuby, so these tools may be worth citing.
 
 ## Copyright
 
-Copyright (c) 2012-2014 Ben J. Woodcroft. See LICENSE.txt for
+Copyright (c) 2012-2015 Ben J. Woodcroft. See LICENSE.txt for
 further details.
 

data/VERSION

@@ -1 +1 @@
-0.0.6
+0.0.7

data/finishm.gemspec

@@ -2,17 +2,17 @@
 # DO NOT EDIT THIS FILE DIRECTLY
 # Instead, edit Jeweler::Tasks in Rakefile, and run 'rake gemspec'
 # -*- encoding: utf-8 -*-
-# stub: finishm 0.0.6 ruby lib
+# stub: finishm 0.0.7 ruby lib
 # stub: ext/mkrf_conf.rb
 
 Gem::Specification.new do |s|
   s.name = "finishm"
-  s.version = "0.0.6"
+  s.version = "0.0.7"
 
   s.required_rubygems_version = Gem::Requirement.new(">= 0") if s.respond_to? :required_rubygems_version=
   s.require_paths = ["lib"]
   s.authors = ["Ben J. Woodcroft"]
-  s.date = "2015-08-31"
+  s.date = "2015-09-07"
   s.description = "De-novo assemblies generally only provide draft genomes. FinishM is aimed at improving these draft assemblies."
   s.email = "donttrustben near gmail.com"
   s.executables = ["finishm"]

data/lib/assembly/contig_printer.rb

@@ -84,10 +84,6 @@ module Bio
         # beginning and ending probes
         begin
           example_path = anchored_connection.paths[0]
-          path_sequence, variants = sequences_to_variants_conservative(
-            anchored_connection.paths.collect{|path| path.sequence}
-            )
-          log.debug "Reference path has a sequence length #{path_sequence.length}" if log.debug?
 
           # Find start index
           begin_onode = example_path[0]
@@ -108,6 +104,20 @@ module Bio
             offset_of_begin_probe_on_path = begin_noded_read.offset_from_start_of_node
           end
 
+          path_sequence, variants = sequences_to_variants_conservative(
+            anchored_connection.paths.collect do |path|
+              seq = nil
+              begin
+                seq = path.sequence
+              rescue Bio::Velvet::Graph::OrientedNodeTrail::InsufficientLengthException => e
+                log.warn "Failed to join two contigs together because of inability to get sequence out of a trail of nodes. In the past this has been caused by low coverage thus making finishM inappropriate, so returning an unconnected contig now. However, this may be legitimate in the case of an unlucky misassembly at both ends of the contigs being joined, so please report this error to the author."
+                return nil, nil
+              end
+              seq
+            end
+            )
+          log.debug "Reference path has a sequence length #{path_sequence.length}" if log.debug?
+
           # Correct variants' positions to be relative to the full contig,
           # not just the path sequence
           variants.each do |variant|

data/lib/finishm/roundup.rb

@@ -191,27 +191,35 @@ the finishm_roundup_results directory in FASTA format. The procedure is then rep
                     )
                   second_sequence = genome.scaffolds[contig.sequence_index].contigs[0].sequence
                   log.debug "Found #{aconn.paths.length} connections between #{last_name} and #{current_name}" if log.debug?
-                  if aconn.paths.length == 0
-                    # when this occurs, it is due to there being a circuit in the path, so no paths are printed.
-                    # (at least for now) TODO: this could be improved.
-                    # Just arbitrarily put in 100 N characters, to denote a join, but no gapfill
-                    scaffold_sequence = scaffold_sequence+('N'*100)+rhs_sequence
-                  else
+                  connected = false
+                  if aconn.paths.length > 0
                     aconn.collapse_paths_to_maximal_coverage_path! if options[:gapfill_with_max_coverage]
-                    scaffold_sequence, variants = printer.ready_two_contigs_and_connections(
+                    scaffold_sequence2, variants = printer.ready_two_contigs_and_connections(
                       master_graph.graph,
                       scaffold_sequence,
                       aconn,
                       rhs_sequence,
                       master_graph.velvet_sequences
                       )
-                    # Print variants
-                    # TODO: need to change coordinates of variants, particularly when >2 contigs are joined?
-                    variants.each do |variant|
-                      variant.reference_name = superscaffold_name
-                      variants_file.puts variant.vcf(scaffold_sequence)
+                    if !scaffold_sequence2.nil?
+                      scaffold_sequence = scaffold_sequence2
+                      connected = true
+                      # Print variants
+                      # TODO: need to change coordinates of variants, particularly when >2 contigs are joined?
+                      variants.each do |variant|
+                        variant.reference_name = superscaffold_name
+                        variants_file.puts variant.vcf(scaffold_sequence)
+                      end
                     end
                   end
+                  if !connected
+                    # when this occurs, it is due to there being a circuit in the path, so no paths are printed.
+                    # (at least for now) TODO: this could be improved.
+                    # Just arbitrarily put in 100 N characters, to denote a join, but no gapfill
+
+                    # (or, it could be impossible to join because of low coverage resulting in inability to get sequence from the node trail)
+                    scaffold_sequence = scaffold_sequence+('N'*100)+rhs_sequence
+                  end
                 end
                 last_contig = contig
               end
@@ -228,6 +236,7 @@ the finishm_roundup_results directory in FASTA format. The procedure is then rep
                 genome.scaffolds[contig.sequence_index].name
               end
               output_file.puts ">#{superscaffold_name} #{descriptor} #{scaffold_names.join(':') }"
+              scaffold_sequence.gsub! '-', '' #remove dashes since these make things fail downstream
               output_file.puts scaffold_sequence
             end
 
@@ -292,7 +301,7 @@ the finishm_roundup_results directory in FASTA format. The procedure is then rep
       options[:interesting_probes].include?(probe2.number)
       connections.push gapfiller.gapfill(master_graph, probe1.index, probe2.index, options)
     end
-    log.debug "Found #{connections.length} connections" if log.debug?
+    log.debug "Found #{connections.length} connections gapfilling in scaffold #{scaffold_index}" if log.debug?
 
     all_variants = []
     num_gapfills = 0
@@ -329,18 +338,22 @@ the finishm_roundup_results directory in FASTA format. The procedure is then rep
   def piece_together_gapfill(printer, master_graph, first_sequence, aconn, second_sequence, gap_length, max_gapfill_paths, options={})
     scaffold_sequence = nil
     gapfilled = -1
-    if aconn.paths.length == 0 or aconn.paths.length > max_gapfill_paths
-      # No paths found. Just fill with Ns like it was before
-      scaffold_sequence = first_sequence + 'N'*gap_length + second_sequence
-      gapfilled = false
-    else
-      acon.collapse_paths_to_maximal_coverage_path! if options[:gapfill_with_max_coverage]
+    if aconn.paths.length != 0 and aconn.paths.length <= max_gapfill_paths
+      aconn.collapse_paths_to_maximal_coverage_path! if options[:gapfill_with_max_coverage]
       scaffold_sequence, variants = printer.ready_two_contigs_and_connections(
         master_graph.graph, first_sequence, aconn, second_sequence, master_graph.velvet_sequences
         )
-      gapfilled = true
+      if !scaffold_sequence.nil? #sometimes it is just impossible even if there is paths
+        scaffold_sequence.gsub!('-','') #remove gaps i.e. where the consensus is a gap
+        gapfilled = true
+      end
     end
-    scaffold_sequence.gsub!('-','') #remove gaps i.e. where the consensus is a gap
+    if gapfilled != true
+      # No paths found, or gapfilling failed at the final step Just fill with Ns like it was before
+      scaffold_sequence = first_sequence + 'N'*gap_length + second_sequence
+      gapfilled = false
+    end
+
     return scaffold_sequence, variants, gapfilled
   end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: finishm
 version: !ruby/object:Gem::Version
-  version: 0.0.6
+  version: 0.0.7
 platform: ruby
 authors:
 - Ben J. Woodcroft
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-08-31 00:00:00.000000000 Z
+date: 2015-09-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bio-ipcress