genomer-plugin-view 0.0.2

data/genomer-plugin-view.gemspec

@@ -0,0 +1,34 @@
+# -*- encoding: utf-8 -*-
+require File.expand_path("../lib/genomer-plugin-view/version", __FILE__)
+
+Gem::Specification.new do |s|
+
+  s.name        = "genomer-plugin-view"
+  s.version     = GenomerViewPlugin::VERSION
+  s.platform    = Gem::Platform::RUBY
+  s.homepage    = "http://github.com/michaelbarton/genomer-plugin-view"
+  s.license     = "MIT"
+  s.authors     = ["Michael Barton"]
+  s.email       = ["mail@michaelbarton.me.uk"]
+  s.summary     = %Q{Provide different views of scaffold.}
+  s.description = %Q{Convert genome scaffold into different sequence format views}
+
+  s.required_rubygems_version = "~> 1.8.0"
+  s.rubyforge_project         = "genomer-view-plugin"
+
+  s.add_dependency "genomer", ">= 0.0.5"
+
+  # Specs
+  s.add_development_dependency "rspec",                   "~> 2.9.0"
+  s.add_development_dependency "rr",                      "~> 1.0.4"
+  s.add_development_dependency "scaffolder-test-helpers", "~> 0.4.1"
+  s.add_development_dependency "heredoc_unindent",        "~> 1.1.2"
+
+  # Features
+  s.add_development_dependency "cucumber", "~> 1.1.9"
+  s.add_development_dependency "aruba",    "~> 0.4.11"
+
+  s.files        = `git ls-files`.split("\n")
+  s.executables  = `git ls-files`.split("\n").map{|f| f =~ /^bin\/(.*)/ ? $1 : nil}.compact
+  s.require_path = 'lib'
+end

data/lib/genomer-plugin-view/agp.rb

@@ -0,0 +1,62 @@
+require 'genomer'
+
+class GenomerPluginView::Agp < Genomer::Plugin
+
+  def run
+    header = "##agp-version	2.0"
+    entries.unshift(header).join("\n") + "\n"
+  end
+
+  def locations(seq,regex)
+    seq.upcase.enum_for(:scan, regex).map do
+      (Regexp.last_match.begin(0)+1)..(Regexp.last_match.end(0))
+    end
+  end
+
+  def entries
+    cumulative_length  = 1
+    count   = 0
+    contigs = 0
+
+    scaffold.map do |entry|
+      case entry.entry_type
+      when :unresolved then
+        length = entry.sequence.length
+        count += 1
+        start = cumulative_length
+        stop  = cumulative_length += length
+        gap(start, stop - 1, count, 'specified')
+      when :sequence   then
+        seq = entry.sequence.upcase
+        seq_regions = locations(seq,/[^N]+/).map{|i| [:contig,i]}
+        gap_regions = locations(seq,/N+/).map{|i| [:gap,i]}
+        entries = (seq_regions + gap_regions).sort_by{|_,loc| loc.to_a}
+
+        entries.map do |(type,location)|
+          count += 1
+          length = (location.end - location.begin)
+          entry = case type
+                  when :contig then
+                    contigs += 1
+                    contig(length, cumulative_length, count, contigs)
+                  when :gap    then
+                    start = cumulative_length
+                    stop  = cumulative_length + length
+                    gap(start, stop, count, 'internal')
+                  end
+          cumulative_length += length + 1
+          entry
+        end
+      end
+    end
+  end
+
+  def contig(length, cum, count, no)
+    %W|scaffold #{cum} #{cum + length} #{count} W #{sprintf("contig%05d",no)} 1 #{length+1} +| * "\t"
+  end
+
+  def gap(start, stop, count, type)
+    %W|scaffold #{start} #{stop} #{count} N #{stop - start + 1} scaffold yes #{type}| * "\t"
+  end
+
+end

data/lib/genomer-plugin-view/fasta.rb

@@ -0,0 +1,36 @@
+require 'genomer'
+
+class GenomerPluginView::Fasta < Genomer::Plugin
+
+  def run
+    if flags[:contigs]
+      flags.delete(:contigs)
+
+      sequence.
+        split(/[Nn]+/).
+        map{|s| Bio::Sequence.new(s) }.
+        each_with_index.
+        map{|s,i| s.output(:fasta,:header => header(sprintf("contig%05d",i+1))) }.
+        join
+    else
+      Bio::Sequence.new(sequence).output(:fasta,:header => header(identifier))
+    end
+  end
+
+  def header(identifier)
+    (identifier + ' ' + header_flags).strip
+  end
+
+  def identifier
+    flags[:identifier] ? flags.delete(:identifier) : '.'
+  end
+
+  def header_flags
+    flags.map{|k,v| "[#{k}=#{v}]" }.join(' ')
+  end
+
+  def sequence
+    scaffold.map{|entry| entry.sequence}.join
+  end
+
+end

data/lib/genomer-plugin-view/gff_record_helper.rb

@@ -0,0 +1,61 @@
+require 'bio'
+
+module GenomerPluginView::GffRecordHelper
+
+
+  DEFAULT_GFF_MAPPING = {'product' => 'product', 'Note' => 'note' }
+
+  GFF_TO_TABLE = {
+    'gene' => {
+      'ID'   => 'locus_tag',
+      'Name' => 'gene'
+    },
+    'CDS' => DEFAULT_GFF_MAPPING.merge({
+      'ID'        => 'protein_id',
+      'ec_number' => 'EC_number',
+      'function'  => 'function',
+    }),
+    'miscRNA' => DEFAULT_GFF_MAPPING,
+    'rRNA'    => DEFAULT_GFF_MAPPING,
+    'tmRNA'   => DEFAULT_GFF_MAPPING,
+    'tRNA'    => DEFAULT_GFF_MAPPING
+  }
+
+  def negative_strand?
+    self.strand == '-'
+  end
+
+  def coordinates
+    if negative_strand?
+      [self.end,self.start,self.feature]
+    else
+      [self.start,self.end,self.feature]
+    end
+  end
+
+  def to_genbank_table_entry
+
+    delimiter = "\t"
+    indent    = delimiter * 2
+
+    entries = table_attributes.inject([coordinates]) do |array,atr|
+      array << atr.unshift(indent)
+    end
+    return entries.map{|line| line * delimiter} * "\n" + "\n"
+  end
+
+  def valid?
+    GFF_TO_TABLE.include?(feature)
+  end
+
+  def table_attributes
+    raise Genomer::Error, "Unknown feature type '#{feature}'" unless valid?
+    attributes.map do |(k,v)|
+      k = GFF_TO_TABLE[feature][k]
+      k.nil? ? nil : [k,v]
+    end.compact
+  end
+
+end
+
+Bio::GFF::GFF3::Record.send(:include, GenomerPluginView::GffRecordHelper)

data/lib/genomer-plugin-view/mapping.rb

@@ -0,0 +1,14 @@
+require 'genomer'
+
+class GenomerPluginView::Mapping < Genomer::Plugin
+
+  def run
+    original = annotations.map(&:id).map(&:clone)
+    updated  = annotations(GenomerPluginView.convert_command_line_flags(flags)).map(&:id)
+
+    original.zip(updated).
+      map{|i| i.join("\t") }.
+      join("\n")
+  end
+
+end

data/lib/genomer-plugin-view/table.rb

@@ -0,0 +1,56 @@
+require 'genomer'
+require 'genomer-plugin-view/gff_record_helper'
+
+class GenomerPluginView::Table < Genomer::Plugin
+
+  def run
+    options = GenomerPluginView.convert_command_line_flags(flags)
+
+    header = ">Feature\t#{options[:identifier]}\tannotation_table\n"
+
+    attns = annotations(options)
+    attns = create_encoded_features(attns, options[:encoded]) if options[:encoded]
+
+    attns.inject(header) do |table,attn|
+      table << attn.to_genbank_table_entry
+    end
+  end
+
+  SUPPORTED_FEATURE_TYPES = ['CDS','rRNA','tRNA','miscRNA','tmRNA']
+
+  def create_encoded_features(genes,prefix)
+    features = genes.map do |gene|
+      feature = gene.clone
+      attrs   = Hash[feature.attributes]
+
+      if id = attrs['ID']
+        attrs['ID'] = (prefix.is_a?(String) ? prefix + id : id)
+      end
+
+      feature.feature = attrs['feature_type'] || 'CDS'
+
+      unless SUPPORTED_FEATURE_TYPES.include?(feature.feature)
+        raise Genomer::Error, "Unknown feature_type '#{feature.feature}'"
+      end
+
+      if feature.feature == "CDS"
+        name, prdt, ftn = attrs['Name'], attrs['product'], attrs['function']
+
+        if name
+          name = name.clone
+          name[0] = name[0].upcase
+          prdt, ftn = name,prdt
+        end
+
+        attrs.delete('Name')
+        attrs['product']  = prdt
+        attrs['function'] = ftn
+      end
+
+      feature.attributes = attrs.to_a.reject{|(_,value)| value.nil? }
+      feature
+    end
+    genes.zip(features).flatten
+  end
+
+end

data/lib/genomer-plugin-view/version.rb

@@ -0,0 +1,3 @@
+class GenomerViewPlugin
+  VERSION = "0.0.2"
+end

data/lib/genomer-plugin-view.rb

@@ -0,0 +1,29 @@
+require 'genomer'
+
+class GenomerPluginView < Genomer::Plugin
+
+  def run
+    self.class.fetch_view(arguments.shift).new(arguments,flags).run
+  end
+
+  def self.fetch_view(view)
+    require 'genomer-plugin-view/' + view
+    const_get(view.capitalize)
+  end
+
+  def self.convert_command_line_flags(flags)
+    flags.inject(Hash.new) do |hash,(k,v)|
+      k = case k
+      when :identifier                then k
+      when :prefix                    then k
+      when :generate_encoded_features then :encoded
+      when :reset_locus_numbering     then :reset
+      else nil
+      end
+
+      hash[k] = v if k
+      hash
+    end
+  end
+
+end

data/man/genomer-view-agp.ronn

@@ -0,0 +1,46 @@
+genomer-view-agp(1) -- Generate agp file views of scaffold
+==========================================================
+
+## SYNOPSIS
+
+`genomer view agp`
+
+## DESCRIPTION
+
+**Genomer-view-agp** produces an AGP view of a scaffold. This format shows the 
+positions of gaps and contigs in the scaffold. More details on this format can
+be found on the [AGP specification page](http://www.ncbi.nlm.nih.gov/projects/genome/assembly/agp/AGP_Specification.shtml "AGP Specification").
+
+This command converts the scaffold into AGP format as follows:
+
+  * contigs:
+    Contiguous sequences of non-N nucleotides generate corresponding sequence 
+    entries in the AGP file. Note: two sequence positioned next to each other 
+    in the scaffold file do not however produce a single contig entry.
+
+  * internal contig gaps:
+
+    Regions of N characters in scaffold sequences are converted 'scaffold' gaps
+    in the AGP file. The 'Linkage Evidence' field is set to "internal" and this
+    should be set to the correct AGP field type after generation. See the AGP
+    Specification for allowed evidence types.
+
+  * unresolved regions:
+
+    'Unresolved' entries in the 'scaffold.yml' file result in a **scaffold**
+    gap entry in AGP file. The 'Linkage Evidence' field is set to "specified"
+    and this should be set to the correct AGP field type after generation. See
+    the AGP Specification for allowed evidence types.
+
+## EXAMPLES
+
+    $ genomer view agp
+
+## BUGS
+
+**Genomer-view** is written in Ruby and depends on the genomer gem. See the 
+Gemfile in the genomer-plugin-view gem install directory for version details.
+
+## COPYRIGHT
+
+**Genomer** is Copyright (C) 2012 Michael Barton <http://michaelbarton.me.uk>

data/man/genomer-view.ronn

@@ -0,0 +1,153 @@
+genomer-view(1) -- Generate file format views of scaffold and annotations
+=========================================================================
+
+## SYNOPSIS
+
+`genomer view` <flatfile-type> [<options>...]
+
+## DESCRIPTION
+
+**Genomer-view** assembles the scaffold and associated annotations to produce 
+common database file formats. The generated file format view is specified by 
+the **flat-file** argument.
+
+## OPTIONS
+
+  * `--identifier`=[<identifier>]:
+    The sequence identifier to include in generated flatfile outputs.
+
+  * `--strain`=[<strain>]:
+    The strain of the source organism.
+    
+  * `--organism`=[<organism>]:
+    The genus and species, enclosed in single quotes, of the source organism.
+
+  * `--prefix`=[<gene-prefix>]:
+    Prepend all ID attributes from the annotation file with <gene-prefix> in 
+    the generated output.
+
+  * `--reset_locus_numbering`:
+    Reset gene ID to begin at 1 from the start of the sequence in the generated 
+    output file.
+
+  * `--generate_encoded_features`=[<feature-prefix>]:
+    Generate corresponding 1:1 encoded feature entries from the genes entries 
+    in the annotation file. These will commonly be CDS entries but RNA type 
+    entries are also supported. The feature IDs are generated from the 
+    corresponding gene ID prefixed with the <feature-prefix>.
+
+## GFF NINTH COLUMN ATTRIBUTES
+
+The annotation file should be in GFF3 format and contain the annotations for 
+the scaffolded contigs. The default location for this file is
+**assembly/annotations.gff**. The following attributes in the GFF3 file are 
+treated specially by genomer when generating flat file output.
+
+### GFF DEFINED ATTRIBUTES
+
+These attributes have a predefined meaning in the GFF specification. These all 
+begin with an upper case letter.
+
+  * `ID`:
+    Used to specify the ID of annotations in the output. If the 
+    `--generate_encoded_features` option is passed, the encoded features have 
+    an ID generated from this field prefixed with the <feature-prefix> 
+    argument. This field should be unique in the annotation file.
+
+  * `Name`:
+    Used to specify the four letter annotation name, e.g. pilO. The lower case 
+    version is used for gene names. If the `--generate_encoded_features` option 
+    is passed, additonal encoded feature entries have the `product` field 
+    generated from this capitalised version of this attribute. This need not be 
+    unique in the file.
+
+  * `Note`:
+    Used to populate the **Note** field for entries when the 
+    `--generate_encoded_features` option is passed.
+
+### GENOMER ATTRIBUTES
+
+These attributes are specific to genomer and should begin with a lower case 
+letter. Many of these attributes have a corresponding relationship with fields 
+in genbank table format, however a caveat to this is outlined in the next 
+section.
+
+  * `product`:
+    Used to populate the **product** field for encoded features when the 
+    `--generate_encoded_features` option is passed. If the **Name** attribute 
+    is also present then the **funtion** field is instead populated with this 
+    value.
+
+  * `entry_type`:
+    When the gene product is not a CDS this field can be used, when the 
+    `--generate_encoded_features` option is passed, as the corresponding entry 
+    type instead of `CDS`. The genbank specification list examples for `rRNA`, 
+    `tmRNA`, `tRNA`, and `miscRNA`. If you require other feature type 
+    implemented, please contact me through the website below.
+
+  * `ec_number`:
+    Used to populate the protein **EC\_number** field for CDS entries when the 
+    `--generate_encoded_features` option is passed.
+
+  * `function`:
+    Used to populate the **function** field for encoded entries when the 
+    `--generate_encoded_features` option is passed. This is overwritten in the 
+    table output by the **product** attribute if both the **Name** and 
+    **product** attributes are present. See the next section for an explanation 
+    of this.
+
+
+### OVERLAP BETWEEN NAME, PRODUCT AND FUNCTION FIELD
+
+The genbank annotation table **product** fields may contain either a short four 
+letter name (e.g. pilO) or a longer gene description (e.g. pilus assembly 
+protein). This presents a problem where data may need to be juggled between the
+**Name**, **product** and **function** fields depending on what is information 
+is avaiable.
+
+Genomer view solves this problem by prioritising these fields in the following 
+order: **Name** > **product** > **function**. If the **Name** attribute is 
+present this will be used for the **product** field in the resulting genbank 
+table. If the **product** attribute is also present at the same time this will 
+instead be used to fill out the **function** field in the genbank table. If 
+only the **product** and **function** attributes are present then these then 
+map to corresponding fields in genbank table. 
+
+### RECOMMENDED FORMAT FOR ANNOTATIONS
+
+All entries should contain a unique `ID` attribute. A `Name` field be used 
+whenever an appropriate four letter name is also available, e.g. 'pilO'. The ID 
+field alone is sufficent for generating a gene-only annotation table. Generally 
+however you will want to generate the encoded annotations also using the 
+`--generate_encoded_annotations` command line flag..
+
+The majority of encoded annotations will be CDS entries but most genomes will 
+also contain RNA non-coding features. CDS annotations should contain either a 
+`product` and/or `Name` field to match the genbank requirements. In general it 
+may be easier to fill out all the `product` field for entries then add names 
+for entries where possible.
+
+## EXAMPLES
+
+Assemble the scaffold sequence into Fasta format. Set the Fasta header to 
+include the sequence identifier, strain, and organism.
+
+    $ genomer view fasta --identifier PRJNA68653 --strain='R124' \
+      --organism='Pseudomonas fluorescens'
+
+Assemble annotations into GenBank Table format suitable for use with `tbl2asn`. 
+Reset the gene order numbering to begin at the sequence start and prefix each 
+gene ID with 'I1A\_'. Set the organism identifier at the top of the feature 
+table to be 'PRJNA68653'.
+
+    $ genomer view table --identifier PRJNA68653 --reset_locus_numbering \
+        --prefix='I1A_'
+
+## BUGS
+
+**Genomer-view** is written in Ruby and depends on the genomer gem. See the 
+Gemfile in the genomer-plugin-view gem install directory for version details.
+
+## COPYRIGHT
+
+**Genomer** is Copyright (C) 2012 Michael Barton <http://michaelbarton.me.uk>