Hard coded script to produce genbank file for a genome from IMG output

transform.rb

# Produce some kind of genbank file output
#!/usr/bin/env ruby

require 'bio'
require 'fastercsv'

proteins = Bio::FlatFile.auto('annotation/proteins.faa').inject({}) do |h,p|
  h[p.definition.split.first] = p.seq
  h
end


fasta = Bio::FlatFile.auto('assembly/build.fna')
record = Bio::Sequence.new(fasta.first.seq)
record.definition = "Pseudomonas fluorescens R124, complete genome"
record.species = "Pseudomonas fluorescens R124"

record.features = []

FasterCSV.open('annotation/gene_list.csv','r',:headers => true).each do |e|
  coordinates = "#{e['Start Coord']}..#{e['End Coord']}"
  if e['Strand'] == '-'
    coordinates = "complement(#{coordinates})"
  end

  id = e['gene_oid']
  
  qualifiers = []
  qualifiers << Bio::Feature::Qualifier.new('gene', id)
  record.features << Bio::Feature.new('gene',coordinates,qualifiers.clone)

  if e['Description']
    qualifiers << Bio::Feature::Qualifier.new('function', e['Description'])
  end

  qualifiers << Bio::Feature::Qualifier.new('translation', proteins[id])

  record.features << Bio::Feature.new('CDS',coordinates,qualifiers)
  
   
end
File.open('R124.gb','w') do |out|
  out.print record.output(:genbank)
end