genome2prolog.rb

#!/usr/bin/env ruby
# Point it to the directory where the genbank files are and it will generate
# Prolog representations of the files.

require 'rubygems'
require 'bio'

class String
  def to_prolog
    "'" + self.gsub("'",'\\\\\'') + "'"
  end
end

class Array
  def to_prolog
    "[" + (self.map { |e| e.to_prolog}).join(',') + "]"
  end
end

class Genome2Prolog
  def initialize(datadir, outputfile)
    @datadir, @outputfile = datadir, outputfile
    @gene_facts = []
    @genome_facts = []
    process_dir(datadir)
    write_prolog_file(outputfile)
  end
  
  def process_dir(dir)
    Dir.open(dir).each do |f|
      process_individual_genes(Bio::FastaFormat.open(dir+"/"+f)) if f =~ /.*\.ffn/
      process_genomes(Bio::FastaFormat.open(dir+"/"+f)) if f =~ /.*\.fna/
    end
  end
  
  def process_genomes(genomes)
    genomes.each { |genome| @genome_facts << genome2prolog(genome) }
  end
  
  def process_individual_genes(genes)
    genes.each { |gene| @gene_facts << gene2prolog(gene) }
  end
  
  def gene2prolog(gene)
    defline = Bio::FastaDefline.new(gene.definition) 
    if defline.list_ids[0][2] =~ /:(c?)(\d+)-(\d+)/
      "gene('#{defline.to_s}',#{$2},#{$3},#{($1=="c" ? "complementary" : "primary")}, #{gene.data.gsub("\n", "").downcase.split("").to_a.to_prolog})."
    end
  end

  def genome2prolog(genome)
    defline = Bio::FastaDefline.new(genome.definition)
    "genome(#{defline.list_ids[0][1]}, #{genome.data.gsub("\n", "").downcase.split("").to_a.to_prolog})."
  end
  
  def write_prolog_file(filename)
    File.open(filename, "w") do |f|
      f << @gene_facts.sort.join("\n")
      f << "\n%% Complete genome: \n"
      f << @genome_facts.sort.join("\n")
    end
  end
end

datadir = ARGV.shift
outputfile = ARGV.shift
g2p = Genome2Prolog.new(datadir,outputfile)