Find gap regions containing Ns in fasta files

print_gap_regions.rb

#!/usr/bin/env ruby
 
require 'rubygems'
require 'bio'
 
def find_gaps(file,buffer=200)
  re = Regexp.compile("[ATGC]{#{buffer}}N+[ATGC]{#{buffer}}")
  return Bio::FastaFormat.open(file).inject(Hash.new([])) do |hash,entry|
    hash[entry.definition.split.first] = entry.seq.scan(re)
    hash
  end
end

# This bit prints the sequences to STDOUT as a series of fasta entries
find_gaps(ARGV[0]).each do |scaffold,sequences|
  sequences.each_with_index {|s,i| puts s.to_fasta("#{scaffold} gap#{i+1}")}
end