Bug fix:

  Make sure that transcripts created by the adaptor has their
  'edits_enabled' set to true.  This ensures that post-translational
  modifications takes effect, crucially translating selenocysteines into
  'U' rather than into stop codons ('*').