The genetic code is not universal

Recently, a new genetic code with 62 sense codons, coding for 21 amino acids, and only 2 termination codons has been identified in archaea. The authors argue that the appearance of this variant of the genetic code is due to the relatively recent and complete recoding of all UAG stop codons to codons encoding for pyrrolysine. I re-evaluate this discovery by presenting arguments that favour the early, i.e. ancestral, appearance of this variant of the genetic code during the origin of the genetic code itself. These arguments are capable of supporting that during the origin of the organization of the genetic code, at least two versions of the genetic code evolved in the domain of the Archaea. Thus, the genetic code would not be absolutely universal.