In their PLOS One paper titled Stylus: A System for Evolutionary Experimentation Based on a Protein/Proteome Model with Non-Arbitrary Functional Constraints, Douglas D. Axe, Brendan W. Dixon and Philip Lu of the Biologic Institute, introduced new software known as Stylus. Stylus is intended to more realistically model proteins and evolutionary constraints with more accurate linkage of protein structure to function. The nature of some current models, compromises accurate assessments of protein transition possibilities and the mapping of sequences to changes in function.
Human languages have been used for the purpose of analogy to biological polymers like polypeptides. The analogous nature of letters to amino acids and whole sentences to function, suggests linguistic models. Stylus however is based, not on alphabets common to European languages, but rather on characters of Chinese origin. This is intended to generate more relevant biological models.
The intent of Stylus is to represent proteins which have a 20 letter alphabet. Unlike DNA, the manner in which proteins fold is critical to their function because protein structure (a three dimensional protein property) is essential to understanding protein function whereas sequencing and nucleotide identity explain DNA coding properties. The utility of Chinese pictograms lies in their analogy to structure. Quoting from the paper:
One such insight is that protein-like models (in contrast to RNA models) tend to show sparse connectivity between regions of sequence space that encode different structures . In other words, stepwise paths through sequence space that accomplish a structural transformation without passing through unstructured intermediates appear to be rare. This clearly fits expectations for real proteins, where reorganization of core structure would seem to require complete loss of structure (and therefore function) along the way . It also fits experimental observations, which show that the expected deterioration is common not only for transitions between different folds  but also, more surprisingly, for transitions between different sequences encoding the same fold .
The authors are pointing out that gradual changes, that would produce new protein innovations, are made problematic by properties innate to proteins. Even small alterations of proteins can radically alter protein structure. Effects of slight sequence changes on protein folding also are difficult to predict. Differences between protein families can entail few transition possibilities. A natural inference is an expectation of distinct structural origins for different functional groups of proteins.