Evolution: complexity, protein language & information (Introduction)

by David Turell , Friday, October 09, 2020, 19:49 (1293 days ago) @ David Turell

A very interesting review of the research:

https://inference-review.com/article/the-internal-language-of-proteins

"THE QUESTION OF WHETHER proteins share a common language across the major branches of life is an intriguing one. In their paper, Lijia Yu et al. investigated the arrangements of protein domains, considering them as a kind of grammar that orders the internal structure of proteins. They found that only a small subset of all the possible orderings occur, and that the properties of such a grammar are common across most of the major taxonomic groupings.

***

"The concept of a language for protein assembly has been widely discussed in different contexts. This should come as no surprise, because biology is increasingly being viewed as an information science.

***

"The researchers found that there was a reasonably consistent difference between random sets and actual genomes, which they interpreted as the minimal information gain required to maintain a functioning living cell. Although not immediately obvious from first principles, this finding makes sense in light of the data. It is conceivable that the information encoded in genomes might not be crucially dependent on domain architecture, so that information could be randomized without issue.

***

"Yu et al. find that animal genomes “show the highest information gain among the analyzed groups.” They claim that this accords “with the notion that domain architectures in animals are more elaborate and evolve under stronger constraints than those in other organisms.”

***

"A few decades ago it was thought that almost all genes were ancient and that modern genes were derived through processes such as duplication and divergence.11 It has since become clear that many genes emerged from noncoding sequences over the course of evolution. Such genes would presumably begin with a minimal number of functional units and accumulate additional domains over time. The resulting protein domain architectures reflect the relative contributions of de novo gene processes and the number of domain shuffling events that are retained.

"Two conclusions can be drawn from the work of Yu et al. First, there is apparently an information increase within proteins at the level of domain arrangement, which is associated with functional cells. Second, complexity can be objectively measured. This paper adds to the growing body of evidence that there have been genuine increases in complexity over the course of evolutionary history, and that this is particularly evident in animals. Biologists have become so accustomed to considering notions of human uniqueness as thoroughly debunked that any hint of so-called progress within evolution is treated with great skepticism. Whatever one makes of such a loaded term, increased complexity in some lineages is observable across multiple biological features, including protein domain architecture.

Comment: This study review relates to the steady increase in information and complexity during the process of evolution. Note the bold. dhw take notice. It is a key to understanding the real process of how evolution advances.