Principles and Mechanisms of Biological Evolution - Molecular Theory of the Living Cell - page 502

with the expression “weakest link” used in the Zeldovich et al. model (Zeldovich
et al. 2007a, b; Zeldovich and Shakhnovich 2008). We may refer to this suggestion
as the XYZ postulate. The XYZ postulate can be viewed as a species (or an
example) of HVC, since X represents a set of genes sharing some common features
and since the malfunctioning of X will lead to the inhibition of Y and hence to cell
death under environmental condition Z. In other words, the set of genes constituting
X can “behave in a coherent fashion within the genome” as required by HVC.
Furthermore, if we represent the probability of X as P(X), then the probability of
cell death can be calculated using Eq. 14.34 with P(x) replacing P nat (i) :

d ¼ d 0 1 ð P(X) Þ (14.37)

Although I believe Eq. 14.36 is biologically more realistic than the Zeldovich--
Shakhnovich model, Eq. 14.34, it is probably computationally less realistic, since P
(X) is the probability of a process which is likely “catalyzed” by a set of proteins
encoded in a gene family and hence unlikely to be as simple as the Boltzmann
distribution, Eq. 14.35, underlying the Zeldovich-Shankhnovich model (2007a, b;
Zeldovich and Shakhnovich 2008).

Observation (2) According to the cell language theory , cells utilize microscopic or
molecular language called the cell language or cellese whose principles are found
similar (or isomorphic) to those operating in the human language or humanese (Sect.
6.1.2 ) . Thus, genes (proteins) are akin to words, complexes of genes (proteins) are
akin to sentences, and a DNA molecule or a genome is akin to a book or a text of
instructions for survival under a given set of environmental conditions. A language,
either natural or cellular, is thought to obey the Law of Requisite Variety (LRV) (Ji
1997a): that is, no simple language can describe complex environment, and no
simple organism can survive complex environment (see Sect. 14.2 ). Therefore,
different genomes have different sizes most likely because they contain the
instructions that are needed for organisms to survive their environment that have
different amounts of complexities (as measured in bits; see algorithmic information
discussed in Sect. 4.3 ). For example, S. cerevisiae has a larger genome (with 6,218
predicted protein-coding regions) than E. coli (with 1,735 predicted structural
genes), probably because the complexity of the environment under which
S. cerevisiae can survive is much greater than the complexity of the environment
under which E. coli can survive. Based on this reasoning, we can conclude that the
size of a genome reflects the algorithmic complexity (or the algorithmic information
content) (see Fig. 4.1 ) of the environment under which the associated organism has
survived. Since the size of a genome is related to the AUC of the Huynen-Nimwegen
plot, we can therefore infer that the AUC of the Huynen-Nimwegen plot reflects the
complexity of the environment under which a species has survived:

(14.38)