quantum events coexist with classical dynamics, including classical randomness, a hardly treated combination in physics – they should be understood in conjunction, for lack of a unified understanding of the respective fields. Finally, both forms of randomness contribute to the interference between levels of organization, due to regulation and integration, called bio-resonance in Buiatti and Longo (2013). Bio-resonance is part of the stabilization of organismal and ecosystemic structures, but may also be viewed as a form of proper biological randomness, when enhancing unpredictability. It corresponds to the interaction of different levels of organization, each possessing its own form of determination. Cell networks in tissues, organs as the result of morphogenetic processes, including intracellular organelles, are each given different forms of statistical or equational descriptions, mostly totally unrelated. However, an organ is made of tissues, and both levels interact during their genesis, as well as along their continual regeneration.
Second, for classical randomness, besides the cell-to-cell interactions within an organism (or among multicellular organisms in an ecosystem) or the various forms of bio-resonance (Buiatti and Longo 2013), let us focus on macromolecules’ Brownian motion. As a key aspect of this approach, we observe that Brownian motion and related forms of random molecular paths and interactions must be given a fundamental and positive role in biology. This random activity corresponds to the thermic level in a cell, thus to a relevant component of the available energy: it turns out to be crucial for gene expression.
The functional role of stochastic effects has long since been known in enzyme induction (Novick and Weiner 1957), and even theorized for gene expression (Kupiec 1983). In the last decade (Elowitz et al. 2002), stochastic gene expression finally came into the limelight. The existing analyses are largely based on the classical Brownian motion (Arjun and van Oudenaarden 2008), while local quantum effects cannot be excluded.
Increasingly, researchers have found that even genetically identical individuals can be very different, and that some of the most striking sources of this variability are random fluctuations in the expression of individual genes. Fundamentally, this is because the expression of a gene involves the discrete and inherently random biochemical reactions involved in the production of mRNA and proteins. The fact that DNA (and hence the genes encoded therein) is present in very low numbers means that these fluctuations do not just average away but can instead lead to easily detectable differences between otherwise identical cells; in other words, gene expression must be thought of as a stochastic process. (Arjun and van Oudenaarden 2008)
Different degrees of stochasticity in gene expression have been observed – with major differences in ranges of expression – in the same population (in the same organ or even tissue) (Chang et al. 2008).
A major consequence that we can derive from this view is the key role that we can attribute to this relevant component of the available energy, heath. The cell also uses it for gene expression instead of opposing to it. As a matter of fact, the view that DNA is a set of “instructions” (a program) proposes an understanding of the cascades from DNA to RNA to proteins in terms of a deterministic and predictable, thus programmable, sequence of stereospecific interactions (physico-chemical and geometric exact correspondences). That is, gene expression or genetic “information” is physically transmitted by these exact correspondences: stereospecificity is actually “necessary” for this (Monod 1970). The random movement of macromolecules is an obstacle that the “program” constantly fights. Indeed, both Shannon’s transmission and Turing’s elaboration of information, in spite of their theoretical differences, are both designed to oppose noise (see Longo et al. 2012b). Instead, in stochastic gene expression, Brownian motion, thus heath, is viewed as a positive contribution to the role of DNA.
Clearly, randomness, in a cell, an organism and an ecosystem, is highly constrained. The compartmentalization in a cell, the membrane, the tissue tensegrity structure, the integration and regulation by and within an organism, all contribute to restricting and canalizing randomness. Consider that an organism like ours has about 1013 cells, divided into many smaller parts, including nuclei: few physical structures are so compartmentalized. So, the very “sticky” oscillating and randomly moving macromolecules are forced within viable channels. Sometimes, though, it may not work, or it may work differently. This belongs to the exploration proper to biological dynamics: a “hopeful monster” (Dietrich 2003), if viable in a changing ecosystem, may yield a new possibility in ontogenesis, or even a new branch in evolution.
Activation of gene transcription, in these quasi-chaotic environments, with quasi-turbulent enthalpic oscillations of macro-molecules, is thus canalized by the cell structure, in particular in eukaryotic cells, and by the more or less restricted assembly of the protein complexes that initiate it (Kupiec 2010). In short, proteins can interact with multiple partners (they are “sticky”) causing a great number of combinatorial possibilities. Yet, protein networks have a central hub where the connection density is the strongest and this peculiar canalization further forces statistical regularities (Bork et al. 2004). The various forms of canalization mentioned in the literature include some resulting from environmental constraints, which are increasingly acknowledged to produce “downwards” or Lamarckian inheritance or adaptation, mostly by a regulation of gene expression (Richards 2006). Even mutations may be both random and not random, highly constrained or even induced by the environment. For example, organismal activities, from tissular stresses to proteomic changes, can alter genomic sequences in response to environmental perturbations (Shapiro 2011).
By this role of constrained stochasticity in gene expression, molecular randomness in cells becomes a key source of the cell’s activity. As we hinted above, Brownian motion, in particular, must be viewed as a positive component of the cell’s dynamics (Munsky et al. 2009), instead of being considered as “noise” that opposes the elaboration of the genetic “program” or the transmission of genetic “information” by exact stereospecific macro-molecular interactions. Thus, this view radically departs from the understanding of the cell as a “Cartesian Mechanism” occasionally disturbed by noise (Monod 1970), as we give heath a constitutive, not a “disturbing” role (also for gene expression and not only for some molecular/enzymatic reactions).
The role of stochasticity in gene expression is increasingly accepted in genetics and may be generally summarized by saying that “macromolecular interactions, in a cell, are largely stochastic, they must be given in probabilities and the values of these probabilities depend on the context” (Longo and Montévil 2015). The DNA then becomes an immensely important physico-chemical trace of history, continually used by the cell and the organism. Its organization is stochastically used, but in a very canalized way, depending on the epigenetic context, to produce proteins and, at the organismal level, biological organization from a given cell. Random phenomena, Brownian random paths first, crucially contribute to this.
There is a third issue that is worth being mentioned. Following Gould (1997), we recall how the increasing phenotypic complexity along evolution (organisms become more “complex”, if this notion is soundly defined) may be justified as a random complexification of the early bacteria along an asymmetric diffusion. The key point is to invent the right phase space for this analysis, as we hint: the tridimensional space of “biomass × complexity × time” (Longo and Montévil 2014b).
Note that the available energy consumption and transformation, thus entropy production, are the unavoidable physical processes underlying all biological activities, including reproduction with variation and motility, organisms’ “default state” (Longo et al. 2015). Now, entropy production goes with energy dispersal, which is realized by random paths, as with any diffusion in physics.
At the origin of life, bacterial exponential proliferation was (relatively) unconstrained, as other forms of life did not contrast it. Increasing diversity, even in bacteria, by random differentiation started the early divergence of life, a process that would never stop – and a principle for Darwin. However, it also produced competition within a limited ecosystem and a slower exponential growth.
Gould
