Universal biology and the statistical mechanics of early life

Abstract

All known life on the Earth exhibits at least two non-trivial common features: the canonical genetic code and biological homochirality, both of which emerged prior to the Last Universal Common Ancestor state. This article describes recent efforts to provide a narrative of this epoch using tools from statistical mechanics. During the emergence of self-replicating life far from equilibrium in a period of chemical evolution, minimal models of autocatalysis show that homochirality would have necessarily co-evolved along with the efficiency of early-life self-replicators. Dynamical system models of the evolution of the genetic code must explain its universality and its highly refined error-minimization properties. These have both been accounted for in a scenario where life arose from a collective, networked phase where there was no notion of species and perhaps even individuality itself. We show how this phase ultimately terminated during an event sometimes known as the Darwinian transition, leading to the present epoch of tree-like vertical descent of organismal lineages. These examples illustrate concrete examples of universal biology: the quest for a fundamental understanding of the basic properties of living systems, independent of precise instantiation in chemistry or other media.This article is part of the themed issue 'Reconceptualizing the origins of life'.

Keywords: evolution; genetic code; homochirality; horizontal gene transfer.

Figure 1.

A (conjectured) brief sketch of the history of life. At present, life is divided into the three domains: Bacteria, Archaea and Eukaryota. Following the lineages of the three domains backward in time (solid lines), we find that they coalesce into the Last Universal Common Ancestor (LUCA), approximately 3.8 Gya. The dashed red line indicates the point in time where it is thought that the Darwinian transition occurred: before that, life was evolving in a communal way (progenote); after the Darwinian transition, life evolved as described by the Modern Synthesis. (Online version in colour.)

Figure 2.

Single run dynamics obtained by stochastic simulations of N=10⁴ digital organisms, initialized with a random distribution in genome space. Each organism is characterized by a binary genome composed of L=7 symbols, a mutation rate μ=0.1 and a HGT strength h=20. (a) Genome abundances (displayed using a scale of brown) of a typical system dynamics. In the beginning, organisms populate almost uniformly the genome space until a certain point in time (DT), after which the system exhibits vertical descent around the fittest genome. (b) Fitness landscape is represented in two dimensions (see the electronic supplementary material). (c) The total population fitness (green dots), obtained from the run displayed in (a), is plotted versus time. The DT corresponds to the inflection point of the corresponding spline. (d,e) Snapshots of the genome abundances corresponding to the progenote phase and the phase where the system clusters around the most fit genome (speciation). (Online version in colour.)

Figure 3.

Stochastic simulation results of an interacting population of digital organisms evolving under a double-peak fitness landscape for various values of HGT strengths h. Genomes A and B have equal (high) fitness, whereas other genomes have low fitness (see the electronic supplementary material). Other parameter values are as in figure 2. (a–c) Asymptotic system dynamics when h=0 (a), h=medium (b), h=high (c). Panel (d) summarizes this main effect. (e) The long-time difference between the number of individuals with genome A and those with genome B, displayed against the HGT strength h. The dashed red line corresponds to the critical HGT strength which separates two regimes: a linear increase and a nonlinear saturation. (Online version in colour.)