Are there laws of genome evolution?

Research in quantitative evolutionary genomics and systems biology led to the discovery of several universal regularities connecting genomic and molecular phenomic variables. These universals include the log-normal distribution of the evolutionary rates of orthologous genes; the power law-like distributions of paralogous family size and node degree in various biological networks; the negative correlation between a gene’s sequence evolution rate and expression level; and differential scaling of functional classes of genes with genome size. The universals of genome evolution can be accounted for by simple mathematical models similar to those used in statistical physics, such as the birth-death-innovation model. These models do not explicitly incorporate selection, therefore the observed universal regularities do not appear to be shaped by selection but rather are emergent properties of gene ensembles. Although a complete physical theory of evolutionary biology is inconceivable, the universals of genome evolution might qualify as ’laws of evolutionary genomics’ in the same sense ’law’ is understood in modern physics.

💡 Research Summary

**
The paper asks whether genome evolution obeys any laws comparable to those in physics and argues that several quantitative regularities uncovered by comparative genomics and systems biology can indeed be regarded as such. Four “universals” are highlighted: (1) the evolutionary rates of orthologous genes follow a log‑normal distribution that is virtually identical across bacteria, archaea and eukaryotes; (2) the sizes of paralogous gene families and the node‑degree distributions of diverse biological networks (protein‑protein interaction, metabolic, regulatory, etc.) display power‑law‑like behavior; (3) a robust negative correlation exists between a gene’s sequence‑level evolutionary rate and its expression level (or protein abundance); and (4) functional classes of genes scale with total genome size in a non‑linear, class‑specific manner (e.g., metabolic enzymes scale linearly, regulators scale quadratically).

To explain these patterns, the author presents a series of minimalist, physics‑inspired mathematical models that do not explicitly invoke natural selection. The central framework is the birth‑death‑innovation (BDI) model, which incorporates three elementary processes: gene duplication (birth), gene loss (death), and the acquisition of entirely new families (innovation, e.g., via horizontal transfer). By tuning the balance between these rates and allowing birth and death probabilities to depend on family size, the BDI model reproduces the observed power‑law distribution of gene‑family sizes and the differential scaling of functional categories. An extended version of the model links family‑size exponents to the scaling exponents of functional classes, a relationship that is confirmed in prokaryotic genome data.

The negative correlation between evolutionary rate and expression is addressed through the “misfolding‑driven” hypothesis. Here, the dominant fitness cost of mutations is assumed to be protein misfolding, which is especially detrimental for highly expressed proteins. Embedding this assumption in an off‑lattice protein‑folding simulation yields both the log‑normal distribution of rates and the rate‑expression anticorrelation, suggesting that the physics of protein stability can generate these genome‑level patterns without invoking adaptive explanations.

The paper also examines network architecture. Biological networks are shown to be scale‑free, small‑world, modular, and hierarchically organized—features that also arise in non‑biological systems such as the Internet. Preferential attachment and duplication‑subfunctionalization mechanisms, both neutral with respect to fitness, generate the observed degree distributions and modularity. Empirical support comes from mutation‑accumulation lines of Caenorhabditis elegans, where network topology is indistinguishable from that of natural isolates despite the near‑absence of selection, indicating that global network features are emergent rather than selected.

These findings lead to the central claim that the universals of genome evolution are emergent properties of large ensembles of weakly interacting “particles” (genes or proteins), analogous to the ideal‑gas approximation in statistical physics. While higher‑order interactions such as epistasis are acknowledged, the success of simple stochastic models demonstrates that many macro‑evolutionary patterns can be captured by statistical ensemble theory.

The author concedes that a complete physical theory of evolution is impossible because evolution is historically contingent and involves adaptive tinkering. Nevertheless, the existence of reproducible quantitative regularities and the ability of parsimonious, selection‑free models to explain them suggest that a set of “laws of evolutionary genomics” may be attainable, occupying a status comparable to physical laws in their generality and predictive power.