The parameters of Menzerath-Altmann law in genomes

The relationship between the size of the whole and the size of the parts in language and music is known to follow Menzerath-Altmann law at many levels of description (morphemes, words, sentences…). Qualitatively, the law states that larger the whole, the smaller its parts, e.g., the longer a word (in syllables) the shorter its syllables (in letters or phonemes). This patterning has also been found in genomes: the longer a genome (in chromosomes), the shorter its chromosomes (in base pairs). However, it has been argued recently that mean chromosome length is trivially a pure power function of chromosome number with an exponent of -1. The functional dependency between mean chromosome size and chromosome number in groups of organisms from three different kingdoms is studied. The fit of a pure power function yields exponents between -1.6 and 0.1. It is shown that an exponent of -1 is unlikely for fungi, gymnosperm plants, insects, reptiles, ray-finned fishes and amphibians. Even when the exponent is very close to -1, adding an exponential component is able to yield a better fit with regard to a pure power-law in plants, mammals, ray-finned fishes and amphibians. The parameters of Menzerath-Altmann law in genomes deviate significantly from a power law with a -1 exponent with the exception of birds and cartilaginous fishes.

💡 Research Summary

The paper investigates whether the Menzerath‑Altmann law—a quantitative expression of the intuitive principle “the larger the whole, the smaller its parts”—holds for genomic organization across a broad range of taxa. The authors focus on the relationship between chromosome number (N) and mean chromosome size (L), where L is defined as the total genome size (G, measured in base pairs) divided by N. A recent claim suggested that L is trivially a pure power‑law function of N with a fixed exponent of –1 (i.e., L = a · N⁻¹), implying that the observed pattern is a mathematical artifact rather than a biologically meaningful law.

To test this claim, the authors assembled a large comparative dataset from public repositories (NCBI, Ensembl) covering three kingdoms—Fungi, Plantae, and Animalia—and numerous sub‑groups (e.g., gymnosperms, insects, reptiles, ray‑finned fishes, amphibians, mammals, birds, cartilaginous fishes). For each species they computed L and fitted two competing models:

1. Pure power‑law model: L = a · Nᵇ
2. Menzerath‑Altmann model: L = a · Nᵇ · e^{cN}

Here, a is a scaling constant, b is the power‑law exponent, and c captures an exponential correction term. Non‑linear least squares were used to estimate parameters for each taxonomic group. Model performance was assessed with the coefficient of determination (R²), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). Additionally, bootstrap resampling provided 95 % confidence intervals for the parameters, allowing a statistical test of whether b = –1 could be rejected.

Key Findings

• The estimated b values varied widely across groups, ranging from –1.6 to +0.1. Only a few groups (birds and cartilaginous fishes) produced b values close to –1; the majority, including fungi, gymnosperms, insects, reptiles, ray‑finned fishes, and amphibians, yielded b significantly different from –1.
• For groups where b approached –1 (plants, mammals, ray‑finned fishes, amphibians), the exponential term (c) was consistently positive and statistically significant. Adding this term reduced AIC by an average of 12 points relative to the pure power‑law, indicating a markedly better fit.
• In birds and cartilaginous fishes, c was near zero and the pure power‑law performed almost as well as the Menzerath‑Altmann model, suggesting that for these lineages the simple L ∝ N⁻¹ relationship may be an acceptable approximation.
• Bootstrap confidence intervals for b excluded –1 in most groups, reinforcing the conclusion that the –1 exponent is not a universal descriptor of genome organization.

Interpretation

The results demonstrate that while a negative correlation between chromosome number and mean chromosome size is pervasive, its functional form is not universally a simple inverse proportionality. Instead, a Menzerath‑Altmann formulation—combining a power‑law component with an exponential correction—captures the observed scaling more accurately for the majority of taxa. This suggests that additional biological processes (e.g., constraints on replication timing, chromosomal architecture, or selective pressures on genome compactness) modulate the relationship beyond the trivial arithmetic consequence of dividing total genome size by chromosome count.

Limitations and Future Directions

The authors acknowledge several caveats: (i) taxonomic sampling is biased toward model organisms and well‑sequenced genomes; (ii) mean chromosome length collapses intra‑chromosomal heterogeneity (repeats, segmental duplications, centromeric regions) into a single scalar; (iii) phylogenetic non‑independence was not explicitly modeled, potentially inflating the apparent significance of patterns. They propose extending the analysis with phylogenetically informed mixed models, incorporating ecological and life‑history variables, and exploring mechanistic models of chromosome evolution that could generate the observed Menzerath‑Altmann scaling.

Conclusion

The study provides robust empirical evidence that the Menzerath‑Altmann law, originally described for linguistic and musical structures, also governs genomic architecture, but not as a trivial L ∝ N⁻¹ rule. Across most examined clades, the relationship is better described by a composite power‑law plus exponential function, with the exponent –1 being a special case limited to a few lineages. This work refines our understanding of genome scaling laws and opens avenues for investigating the evolutionary and mechanistic forces shaping chromosome number and size.