Influence of effective population size on animal genome architecture – NeGA

The project is based on the study of biological groups where the impact of multiple population size variations between species is investigated. One of these groups consists of in silico organisms.

In addition to this approach focused on specific groups of interest, the project rapidly integrated a large-scale approach (e.g. metazoans) to study at this scale the relationship between population size and genome architecture. This large-scale approach combines both macroevolutionary and microevolutionary analyses.

The genomic traits studied in the project are diverse: polymorphism, protein evolution rates, splicing rates, base composition, size and diversity of the repeatome, etc. For example, as part of Alba Marino's PhD thesis, we assembled a dataset of 773 animal genomes. Transposable elements were annotated for 672 species.

The resources mobilised are first and foremost human: hiring and hosting of young researchers (4 PhD students) and one engineer. Finally, significant resources have been devoted to the acquisition of new genomic data.

1. Micro and macroevolutionary effects of Ne (M. Bastian's PhD thesis)

The objective was to test the nearly neutral theory on the evolution of coding sequences in 144 placental mammals, integrating inter-species divergence and intra-species polymorphism for ~8,000 orthologous genes (heterozygosity and dN/dS). The results confirm all predictions of the nearly neutral theory, for the first time at both micro and macroevolutionary scales (Bastian et al., 2026).

2. Ne, repeated elements and genome size (A. Marino's PhD thesis)

The objective was to test the relationship between Ne, genome size and the dynamics of repeated elements (~800 species and annotation of transposable elements). The results show no link between Ne and genome size, nor with transposable element activity. Ne is not a global evolutionary factor influencing genome size in animals (Marino et al., 2025).

3. Ne, gene composition and expression (F. Bénitière's PhD thesis)

The objective was to evaluate the extent to which the drift barrier influences the evolution of gene base composition (223 metazoan species with measurement of translational selection) as well as the complexity of produced isoforms (53 metazoan species, GTDrift database, Bénitière et al., 2024a). The results show strong variation in alternative splicing rates across animals, with an inverse correlation between Ne and alternative splicing rate: species with low Ne exhibit higher alternative splicing (Bénitière et al., 2024b). We also observe a weak signal of translational selection in most metazoans, with stronger selection in species with large Ne, but absent in some despite high Ne. Selection on synonymous codons may be modulated by other factors such as growth rate (Bénitière et al., 2025).

4. Ne and genomic robustness (J. Luiselli's PhD thesis)

Alongside comparative studies, a modelling approach was initiated to test in silico the effect of Ne on genome structure. Simulations of circular haploid genomes (bacterial model) and a generalisation to eukaryotic genomes (diploid, linear, sexual) were used. The results show that with the "bacterial" model, chromosomal rearrangements promote long-term adaptation and stabilisation of genome size (Banse et al., 2024). The pressure on genome size is modulated by mutation rate and population size (Luiselli et al., 2024). This result is confirmed by a mathematical model: these two parameters determine an equilibrium proportion of non-coding genome (Luiselli et al., 2025). We generalised these results to diploid sexually reproducing organisms with meiotic recombination: eukaryotic-type genomes respond to changes in mutation rate and Ne in an analogous manner.

Overall, these findings highlight a paradox. While the micro-evolutionary effects of Ne are confirmed by both empirical and theoretical data, only some of the patterns predicted by Lynch's theory are recovered at large scales. The relationship between Ne and genome size is clear in certain groups but is lost globally. For other genomic traits, a relationship is found (e.g. alternative splicing rate), but remains weak for others (e.g. translational selection).

These findings open perspectives on the regulation of transposable element (TE) activity and on why certain lineages undergo dramatic genome size increases. Selection may only act when genomes are actively expanding: active TEs could be more efficiently counter-selected in large populations, whereas inactive TEs may no longer affect fitness, leaving non-coding DNA without strong selective pressure for reduction. Counter-selection of active TEs can thus limit genome expansion in some populations.

Another perspective is the direct reuse of project-generated data. For instance, the ecological genomics model in subterranean environments will be applied to study the impact of phenotypic trait loss and trophic changes on symbiotic interactions.

The theoretical approaches developed during the project also offer rich prospects for exploring co-evolution between mutation rates and patterns, recombination landscapes, and genome size and complexity.

Finally, work conducted at the micro-macro interface serves as a starting point for new research projects centred on (i) building integrated micro-macro data matrices across all orthologous genes; and (ii) integrative modelling of short- and long-term Ne evolution across the phylogeny and its impact on coding sequence evolution.