Aevol-9: A simulation platform to decipher the evolution of genome architecture
Aevol-9 is a forward-in-time simulation platform that models populations with explicit genomes subject to diverse mutational operators, enabling researchers to investigate how evolutionary conditions influence genome architecture and organization.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Life on Earth is written in a code, a long string of chemical letters that tells a cell how to build itself and survive. For decades, biologists have understood how this code changes at the smallest level, swapping one letter for another or adding a single extra character. But the code is also organized into a larger structure, a blueprint of how genes are arranged, how many of them exist, and how much "empty space" sits between them. This larger structure, known as genome architecture, varies wildly across the tree of life. Some organisms pack their genetic instructions tightly into tiny circles, while others carry massive, sprawling genomes filled with vast stretches of non-coding DNA. Scientists have long wondered what forces shape these different blueprints. Is it the pressure of natural selection, which favors efficiency? Or is it the random roll of the dice, known as genetic drift, which can allow extra genetic material to accumulate simply by chance? Testing these ideas in the real world is nearly impossible, as evolution takes millions of years to play out, and observing the same species over such vast timescales is beyond human reach.
To solve this puzzle, a team of researchers has built a digital laboratory called Aevol-9. This is not a tool for observing the past, but a forward-looking simulator that allows scientists to watch evolution happen in fast-forward. In this virtual world, the researchers created populations of digital organisms, each carrying a structured, double-stranded genome. These digital life forms compete for survival, reproduce, and pass their genetic instructions to the next generation. The key innovation of this system is that it does not simply assign a score to a mutation based on a pre-written list. Instead, the software reads the organism's entire genome, translates it into a set of proteins, and then calculates how well those proteins help the organism meet a specific survival goal. If a mutation changes the genome in a way that messes up this process, the organism's fitness drops. If the change improves the process, fitness rises. This allows the system to naturally discover which genetic arrangements work best under different conditions, from tiny populations where chance rules, to massive ones where selection is strict.
The researchers used this platform to run a massive experiment to see how population size influences the shape of a genome. They started with a digital strain that had already evolved for ten million generations in a stable environment. They then split this population into three groups: one that stayed the same size, one that grew four times larger, and one that shrank to a quarter of its original size. They let these groups evolve for another 1.1 million generations, a span of time that would take real bacteria thousands of years to cover. The results were striking. In the small populations, where random chance plays a bigger role, the genomes grew larger, accumulating more non-coding DNA that did not seem to serve a direct purpose. In the large populations, where natural selection is more efficient at weeding out inefficiencies, the genomes shrank. The organisms in these large groups streamlined their genetic code, shedding the extra baggage to become more compact.
This finding supports a long-standing idea in biology called the mutational-hazard hypothesis, which suggests that larger populations are better at purging useless genetic material, while smaller populations are more likely to let it pile up. However, the simulation revealed a nuance that was not immediately obvious. While the total size of the genome changed dramatically, the amount of actual coding DNA—the parts that build proteins—remained surprisingly stable across all group sizes. The difference lay entirely in the non-coding regions. Furthermore, the way the genes were organized shifted. In the smaller populations, the genes tended to be spread out over longer stretches of DNA. In the larger, more efficient populations, the genes were packed more tightly together, often grouped into clusters that functioned as single units. This suggests that evolution does not just add or remove genes; it constantly rearranges the furniture of the genome to fit the constraints of the population.
The power of Aevol-9 lies in its ability to track every single change, from a tiny swap of a letter to a massive deletion of a large chunk of the genome. Because the researchers can see the entire history of every digital organism, they can trace exactly how these large-scale changes happened and why they stuck. The simulation showed that these changes are not random accidents but are driven by the interplay between mutation, selection, and the size of the population. The study suggests that the complex architecture of a genome is not a fixed trait but a dynamic response to the environment. By running thousands of these digital experiments, the team has provided a clear view of how life might organize its genetic instructions over deep time, offering a new way to understand the rules that govern the diversity of life on Earth.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.