Posted by

Santiago Campillo Brocal

Biologist. Master in Molecular Biology and Biotechnology, Director of Muy Interesante Digital

Created:
1.06.2026 | 08:49

Updated:
1.06.2026 | 09:13

To understand why this matters, we must return to an idea that has been at the heart of molecular biology for more than half a century. The Neutral Theory of Molecular Evolution, formulated in the 1960s by Japanese geneticist Motoo Kimura, proposed something apparently reasonable: most of the genetic changes that accumulate in a population over time are neither good nor bad. They are simply neutral. Natural selection acts strongly on clearly harmful mutations, eliminates them, and lets the rest pass through almost without filtering them. That “remnant,” in Kimura’s view, is what shapes the molecular record of evolution.

For decades, the data seemed to prove him right. When comparing genomes from different species, the patterns of molecular change fit well with a model in which beneficial changes were extraordinarily rare. But a new study led by evolutionary biologist Jianzhi Zhang of the University of Michigan and published in Nature Ecology & Evolutionjust questioned that image.

“We are saying that the result was neutral, but that the process was not,” explains Zhang. “Our model suggests that natural populations are not truly adapted to their environments because environments change so quickly, and populations always lag behind.”

More useful mutations than anyone expected

Zhang’s team analyzed deep mutational scanning data sets (a technique that involves making systematic changes to a gene and measuring how each alteration affects the organism) from their own laboratory and other groups. The organisms used were yeasts of the genus Saccharomyces and the bacteria Escherichia coliboth classics of experimental genetics.

What they found was a piece of information that flies in the face of decades of theory: More than 1% of mutations that change an amino acid turned out to be beneficial to the organism in that particular environment. It may sound like little. It is not.

In terms of evolutionary theory, that percentage implies that more than 99% of amino acid substitutions should be adaptive. And if that were true, molecular evolution would have to occur at a rate much higher than what is observed in nature. Therein lies the contradiction. Something didn’t add up.

Evolution pursues a target that never stops moving

The solution proposed by Zhang’s team has a technical name that can be summarized: Adaptive Tracking with Antagonistic Pleiotropythat is, adaptive tracking with antagonistic pleiotropy. In practical terms, it means that a mutation can be advantageous today and become a burden tomorrow, depending on what the environment does in the meantime.

If the environment changes before that useful mutation spreads throughout the population, the advantage evaporates. The mutation does not become “fixed”, that is, to become universal in the species. And so, paradoxically, an evolution full of beneficial mutations can leave a molecular record that appears neutral: traces of the process disappear before they can be seen.

To test it under controlled conditions, the researchers divided yeast populations into two groups. One evolved over 800 generations in a stable environment. The other rotated through ten different growing media, spending 80 generations in each, until also completing 800. In the changing environment group, beneficial mutations appeared, but they did not have time to spread before the terrain changed again. The result: an evolutionary pattern that, seen from the outside, is reminiscent of that predicted by the Neutral Theory.

Beneficial mutations existed. It’s just that the environment didn’t give them time.

Be careful not to read this as Kimura’s funeral. The new model does not say that the Neutral Theory was wrong, but rather it reconciles something that until now seemed impossible to reconcile: genomes look like what neutrality predicts, but the machinery that produces them is continually in adaptive motion.

And us?

Zhang is not limited to single-celled organisms. He points out that human beings are a particularly interesting case of this dynamic, because our environment has changed drastically and rapidly in the last thousands of years. Some genes that could be advantageous in the environments where our ancestors lived could today be in a state of mismatch with current conditions.

It is not an entirely new idea: studies on cardiovascular disease, obesity or type 2 diabetes have been pointing out for years that certain physiological traits that in another context would have been adaptive are today problematic in environments of abundance and sedentary lifestyle. But Zhang offers a concrete molecular mechanism that would explain why this mismatch is almost inevitable.

“Any time you look at a natural population, depending on when the last major environmental change was, that population may be very poorly adapted or relatively well adapted,” Zhang says. “But we will probably never see any population fully adapted to its environment, because complete adaptation would require more time than almost any natural environment can remain constant.”

What remains to be checked

Talking about biology is complicated, and it is important to be precise about the scope of what this study demonstrates, and what it does not yet demonstrate. Most of the available deep mutational scanning data comes from single-celled organisms: yeast and bacteria. There is no reason to automatically assume that the same rates of beneficial mutations occur in complex organisms such as animals, plants, or humans, whose genetic biology is considerably more tangled.

Zhang himself recognizes this. The lab’s next step will be to look for mutational scanning data in multicellular organisms to see if the pattern repeats itself. And it leaves open a question that the model does not fully resolve: why, even when the environment remains relatively stable for a long time, it still takes so long for populations to reach something resembling complete adaptation.

What Zhang and his team have proposed is the right question. The answer will surely have more layers.

What the study does solidly establish is that Neutral Theory and adaptive selection are not irreconcilable rivals. They can be, simultaneously, true: the evolutionary process is charged with useful movement, and the record it leaves seems immobile. Evolution is not going towards perfection. He is after a target that never stops.