Scientists say evolution may work differently than we thought

A University of Michigan study challenges that picture. Led by evolutionary biologist Jianzhi Zhang, the research suggests that helpful mutations may be far more common than long standing theory predicts. But there is a catch. Many of those useful mutations may not last long enough to become permanent.

A Major Evolutionary Theory Faces a New Test

During evolution, mutations arise by chance. Some disappear. Others spread until every member of a population carries them, a process known as fixation.

For more than half a century, one of the most influential ideas in molecular evolution has been the Neutral Theory of Molecular Evolution. First proposed in the 1960s, the theory holds that most fixed genetic changes at the level of genes and proteins are neutral. In this view, harmful mutations are usually removed by natural selection, while truly beneficial mutations are so rare that most lasting molecular changes are expected to be neutral.

Zhang and his colleagues set out to examine a key assumption behind that theory. Are beneficial mutations really that scarce?

Their results suggest the answer may be no.

Helpful Mutations May Be Surprisingly Common

Using large deep mutational scanning datasets from their own lab and others, the team looked at the effects of many mutations in model organisms such as yeast and E. coli. In deep mutational scanning, scientists create many mutations in a gene or region of the genome and then measure how those changes affect the organism.

The researchers tracked organisms over many generations and compared them with the wild type, or the version most commonly found in nature. By measuring growth, they could estimate whether a mutation helped, hurt, or had little effect.

They found that more than 1% of the amino acid changing mutations they examined were beneficial. That may sound small, but in evolutionary theory it is enormous. If that many mutations are helpful, the team calculated that more than 99% of amino acid substitutions should be adaptive. Gene evolution should also happen much faster than scientists actually observe in nature.

That mismatch forced the researchers to rethink one of their assumptions. The problem, they concluded, may be that environments do not stay still.

Evolution Chases a Moving Target

A mutation can be useful in one setting and harmful in another. If the environment changes before a beneficial mutation spreads through an entire population, that mutation may lose its advantage or even become a liability.

"We're saying that the outcome was neutral, but the process was not neutral," said Zhang, U-M professor of ecology and evolutionary biology. "Our model suggests that natural populations are not truly adapted to their environments because environments change very quickly, and populations are always chasing the environment."

The team calls this framework Adaptive Tracking with Antagonistic Pleiotropy. In plain terms, it means populations may constantly respond to changing surroundings, while many mutations have tradeoffs that depend on the environment.

A mutation that boosts fitness today may reduce fitness tomorrow. As a result, evolution can be filled with beneficial changes that never become permanent.

Yeast Experiments Show What Happens When Conditions Change

To test this idea, Zhang's team compared two groups of yeast over 800 generations. One group evolved in a stable environment. The other evolved in a shifting environment made up of 10 different growth media.

The changing environment group spent 80 generations in the first medium, then 80 in the next, and so on, until it had also completed 800 generations. (each generation lasted 3 hours)

The researchers found far fewer beneficial mutations in the group exposed to changing conditions. Helpful mutations still appeared, but they often did not have enough time to spread through the population before conditions shifted again.

"This is where the inconsistency comes from. While we observe a lot of beneficial mutations in a given environment, those beneficial mutations do not have a chance to be fixed because as their frequency increases to a certain level, the environment changes," Zhang said. "Those beneficial mutations in the old environment might become deleterious in the new environment."

Why Perfect Adaptation May Be Out of Reach

The findings point to a more restless view of evolution. Rather than steadily marching toward a perfect fit between organisms and their environments, populations may often be stuck in pursuit of conditions that keep changing.

Zhang said the idea has broad implications for living things, including humans.

"I think this has broad implications. For example, humans. Our environment has changed so much, and our genes may not be the best for today's environment because we went through a lot of other different environments. Some mutations may be beneficial in our old environments, but are mismatched to today," Zhang said.

He added that the degree of adaptation seen in any population may depend on how recently its environment changed.

"At any time when you observe a natural population, depending on when the last time the environment had a big change, the population may be very poorly adapted or it may be relatively well adapted. But we're probably never going to see any population that is fully adapted to its environment, because a full adaptation would take longer than almost any natural environment can remain constant."

A Bigger Shift in How Scientists Study Mutation

The Neutral Theory emerged at a turning point in biology. Before the 1960s, scientists often studied evolution by examining an organism's shape, structure, and physical traits. As researchers began sequencing proteins and later genes, they could study evolution at the molecular level.

That shift revealed patterns that the Neutral Theory explained well, including why many genetic differences appear to accumulate steadily over time. The Michigan study does not erase that history. Instead, it offers a way to reconcile two observations that seem to conflict.

On one hand, many molecular changes that become fixed still look neutral when scientists compare genomes. On the other hand, experiments suggest that beneficial mutations may be abundant in a given environment. Zhang's team argues that both can be true if beneficial mutations are often temporary.

Recent research in evolutionary genetics has continued to emphasize the importance of changing environments. A 2026 review of adaptation in rapidly changing conditions highlighted how shifts in allele frequencies and traits depend heavily on available genetic variation. Other yeast research has also shown that adaptation can be shaped by environmental stress and that mutations helpful in one setting may carry costs in others.

Together, these findings reinforce a growing theme in evolutionary biology. The effect of a mutation cannot always be understood in isolation. It may depend on the surroundings, the organism's history, and the speed at which conditions change.

The Caveat and the Next Question

Zhang noted an important limitation. Much of the data used in the study came from yeast and E. coli, single-celled organisms that make it easier to measure the fitness effects of mutations. More deep mutational scanning data from multicellular organisms will be needed to see whether the same patterns apply to animals, plants, and humans.

The team also plans to investigate why organisms take so long to fully adapt even when an environment remains constant.

The study was supported by the U.S. National Institutes of Health and published in Nature Ecology and Evolution. Other authors include former U-M graduate students Siliang Song and Xukang Shen and former U-M postdoctoral researcher Piaopiao Chen.

For now, the work points to a striking possibility. Evolution may be less like climbing steadily toward perfection and more like running after a world that keeps moving.