February 19, 2026
Maureen Searcy
UChicago scientists clock a driving factor in the evolution of error correction
Complex biological systems with multistep processes are prone to errors. As life evolved in complexity, so too did error-correcting strategies.
One type of error-correcting mechanism, called kinetic proofreading because it takes time and energy, can avoid or correct mistakes during biological processes. For example, enzymes can cut out incorrect nucleotides during replication, backtrack during transcription, and disassemble protein complexes to try again.
Fixing mistakes takes time. A widely held assumption is that slower replication or assembly, for instance, is the cost for increased accuracy—that evolution favors quality over speed. However, this assumption does not take into account that mistakes jam the line, that the errors themselves slow down the process.
“For example, the assembly of a complex structure such as the ribosome can be stalled when some step in the assembly process goes awry,” said coauthor Jack Szostak, University Professor in the Department of Chemistry. “In such cases, the fastest way to complete the assembly process may be to go back a few steps and start over.”
In a new study published in Science, a team of UChicago physicists and chemists developed a model comparing the time lost to a jam versus the time spent correcting the problem. They found that fixing the error can take less time overall, suggesting that speed alone could be the factor driving the evolution of error correction.
“We found that proofreading mechanisms can evolve just because biology wants to go fast, without having to care about accuracy,” said co–first author Kabir Husain, a postdoc in Arvind Murugan’s lab at the time of this research and now faculty in physics at University College London. “These results imply that error correction is easier to evolve than previously thought and could have arisen earlier during the origins of life.”
The wrong puzzle piece
The study originated from an unrelated experiment in the lab of Arvind Murugan, associate professor in the Department of Physics and the James Franck Institute, which studies the essential functions of life—learning, self-replication, and evolution—in the simplest possible systems.
Murugan, Husain, and other lab members were running experiments looking at the mutation rate of DNA polymerase, the molecular machine that copies DNA. They found that mutated polymerases that made more mistakes slowed down the process, which was the opposite of what they had intuitively thought—that a “sloppy” job could be done quickly.
This effect, called stalling, happens when an uncorrected error slows down subsequent steps, even if the later steps are correct. Husain compared it to placing the wrong piece in a jigsaw puzzle. “You’re stuck, because now the next piece is really hard to put in, and because of that, it takes you longer to finish the puzzle.”
(Left) Many biological systems fix mistakes by going backward, such as proofreading during DNA replication or reassembling macromolecular complexes. These corrections cost time. (Center) Without correction, mistakes cause long stalls; slow error-correcting systems end up finishing first. (Right) Under selection for faster completion alone (blue), evolution drives biology to a trade-off front (black) dictated by stalling, where gaining speed requires evolving error correction. |
Around that time, Riccardo Ravasio, a postdoc in Murugan’s lab and co–first author on the paper, joined the team and started thinking about these questions theoretically. Ravasio developed a physics-based model, stripped of biological elements, to test whether stalling takes longer than error correction.
The model is an extension of a fundamental classical physics model of kinetic proofreading introduced over 50 years ago. “We turned the model on its head by introducing this new ingredient of stalling after mismatches, to investigate whether there actually is a trade-off between speed and accuracy,” said Ravasio.
Using this model, the team simulated the evolution of a DNA polymerase, which showed that proofreading took less overall time than stalling, indicating that evolutionary selection for speed alone leads to more proofreading and improved accuracy, or fidelity, regardless of how harmful or harmless errors might be.
The study also raises further questions: “Is stalling a property that can itself evolve?” asked Ravasio. “What’s driving the emergence of stalling in the first place?” The team believes there are opportunities to better understand the effect found to be widespread in biological processes.
A growing genome
The highly collaborative study, which leveraged the different backgrounds, knowledge, and skills of scholars from UChicago’s Physics Department and Chemistry Department, University College London, Maynooth University, California Institute of Technology, and Université PSL, contributes to UChicago’s Origins of Life initiative.
The results offer a new advancement toward understanding how life developed on Earth and the potential for life elsewhere—the mission of the Chicago Center for the Origins of Life. The Center unites chemists, physicists, astronomers, and Earth and planetary scientists to investigate questions so immense that no one discipline can answer.
“One of the many puzzles about the origin and early evolution of life is how and why the fidelity of genome replication increased so that larger and larger amounts of information could be transmitted from generation to generation,” said Szostak, director of the Center for the Origins of Life. “The surprising results of our study, which show that selection for speed of replication alone can lead to the evolution of error correcting mechanisms, independent of selection for maintenance of the underlying information, gives a new perspective on the evolution of error correction.”
The project is a good example of what happens when UChicago brings different kinds of scientists together. “My group had been all-theory until we tried some wet-lab experiments and couldn't make sense of our results,” said Murugan. “A chance conversation with Jack during his recruitment to the Origins initiative made us realize that this was something bigger, and we now have several follow-ups that span physics, chemistry and evolution.”
Citation: Riccardo Ravasio et al., Evolution of error correction through a need for speed. Science 391, 818-824 (2026).
Funding: Sloan Foundation, Moore Foundation, National Science Foundation, National Institute of General Medical Sciences, National Institutes of Health, European Research Council, Science Foundation Ireland, and Evans Foundation for Molecular Medicine.