AI Unravels Tough Protein Structures

New Berkeley Lab program offers faster, more accurate way to determine protein structure

Published on Mar. 11, 2026

Got story updates? Submit your updates here. ›

Researchers at the Department of Energy's Lawrence Berkeley National Laboratory have developed a new computer program called AQuaRef that uses quantum-mechanical calculations and artificial intelligence to produce higher quality structural information on proteins at a lower computational cost. This breakthrough allows for a more precise understanding of protein function in both healthy and diseased states, with potential applications in drug design, bioenergy production, and more.

Why it matters

Understanding the molecular structure of proteins is crucial for revealing insights into how they function, which can lead to more effective therapeutics and bioenergy production. However, current methods for determining protein structure have limitations. The new AQuaRef program represents a significant advancement, using quantum-level refinement to provide a faster and more accurate way to map protein structures.

The details

AQuaRef, developed by researchers at Berkeley Lab and an international team, uses quantum-mechanical calculations and artificial intelligence to predict the placement of atoms and electrons in a protein's molecular structure. This allows for a more precise understanding compared to existing methods that rely on combining experimental data and a library of known protein structures. In tests, AQuaRef produced higher quality structural information at a lower computational cost while maintaining an equal or better fit to experimental data. The program was also able to correctly determine the structure of a protein linked to Parkinson's disease, which has been notoriously difficult to map.

  • The AQuaRef program was developed over the past five years through a collaborative effort between Berkeley Lab and researchers at Carnegie Mellon University.
  • The findings from the team's work were recently published in the journal Nature Communications.

The players

Nigel Moriarty

A Berkeley Lab researcher and contributor to the recent AQuaRef publication.

Paul Adams

A member of the Phenix team at Berkeley Lab involved in the AQuaRef research.

Billy Poon

A member of the Phenix team at Berkeley Lab involved in the AQuaRef research.

Pavel Afonine

The lead researcher on the AQuaRef project.

Phenix

A comprehensive software suite that generates realistic computer models used by structural biologists around the world to solve macromolecular structures.

Got photos? Submit your photos here. ›

What they’re saying

“We're all basically a bunch of proteins. They do so much in our bodies that detail the processes of life. Understanding their structure can give us insights into the mechanisms that cause disease in humans or produce energy in plants. All of this knowledge can lead to more effective therapeutics and bioenergy production.”

— Nigel Moriarty, Berkeley Lab researcher (Mirage News)

What’s next

The researchers are now aiming to broaden the scope of AQuaRef to include more diverse protein structures, such as those required for pharmaceutical drug design. They also see potential applications in better understanding the mechanisms of photosynthesis for enhanced crop productivity and mapping proteins in plants related to biofuel production.

The takeaway

The AQuaRef program represents a significant advancement in the field of protein structure determination, using quantum-level refinement and AI to provide a faster and more accurate way to map protein structures. This breakthrough could lead to important insights into the mechanisms behind disease and energy production, ultimately enabling more effective therapeutics and bioenergy solutions.