In the PYP test, the machine learning algorithm is fed data from many closely related proteins that have been imaged in sequence. (The researchers couldn’t reuse the same proteins because they were damaged by X-rays.) Remarkably, these images show how electrons inside proteins move in frames just femtoseconds apart. These clips – which the team then slowed down enough to allow the human eye to follow the change – showed electrons moving from one part of the protein to another. Their movement within the molecule tells how the whole thing is changing its structure. “If my thumb moves, the electrons inside it have to move with it,” Ourmazd offered as an analogy. “When I look at the change in fee allocation [of the thumb], it tells me where my thumb used to be and where it’s gone. “
Previously, the protein’s response to light had never been observed in such a small time period. “There is a lot more information in the dataset than people think,” says Ourmazd.
To better understand the motion of electrons, the Wisconsin team worked with physicists at the Deutsches Elektronen-Synchrotron, who performed theoretical simulations of the protein’s response to light. The electrons and atoms in the protein must move according to the laws of quantum mechanics, they act as a rule. Comparing their results with a simulation based on those rules helped the team understand which movement the protein was making was allowed. This brings them closer to understanding why they see the movements they did.
The combination of quantum theory and AI in the new work holds promise for future studies of light-sensitive molecules, said Fromme. She emphasizes that machine learning methods that can extract a lot of insight from experimental data seem limited, which means that future experiments may include fewer days longer than commuting. one more thing in the lab. Mukamel agrees: “This is a most welcome development that provides a new direction for the analysis of ultrafast diffraction measurements.”
Co-author Robin Santra, a physicist at Deutsches Elektronen-Synchrotron and the University of Hamburg, believes the team’s new approach could change scientists’ minds about incorporating data analysis into their work. surname. “The combination of modern experimental techniques with ideas from theoretical physics and mathematics is a promising route towards further advancements. At times, this can require scientists to leave their comfort zones,” he said.
But some chemists want to see the new approach examined in more detail. Massimo Olivucci, a chemist at Bowling Green State University, shows that PYP’s response to light includes a singularity-like point in its energy spectrum – a point where the mathematical equations for The energy properties of the protein are “broken down”. This pattern of occurrence is as important to a quantum chemist as a black hole to an astrophysicist, because it is another example where the laws of physics, as we understand them today , doesn’t tell us exactly what’s going on.
