To see proteins change in billiardths of a second, use AI


In the PYP experiment, the machine learning algorithm received data from several almost identical proteins that had been mapped one after the other. (Researchers couldn’t reuse the same protein because it was damaged by the X-rays.) The AI ​​extracted the details of the process without the blurring of the X-ray flashes and uncovered what had been obscuring the blur. Notably, these images showed how electrons move within the protein within frames that are only femtoseconds apart. These films – which the team later slowed down enough for the human eye to see the change – show electrons moving from one part of the protein to the other. Their movement within the molecule indicates how the whole is changing its structure. “If my thumb moves, the electrons in it have to move with it,” Ourmazd offers for comparison. “When I look at the change in the load distribution [of the thumb], it tells me where my thumb was before and where it has gone. “

The reaction of the protein to light has never before been observed in such small time increments. “There’s a lot more information in data sets than people generally think,” says Ourmazd.

To better understand the movements of electrons, the Wisconsin team worked with physicists from the German Electron Synchrotron, who performed theoretical simulations of the protein’s reaction to light. The electrons and atoms within the protein have to move according to the laws of quantum mechanics, which act like a set of rules. Comparing their results with a simulation based on these rules helped the team understand which of the allowed movements the protein was making. This brought them closer to understanding why they were seeing the movements they were making.

The combination of quantum theory and AI encapsulated in the new work is very promising for future research on light-sensitive molecules, says Fromme. She stresses that a machine learning approach can extract a lot of detailed information from seemingly limited experimental data, which could mean that future experiments could consist of less long days of doing the same thing over and over in the laboratory. Mukamel agrees, “This is a very welcome development that offers a new way of analyzing ultrafast diffraction measurements.”

Co-author Robin Santra, a physicist at the German Electron Synchrotron and the University of Hamburg, believes the team’s novel approach could change the way scientists think about incorporating data analysis into their work. “The combination of modern experimental techniques with ideas from theoretical physics and mathematics is a promising path to further advances. Sometimes this can require scientists to get out of their comfort zone, ”he says.

However, some chemists would like to see the new approach investigated in more detail. Massimo Olivucci, a chemist at Bowling Green State University, suggests that PYP’s response to light involves something like a singularity in its energy spectrum – a point at which the mathematical equations used to calculate the protein’s energy “break”. Such an event is just as important to a quantum chemist as a black hole is to an astrophysicist, because here, too, the laws of physics as we understand them today cannot tell us exactly what is happening.


Comments are closed.