researchers use machine learning to model cancer-related proteins | Regional News/CA
Lawrence Livermore National Laboratory (LLNL) researchers and a multi-institutional team of scientists have developed a machine learning-based model showing the importance of lipids in the signaling dynamics of RAS, a family of proteins whose mutations are linked to many cancers.
Lipids are organic fatty acid compounds that are insoluble in water but soluble in organic solvents.
In a paper published in the Proceedings of the National Academy of Sciences, the researchers detail the methodology behind the Machine-Learned Multiscale Modeling Framework (MuMMI), which simulates the behavior of RAS proteins on a cell membrane, their interactions with lipids – which help make cell membranes – and the activation of RAS signaling at the macro and molecular level.
According to the researchers, the data indicate that lipids – rather than protein interfaces – govern both the orientation of RAS and the accumulation of RAS proteins.
“We always knew lipids were important,” said LLNL computer scientist and lead author Helgi Ingolfsson. “You need some of them, otherwise you don’t have that behavior. But after that, scientists no longer knew what was important about them.
Normally, RAS proteins receive and follow signals to switch between active and inactive states, but as the proteins move along the cell membrane, they combine with other proteins and can activate signaling behavior.
Mutated RAS proteins can get stuck in an uncontrollable, “always-on” state of growth, which is seen in the formation of approximately 30% of all cancers, particularly pancreatic, lung, and colorectal cancers.
The research “shows us that lipids are a key player,” Ingolfsson said. “By modulating lipids and different lipid environments, the RAS changes its orientation, and you can actually change the signaling (between ‘grow’ and ‘not grow’) by changing the lipids underneath.”
The researchers said the MuMMI framework represents a “fundamentally new technology in computational biology” and could be used to improve their basic understanding of RAS protein binding.
The research is part of a pilot project to co-design advanced IT solutions for cancer, a collaboration between the Department of Energy, the National Cancer Institute and other organizations.
Traditional researchers can only simulate a small, fixed number of proteins and one lipid composition at a time, Ingolfsson explained, and they need to know in advance which lipids are important to model. With the MuMMI framework, researchers can simulate thousands of different cellular compositions derived from the macro-model, allowing them to answer questions about RAS-lipid interactions that were previously only possible with multiscale simulation. .
“We’re demonstrating that the old way of doing things is starting to get outdated,” Ingolfsson said. “At Livermore, we have massive computing power, we have a lot of people working on it, and we can show what’s possible.”