UK – DECEMBER 23: Raindrops on glass (Photo by Tim Graham/Getty Images)
It seems ridiculous to think that you could reduce something like amino acid interactions and the resulting proteins to the level of something like Minecraft.
In fact, I don’t think most of us have even thought about how generative AI could be applied to the life sciences.
But a few people behind the scenes are working diligently to find a solution.
Geoff von Maltzahn works at Flagship Pioneering, a company working at the forefront of high-tech life sciences. This venture capital firm has an incubator and teams of scientists and entrepreneurs working on topics such as mRNA and antibody therapies, as well as broader genomics projects. Flagship Pioneering works with companies like Seres Therapeutics and Indigo Agriculture, to innovate where groundbreaking scientific advances can help address the greatest challenges we humans face.
In a compelling TED talk, Maltzahn illustrates how it can all work.
It begins with a basic explanation of protein folding, where individual proteins emerge through a process of evolution that biologists tend to understand quite well, while the rest of us don’t really have a clue. idea.
The world of proteins
Part of what I thought was great about this talk was that Maltzahn really breaks down great scientific ideas into words that many people can easily understand.
For example, in showing how the structure of a protein evolves, he notes that each protein starts as a “noodle” and then folds into useful shapes.
“It’s the equivalent of if your car, your refrigerator, your house came in noodle form and folded into whatever shape you like it in,” he says.
After explaining protein folding, he explains that the new systems are able to generate models of proteins that fold very precisely. Maltzahn discusses the prospect of pursuing a “Shakespearean mastery” of protein writing.
His metaphor about precision and archery, again, boils down this type of innovation to a sort of visual reference for how systems can focus on creating precise antibodies:
“It’s a bit like Robin Hood cutting the first arrow by shooting an arrow right in the center, 50 times in a row,” he says. “And the reason it has the potential to be valuable is this: Every time an antibody shows… that it can provide benefit to patients, humanity benefits from that, the ability to access the best antibody . »
To talk about the best antibodies, he also explains how to react against the Covid virus which created the universal pandemic from which we have all suffered. He explains that to prevent viruses from entering cells, the technology can create antibodies that attack the part that attaches to the receptors.
“This could be important for the next pandemic,” Maltzahn says.
More new science
Additionally, Maltzahn talks about creating new enzymes and analyzing things like the “tree of life of SARS viruses” and how that will lead to some very profound discoveries.
“Perhaps generative AI will not only create great cartoons and limericks to entertain us, each other, and access knowledge that already resides within humanity,” posits he said. “It just might expand our access to new knowledge and new technologies in important ways.” »
How big will it be?
The unexplored space of biology, he explains, is vast. Drawing an analogy from the “great age of exploration” with ships, he asserts that we have worlds of knowledge to discover.
“Even with generous assumptions about everything Mother Nature has had the chance to build and test throughout evolutionary history, her entire laboratory has tested less than the extent of a drop of water compared to all the possible protein oceans on Earth,” he says. “Just think about it for a second, every protein that makes our lives possible, every one of our ancestors. and everything that has ever existed in the living world fits into this one drop of water…if everything we know is in this drop of water, if we find one more drop of water, it will be a really big problem. But there’s really no reason to. believing that it will be limited to this, just imagine what could be.
I’ve seen and heard a lot about AI over the past year, but it’s something quite unique that definitely deserves a close look. We are just beginning to understand how AI can create dynamic, unique and fascinating digital worlds, and how it can think like humans. But being able to create in God’s way, using the building blocks of all life – well, to put it simply, that’s something else.
