EvolutionaryScale, a startup founded by former Meta researchers, has secured $142 million in seed funding to develop AI models capable of generating novel proteins.
The funding round was led by ex-GitHub CEO Nat Friedman, Daniel Gross, and Lux Capital, with significant contributions from Amazon and NVentures, Nvidia’s corporate venture arm.
The company’s flagship AI model, ESM3, aims to transform biology by simulating 500 million years of evolutionary processes.
“ESM3 takes a step toward a future of biology where AI is a tool to engineer from first principles, the way we engineer structures, machines, and microchips and write computer programs,” said Alexander Rives, co-founder and chief scientist at EvolutionaryScale.
ESM3 can generate new proteins by reasoning over their sequence, structure, and function, significantly reducing the time and cost associated with protein design in the lab.
We have trained ESM3 and we’re excited to introduce EvolutionaryScale.
ESM3 is a generative language model for programming biology. In experiments, we found ESM3 can simulate 500M years of evolution to generate new fluorescent proteins.
Read more: https://t.co/iAC3lkj0iV pic.twitter.com/AhWtC4vxlF
— Alex Rives (@alexrives) June 25, 2024
Trained on a dataset of 2.78 billion proteins, the model has already demonstrated its capabilities by generating a new variant of green fluorescent protein (GFP), which typically takes hundreds of millions of years to evolve naturally.
The startup plans to make ESM3 available for non-commercial use through its cloud Forge developer platform and will offer a smaller version of the model for offline use.
Extraordinary! Can it generate interaction partners of a given protein? Could you design with this a new “general” interaction partner framework (like a new class of small programmable binding proteins that would replace (too expensive) antibodies and which would be easy to…
— Garry P. Nolan (@GarryPNolan) June 25, 2024
EvolutionaryScale will also partner with AWS and Nvidia to provide access to ESM3, enabling pharmaceutical and biotech companies to integrate the model into their workflows.
These partnerships will allow customers to fine-tune the model using their own data, enhancing its applicability in drug discovery and materials science.
You guys should ask dang at HN to put this in the “second chance pool” because hardly anyone saw it there and I think they’d find it very interesting.
— Jeffrey Emanuel (@doodlestein) June 25, 2024
Despite the ambitious goals, EvolutionaryScale acknowledges the challenges ahead. The company estimates it could take a decade for generative AI models to integrate fully into therapeutic design processes.
Furthermore, EvolutionaryScale faces competition from established players like DeepMind’s spin-off, Isomorphic Labs, and other biotech firms such as Insitro, Recursion, and Inceptive.
You guys should ask dang at HN to put this in the “second chance pool” because hardly anyone saw it there and I think they’d find it very interesting.
— Jeffrey Emanuel (@doodlestein) June 25, 2024
In their pitch deck, EvolutionaryScale emphasized the importance of scaling their model training to incorporate diverse biological data, aiming to create a general-purpose AI model for various biotech applications.
The company’s approach underscores the growing trend of using increasingly large models, datasets, and computational power to drive AI advancements in biology.
Just thinking about how the AI behind this generative language model can help improve lives gives goosebumps by disrupting healthcare
— Prashant | AI (@Prashant_1722) June 25, 2024
EvolutionaryScale’s efforts represent a significant step toward making biology programmable, with the potential to revolutionize drug discovery and synthetic biology.
The substantial backing from prominent investors and strategic partnerships with industry leaders like Amazon and Nvidia provide a strong foundation for the startup to achieve its ambitious objectives.
For more news and trends, visit AI News on our website.
