/w=3840,quality=90,fit=scale-down)
Raphael Thys
In the wake of Evozyne's hiring of a seasoned tech-company executive as CEO, the Chicago-based biotech company announced it's developed an AI model that it hopes will revolutionize development of new therapeutic proteins, significantly accelerating the time it takes to bring drugs treating rare diseases to market.
The Evozyne model, developed in collaboration with Santa Clara, Calif.-based computing company Nvidia, will use data-driven "deep-learning technology" to narrow down possible novel proteins that may successfully fill therapeutic or sustainability roles better than proteins found in nature, said CEO Mike Gamson, who took the helm of the Paragon Biosciences portfolio company Monday. He was previously CEO of legal software tech company Relativity.
By exponentially increasing the number and quality of synthetic protein designs, Evozyne hopes to distinguish itself as the company engineering new therapeutic targets for untreatable diseases and greatly reducing development time for new treatments, he said.
Ultimately, the big splash that Evozyne's novel proteins could make would be in helping build life-changing drug therapies, or world-changing sustainability solutions, Gamson said. And the big payoff could be in the hundreds of millions, maybe even billions, if an Evozyne-designed protein reaches the market, he said.
Currently, the company is working with a number of large pharmaceutical companies, in which it is paid smaller amounts for success in researching proteins.
For example, in April, Evozyne inked a deal with Takeda Pharmaceuticals to research and develop proteins that could be used in gene therapies for up to four rare disease targets.
The collaboration built upon Evozyne's earlier successes on a smaller scale with Takeda and is an example of how the company would work with Big Pharma to bring the new proteins it engineers to market, ultimately, as drug therapies, Gamson said.
Under the terms of the agreement with Takeda, Evozyne will receive upfront and research funding payments totaling tens of millions, the company said in April. It will also be eligible to receive future developmental, regulatory and commercial milestone payments of up to $400 million if all milestones are achieved, plus tiered royalties on net sales of any commercial product resulting from the collaboration, Evozyne said in a statement.
The process, put simply, involves Evozyne identifying a genetic disorder tied to a protein deficiency, and engineering a new protein to treat the disorder.
Evozyne's computer model, the ProT-VAE large language model, is built on computer processing giant Nvidia's BioNeMo framework. It looks at the thousands of possible proteins that can be made to develop a short list, which the company's bioscientists then test, strain out those that won't and run promising proteins through the AI model, which provides a smaller batch to test again, Gamson said.
The company's AI model represents a trend in which a growing number of biotech firms that are proving just as tech as they are bio.
“The field of biology is quickly fusing science and engineering using the latest breakthroughs in generative AI,” Kimberly Powell, vice president of healthcare at Nvidia, said in a statement. “Nvidia BioNeMo is a fundamental part of the ProT-VAE large language model and Evozyne’s platform, which is paving the way for machine learning-guided protein engineering resulting in synthetic functional proteins that can be used in new therapies, energy sources, materials and beyond.”
At its Lincoln Park offices, Evozyne has a wet lab as well as banks of computers. You'll see techs in white lab coats working alongside pods of computer scientists, Gamson said.
Other companies like Chicago-based Tempus Labs, which has built a massive database of gene data that’s used by doctors to determine how patients might respond to different drugs or treatments, are increasingly bringing big data computing to bear on bioscience.
Gamson says that Chicago will be right in the middle of that trend over the next few years, with players like Evozyne's parent Paragon bringing core AI tech together with life sciences.
Beyond life sciences, the new protein engineering model at Evozyne is also being brought to bear on creating enzymes that can provide solutions to sustainability problems, like carbon capture and improved battery performance.
Nurses talking
hospital generic
health insurance form with stethoscope
Future of Work headshots