A few years ago, Gevorg Grigoryan PhD ’07, then a professor at Dartmouth College, had been pondering an idea for data-driven protein design for therapeutic applications. Unsure how to move forward with launching that concept into a company, he dug up an old syllabus from an entrepreneurship course he took during his PhD at MIT and decided to email the instructor for the class.
He labored over the email for hours. It went from a few sentences to three pages, then back to a few sentences. Grigoryan finally hit send in the wee hours of the morning.
Just 15 minutes later, he received a response from Noubar Afeyan PhD ’87, the CEO and co-founder of venture capital company Flagship Pioneering (and the commencement speaker for the 2024 OneMIT Ceremony).
That ultimately led Grigoryan, Afeyan, and others to co-found Generate:Biomedicines, where Grigoryan now serves as chief technology officer.
“Success is defined by who is evaluating you,” Grigoryan says. “There is no right path — the best path for you is the one that works for you.”
Generalizing principles and improving lives
Generate:Biomedicines is the culmination of decades of advancements in machine learning, biological engineering, and medicine. Until recently, de novo design of a protein was extremely labor intensive, requiring months or years of computational methods and experiments.
“Now, we can just push a button and have a generative model spit out a new protein with close to perfect probability it will actually work. It will fold. It will have the structure you’re intending,” Grigoryan says. “I think we’ve unearthed these generalizable principles for how to approach understanding complex systems, and I think it’s going to keep working.”
Drug development was an obvious application for his work early on. Grigoryan says part of the reason he left academia — at least for now — are the resources available for this cutting-edge work.
“Our space has a rather exciting and noble reason for existing,” he says. “We’re looking to improve human lives.”
Mixing disciplines
Mixed-discipline STEM majors are increasingly common, but when Grigoryan was an undergraduate, little-to-no infrastructure existed for such an education.
“There was this emerging intersection between physics, biology, and computational sciences,” Grigoryan recalls. “It wasn’t like there was this robust discipline at the intersection of those things — but I felt like there could be, and maybe I could be part of creating one.”
He majored in biochemistry and computer science, much to the confusion of his advisors for each major. This was so unprecedented that there wasn’t even guidance for which group he should walk with at graduation.
Heading to Cambridge
Grigoryan admits his decision to attend MIT in the Department of Biology wasn’t systematic.
“I was like, ‘MIT sounds great — strong faculty, good techie school, good city. I’m sure I’ll figure something out,’” he says. “I can’t emphasize enough how important and formative those years at MIT were to who I ultimately became as a scientist.”
He worked with Amy Keating, then a junior faculty member, now head of the Department of Biology, modeling protein-protein interactions. The work involved physics, math, chemistry, and biology. The computational and systems biology PhD program was still a few years away, but the developing field was being recognized as important.
Keating remains an advisor and confidant to this day. Grigoryan also commends her for her commitment to mentoring while balancing the demands of a faculty position — acquiring funding, running a research lab, and teaching.
“It’s hard to make time to truly advise and help your students grow, but Amy is someone who took it very seriously and was very intentional about it,” Grigoryan says. “We spent a lot of time discussing ideas and doing science. The kind of impact that one can have through mentorship is hard to overestimate.”
Grigoryan next pursued a postdoc at the University of Pennsylvania with William “Bill” DeGrado, continuing to focus on protein design while gaining more experience in experimental approaches and exposure to thinking about proteins differently.
Just by examining them, DeGrado had an intuitive understanding of molecules — anticipating their functionality or what mutations would disrupt that functionality. His predictive skill surpassed the abilities of computer modeling at the time.
Grigoryan began to wonder: Could computational models use prior observations to be at least as predictive as someone who spent a lot of time considering and observing the structure and function of those molecules?
Grigoryan next went to Dartmouth for a faculty position in computer science with cross-appointments in biology and chemistry to explore that question.
Balancing industry and academia
Much of science is about trial and error, but early on, Grigoryan showed that accurate predictions of proteins and how they would bind, bond, and behave didn’t require starting from first principles. Models became more accurate by solving more structures and taking more binding measurements.
Grigoryan credits the leaders at Flagship Pioneering for their initial confidence in the possible applications for this concept — more bullish, at the time, than Grigoryan himself.
He spent four years splitting his time between Dartmouth and Cambridge and ultimately decided to leave academia altogether.
“It was inevitable because I was just so in love with what we had built at Generate,” he says. “It was so exciting for me to see this idea come to fruition.”
Pause or grow
Grigoryan says the most important thing for a company is to scale at the right time, to balance “hitting the iron while it’s hot” while considering the readiness of the company, the technology, and the market.
But even successful growth creates its own challenges.
When there are fewer than two dozen people, aligning strategies across a company is straightforward: Everyone can be in the room. However, growth — say, expanding to 200 employees — requires more deliberate communication and balancing agility while maintaining the company’s culture and identity.
“Growing is tough,” he says. “And it takes a lot of intentional effort, time, and energy to ensure a transparent culture that allows the team to thrive.”
Grigoryan’s time in academia was invaluable for learning that “everything is about people” — but academia and industry require different mindsets.
“Being a PI [principal investigator] is about creating a lane for each of your trainees, where they’re essentially somewhat independent scientists,” he says. “In a company, by construction, you are bound by a set of common goals, and you have to value your work by the amount of synergy that it has with others, as opposed to what you can do only by yourself.”
