by MONROVIA, Calif. — The lab at Terray Therapeutics is a symphony of miniaturized automation. Robots spin, pouring tiny tubes of fluid to their stations. Scientists in blue coats, sterile gloves and goggles monitor the machines.
But the real action is happening at the nanoscale: Proteins in solution combine with chemical molecules held in minuscule wells in custom silicon chips that resemble microscopic muffin tins. Every interaction is recorded, millions upon millions every day, generating 50 terabytes of raw data per day – the equivalent of more than 12,000 movies.
The lab, about two-thirds the size of a football field, is a data factory for artificial intelligence drug discovery and development in Monrovia, California. It’s part of a wave of young companies and startups looking to harness AI to produce more effective drugs, faster.
Sign up for The Morning newsletter from the New York Times
The companies are taking advantage of the new technology – which learns from huge amounts of data to generate answers – to try to replicate drug discovery. They’re moving the field from painstaking craft to more automated precision, a shift fueled by AI that learns and gets smarter.
“When you have the right kind of data, the AI can work and become really good,” said Jacob Berlin, co-founder and CEO of Terray.
Most of the early business uses of generational AI, which can produce everything from poetry to computer programs, were to help take the drudgery out of mundane office tasks, customer service and code writing. But drug discovery and development is a huge industry that experts say is ripe for an AI makeover.
AI is a “once-in-a-century opportunity” for the pharmaceutical business, according to consulting firm McKinsey & Co.
Just as popular chatbots like ChatGPT are trained on text from around the internet, and image generators like DALL-E learn from dozens of pictures and videos, AI for drug discovery relies on data. And it’s highly specialized data – molecular information, protein structures and measurements of biochemical interactions. The AI learns from patterns in the data to suggest potentially useful drug candidates, as if matching chemical keys with the right protein locks.
Because AI for drug development is driven by precise scientific data, there is far less chance of toxic “bystanders” than with extensively trained chatbots. And any potential drug must be tested extensively in laboratories and clinical trials before it is approved for patients.
Companies like Terray are building large high-tech labs to generate the information to help train the AI, enabling rapid experimentation and the ability to recognize patterns and predict what might work. .
A generative AI can then digitally design a drug molecule. That design is translated, in a high-speed automated laboratory, into a physical molecule and tested for its interaction with a target protein. The results – positive or negative – are recorded and fed back into the AI software to improve the next design, speeding up the entire process.
Although some AI-developed drugs are in clinical trials, it is still early days.
“AI is revolutionizing the field, but the drug development process is messy and very human,” said David Baker, a biochemist and director of the Institute for Protein Design at the University of Washington.
Drug development has traditionally been an expensive, time-consuming, hit-or-miss endeavor. Studies on the cost of designing a drug and leading clinical trials to final approval vary widely. But the total cost is estimated to be $1 billion on average. It takes 10-15 years. And nearly 90% of candidate drugs that enter human clinical trials fail, usually due to lack of efficacy or unexpected side effects.
The young AI drug developers are trying to use their technology to improve those odds, while cutting time and money.
Their most consistent source of funding comes from the pharmaceutical giants, who have long acted as partners and bankers to smaller research ventures. Today’s AI drugmakers are typically focused on accelerating the preclinical stages of development, which typically take 4-7 years. Some people may try to enter clinical trials themselves. But that stage is when big pharmaceutical corporations usually take over, running the expensive human trials, which can take another seven years.
For the established drug companies, the partnership strategy is a relatively low-cost path to innovation.
“For them, it’s like taking an Uber to get you somewhere rather than buying a car,” said Gerardo Ubaghs Carrión, a former biotech investment banker at Bank of America Securities.
The big pharmaceutical companies pay their research partners to achieve milestones towards drug candidates, which can fetch hundreds of millions of dollars over the years. And if a drug is eventually approved and commercially successful, there is a royalty income stream.
Companies such as Terray, Recursion Pharmaceuticals, Schrödinger and Isomorphic Labs are chasing leads. But there are, in general, two different paths – those who are building large laboratories and those who are not.
Isomorphic, the drug discovery spin-off from Google DeepMind, the tech giant’s core AI group, believes that the better the AI, the less data it needs. And he is betting on his software prowess.
In 2021, Google released DeepMind software that accurately predicted the shapes that strings of amino acids would fold into as proteins. Those 3D shapes determine how a protein functions. This stimulated biological understanding and helped drug discovery, because proteins drive the behavior of all living things.
Last month, Google DeepMind and Isomorphic announced that their latest AI model, AlphaFold 3, could predict how molecules and proteins will interact – another step in drug design.
“We’re focusing on the computational approach,” said Max Jaderberg, chief AI officer at Isomorphic. “We think there’s a lot of potential to unlock.”
Terray, like most drug startups, is a byproduct of years of scientific research, along with more recent developments in AI.
Berlin, who earned his doctorate in chemistry from Caltech, has made advances in nanotechnology and chemistry throughout his career. Terray grew out of an academic project started more than a decade ago at the City of Hope cancer center near Los Angeles, where Berlin had a research group.
Terray is focusing on developing small-molecule drugs, basically any drug that a person can ingest in a pill such as aspirin and statins. Pills are convenient to take and cheap to produce.
Terray’s sleek labs are a far cry from the old days of academia when data was stored on Excel spreadsheets and automation was a distant goal.
“I’m the robot,” recalled Kathleen Elison, co-founder and senior scientist at Terray.
But by 2018, when Terray was founded, the technologies needed to build its industrial-style data lab were advancing rapidly. Terray relies on advances from outside manufacturers to make the microscale chips that Terray designs. Their labs are filled with automated equipment, but almost all of the equipment is custom-made—enabled by advances in 3D printing technology.
From the beginning, Terray’s team recognized the importance of AI being able to make sense of its data store, but the potential for generative AI in drug development only became apparent later – but before ChatGPT emerged in 2022.
Narbe Mardirossian, a senior scientist at Amgen, became Terray’s chief technology officer in 2020—in part because of his wealth of data generated in the lab. Under Mardirossian, Terray has built its data science and AI teams and created an AI model to translate chemical data into mathematics, and back again. The company has released an open source version.
Terray has partnership deals with Bristol Myers Squibb and Calico Life Sciences, a subsidiary of Google’s parent company Alphabet, which focuses on age-related diseases. The terms of those deals are not disclosed.
To expand, Terray will need more than $80 million in venture funding, said Eli Berlin, Jacob Berlin’s younger brother. He left a job in private equity to become the startup’s co-founder and chief financial and operating officer, arguing that the technology could open the door to a lucrative business, he said.
Terray is developing new drugs for inflammatory diseases including lupus, psoriasis and rheumatoid arthritis. The company, said Jacob Berlin, expects to have drugs in clinical trials before 2026 at the earliest.
The drug-making innovations of Terray and her peers can speed things up, but only so much.
“The ultimate test for us, and the field in general, is if you look back in 10 years and you can say that the clinical success rate has gone up a lot and we have better drugs for human health,” said Berlin.
c.2024 The New York Times Company
