MONROVIA, Calif. — Terray Therapeutics’ lab is a symphony of miniaturized automation. Robots carry tiny tubes of fluids to their stations. Scientists monitor the machines. Proteins in solution combine with chemical molecules contained in wells on special silicon chips.
According to the criteria of
Every interaction is recorded, millions and millions every day, generating 50 terabytes of data daily — the equivalent of more than 12,000 movies.
The lab is a data factory for drug development. It is one of several young companies trying to harness artificial intelligence to produce more effective drugs more quickly. Companies are leveraging the new technology to move the field from artisanal to more automated precision.
“Once you have the right kind of data, AI can work and get really good at it,” said Jacob Berlin, co-founder and CEO of Terray.
AI is a “once-in-a-century opportunity” for the pharmaceutical business, according to consulting firm McKinsey & Company.
AI for drug discovery relies on molecular information, protein structures, and measurements of biochemical interactions. It learns from patterns to suggest potential drug candidates. Because it is based on accurate scientific data, Toxic “hallucinations” are much less likely than with chatbots with more extensive training.
Companies like Terray are building labs to generate the data to help train AI, enabling rapid experimentation and the ability to identify patterns and make predictions about what might work. Generative AI could then digitally design a drug molecule. That design is translated into a physical molecule and tested for interaction with a target protein. The results are fed back into the AI software to improve its next design, speeding up the overall process.
Drug development has traditionally been a costly and time-consuming endeavor. Studies vary on the cost of designing a drug. But the total expense is estimated at $1 billion on average. It takes 10 to 15 years. And nearly 90 percent of drug candidates that make it to human clinical trials fail.
AI developers are looking to improve that rate while reducing time and money.
Companies including Terray, Recursion Pharmaceuticals, Schrödinger and Isomorphic Labs, the drug discovery subsidiary of Google DeepMind, the tech giant’s core AI group, are all pursuing breakthroughs.
In 2021, Google DeepMind released software that accurately predicted the shapes in which amino acid chains would fold into proteins. Those shapes determine how a protein functions. In May, Google DeepMind and Isomorphic announced that their latest AI model, AlphaFold 3, can predict how molecules and proteins will interact.
Berlin has pursued advances in nanotechnology and chemistry throughout his career. Terray has created an AI model to translate chemical data into math, and vice versa. The company has released an open-source version. To expand, it will need funding beyond its $80 million in venture funding, said Eli Berlin, Jacob Berlin’s younger brother and its chief financial and operating officer.
Terray is developing new drugs for inflammatory diseases such as lupus, psoriasis and rheumatoid arthritis. The company, Berlin said, hopes to have drugs in clinical trials by early 2026.
“The ultimate test for us, and for the field in general, is whether in 10 years you look back and can say that The clinical success rate has increased significantly and we have better medicines for human health,” Berlin said.
#Artificial #Intelligence #revolutionizes #drug #development