Can AI discover new biology?
It began with an unexpected result in an automated laboratory in Zhangjiang, Shanghai. In traditional drug discovery, laboratory capacity is tightly budgeted, and evaluating unconventional biological ideas is often viewed as an unjustifiable luxury. But when a bundle of in vivo studies we purchased through a contract research organization had more assay capacity than our programs needed, our…
An unexpected discovery in an automated laboratory in Shanghai led to the development of an innovative AI-discovered pain relief candidate. Traditional drug discovery is often viewed as an unjustifiable luxury due to budget constraints, but when a neuroscience team had excess assay capacity in their automated lab, they decided to test their AI target discovery platform, PandaOmics.
To their surprise, the AI surfaced Target Z, a biological protein target that was already being actively developed for a different disease area. The platform linked this protein to pain signaling mechanisms for the first time.
Using unused assay slots, the team evaluated a newly designed molecule engineered to interact with Target Z. The experimental data showed that one molecule outperformed morphine in preclinical pain models, leading to the development of ISM9528, an oral, brain-penetrant, non-opioid candidate. This breakthrough revealed a fundamental truth for business leaders and technologists: the true power of AI in medicine is not just speeding up chemical synthesis, but engineering scientific serendipity at scale.
The biopharmaceutical market is showing strong interest in this platform model, with ISM9528 entering Phase III clinical trials in 2026 as the world's first AI-discovered target and molecule to reach late-stage clinical evaluation. The integration of AI and human expertise in the drug discovery process has transformed the field, moving beyond purely serendipitous discoveries to a systematic, scalable capability.
Written by urgent.news from Fast Company's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
