Former OpenAI science lead founds Warp Sciences, raising $215 million to build robotic drug lab
Former OpenAI science lead Wei Er and three co-founders founded Warp Sciences, raising $215 million and planning to establish a chemistry contract research organization in the United States to provide molecule synthesis and testing services for drug discovery teams. In its first phase, the company will focus on synthesizing analogs for lead compound optimization; clients decide which molecules to study and retain control of their drug programs, and Warp will not develop its own drug pipeline. To train the robots, chemistry researchers will wear camera equipment and sensor gloves to demonstrate and record experimental operations, and the data from each round of experiments will then be used to improve the AI models and subsequent experimental protocols.
Why read this
AI score 57
This funding news lays out the specific division of labor and data feedback loop for the robotic lab, offering a window into how AI and automation are moving into the outsourced drug discovery segment.
Topics
Body · AI translation
Warp Sciences plans to establish a chemistry contract research organization in the United States to provide molecular synthesis and testing services for drug discovery teams. In its first phase, the company will focus on "lead compound optimization analog synthesis," that is, creating different variants around promising molecules and then comparing their properties through experiments to help researchers screen for drug candidates more suitable for continued development. The company says clients decide which molecules to study and retain control over their drug programs; Warp does not intend to develop its own drug pipeline.
To train robots to perform laboratory work, chemistry researchers will wear camera equipment and sensor gloves to demonstrate and record experimental operations. The company hopes to gradually use robots to carry out these tasks so that the laboratory can operate continuously; the data generated by each round of experiments will then be used to improve AI models and subsequent experimental plans, forming a "design, execute, analyze, redesign" loop. The degree of automation and actual efficiency still await the company's subsequent construction and verification.
The startup's business also targets the U.S. biotech industry's reliance on overseas R&D services. In drug discovery, the batch synthesis of thousands of molecular variants is a time-consuming and labor-intensive step, and such work has long been undertaken by contract research organizations in China and elsewhere. Warp hopes to provide an alternative service in the United States, reducing geopolitical supply chain risks and addressing pharmaceutical companies' concerns about intellectual property protection.
Warp Sciences was co-founded by Weir and three co-founders: Saul Reisberg, former head of medicinal chemistry at Terremoto Biosciences; Manohar Paluri, co-founder and former executive of Meta's AI research lab FAIR; and DJ Patil, former U.S. chief data scientist. The company says the team will combine AI, robotics and medicinal chemistry experience to build domestic laboratory infrastructure, allowing physical experimental data in turn to drive AI-assisted drug discovery.