Pfizer launches new collaboration with XtalPi for AI drug modeling

Pfizer plans to expand its work with the Chinese tech startup XtalPi to develop an artificial intelligence-powered platform to model small-molecule drugs as part of its discovery and development efforts.

The project will combine quantum mechanics and machine learning to help predict the pharmaceutical properties of a broad range of molecular compounds.

Pfizer and XtalPi are currently collaborating in crystal structure prediction and screening—using computer models to determine the potential molecular stability of an organic compound—and are looking to advance their work in drug design and solid-form selection.

As part of the project’s development, a portion of the molecular mechanics parameters generated using public-domain compounds will be made available to the public, XtalPi said in a statement.

RELATED: China AI biotech XtalPi lands $15M from Sequoia, Google, Tencent

“The collaboration allows us to apply our expertise in molecular modeling, AI, and cloud computing towards improving existing computational methods while exploring new algorithms to address a wide range of drug design challenges,” said Shuhao Wen, XtalPi's co-founder and chairman of the board.

In January, XtalPi announced it raised $15 million in series B funding from Google, Sequoia China, and Tencent, the Chinese internet conglomerate that led the startup’s series A round. Founded in 2014 by a group of quantum physicists from the Massachusetts Institute of Technology, the company now has locations in Cambridge, Mass., Beijing and Shenzhen, China.

XtalPi’s algorithms are supported by cloud computing, employing an elastic cluster of servers across AWS, Tencent Cloud, Google Cloud and Alibaba Cloud that can deploy up to 1 million cores in seconds, the company told FierceBiotech at the time.

As the biotech and research industry turns to AI to build out their drug development capabilities, so too have tech companies made moves into pharmaceutical research. Earlier this year, BenevolentAI purchased Proximagen’s drug discovery and development operations for an undisclosed sum, aiming to be the first to implement AI across the entire R&D process.

Meanwhile, GlaxoSmithKline tapped Insilico Medicine to use AI, genomics and big data to help identify novel biological targets and pathways last August. Insilico’s own drug discovery programs have focused on cancer, ALS and diabetes, as well as in Parkinson’s and Alzheimer’s disease.

And this week, Worldwide Clinical Trials and Datavant partnered to build a catch-all data set that explores the reasons clinical trials in neurodegenerative diseases fail, including use of electronic health records, insurance claims and diagnostics through a pilot study.