If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Drugmakers embrace AI through deals with specialist firms

Companies seek to accelerate discovery and improve drug candidate quality with high-speed computational models

by Rick Mullin
January 13, 2022 | A version of this story appeared in Volume 100, Issue 2

A worker at a lab computer
Credit: Exscientia
A drug discovery researcher at work at Exscientia

Signaling a pattern for the year ahead, five major drug companies are advancing their use of artificial intelligence in drug discovery through research agreements with AI specialist firms.

Exscientia, an AI developer with a drug pipeline of its own, has signed an agreement with Sanofi in which the French drugmaker will access Exscientia’s AI platform in joint research on 15 small-molecule drug candidates for oncology and immunology indications.

Generate Biomedicines, a 3-year-old specialist in AI for biological engineering, announced a similar deal with Amgen in which the partners will collaborate on protein-based drugs for five targets in several therapeutic areas.

The AI specialist Insilico Medicine has agreed to work with China’s Fosun Pharma on AI-driven drug discovery R&D on four biological targets. The partners will also codevelop Insilico’s QPCTL discovery program, which targets a protein in the CD47 pathway with relevance in immuno-oncology.

Meanwhile, PostEra, a medicinal chemistry AI specialist formed at the University of Cambridge in 2019, announced a deal in which Pfizer will employ PostEra’s technology in oncology and antiviral drug research programs. And Absci announced a partnership in which it will use its AI-based biologics platform to make enzymes tailored for biomanufacturing at Merck & Co.

In the largest of the deals, Exscientia will receive $100 million in up-front payments from Sanofi, and Generate will get $50 million from Amgen. Both deals also entail potential milestone and royalty payments.

Exscientia CEO Andrew Hopkins says his firm’s new contract—its largest so far—marks a significant increase in its engagement with Sanofi. The two have partnered on the design of bispecific small molecules since 2016, and in 2019 Sanofi licensed an Exscientia drug candidate that addresses inflammation and immunology targets.

The partnership will now include drug target discovery as well as drug design using Exscientia’s AI-based personalized medicine platform, which incorporates human tissue samples into early target and drug discovery.

Under its contract with Amgen, Generate will take a lead role in early discovery, generating novel protein sequences computationally and building and testing them in its laboratory. Amgen will take the lead in preclinical and clinical development and manufacturing.

Amgen, which in recent years has supplemented biologics discovery with emerging sequence-based drug design technologies, sees AI as a way to accelerate the discovery process through the rapid generation of large volumes of data, says Raymond J. Deshaies, senior vice president of global research.

“I’m looking at this as a part of a continuum” in which Generate’s AI algorithms will optimize sequencing and high-throughput discovery engines already in place, Deshaies says. “That allows us to create huge data sets, and it is those huge data sets that are instrumental to feeding the AI engine of the future,” he adds.

Industry watchers say major drug companies are likely to make significant progress in incorporating AI into discovery research in the months ahead, relying more on their partnerships with AI platform development firms.

Sam Heaps, a specialist in machine learning and bioinformatics at BioTeam, a health-care informatics consultancy, says the rash of deals reflects drug companies’ increased confidence in the usefulness of AI and bodes well for developing discovery teams composed of computer scientists and traditional drug researchers.

“You need both sides to make AI go, especially in challenging fields like drug discovery and protein design,” Heaps says. “So this is a fantastic first step.”


This story was updated on Jan. 18, 2022, with news about agreements signed by Fosun Pharma, Pfizer, and Merck & Co. plus comment from an analyst. Quotations about the relationship between artificial intelligence and drug discovery from Sanofi and Amgen press releases were removed because they were no longer relevant.


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.