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.



Creyon debuts to disrupt oligonucleotide research

Startup aspires to drugs-on-demand development techniques with an AI-enabled platform

by Rick Mullin
March 8, 2022

A woman working at a bench in a drug discovery laboratory.
Credit: Creyon Bio
Amanda Reyes, senior research associate at Creyon, where scientists collect data from compounds deemed optimal for building machine learning pharmacology models.

Creyon Bio, an oligonucleotide research specialist with the goal of designing precision medicines, has emerged with $40 million from a series A financing round led by DCVC Bio and Lux Capital.

Creyon is headed by former executives of Ionis Pharmaceuticals, a pioneer in antisense oligonucleotide therapies. It plans to apply machine learning to proprietary datasets to identify design rules and engineering principles for drugs that treat individuals as well as large patient populations.

The company says it works across the full range of oligonucleotide-based medicines, including antisense compounds, small interfering RNA, and DNA- and RNA-editing systems. It hopes to bring an on-demand approach to drug discovery and development based on its artificial intelligence-enabled platform.

Oligonucleotide medicines work by binding to messenger RNA to regulate production of unwanted proteins, but finding them can involve a lengthy trial-and-error screening process.

Creyon began working in late 2019 on an alternative way of finding therapeutic oligonucleotides, according to CEO Christopher E. Hart. The impetus for the company, he says, was a realization that while genomics research can identify the underlying molecular basis of disease, the drug industry has been “embarrassingly slow” in bringing drugs based on this research to market.

“People are recognizing that there is a growing disconnect between our capacity to understand biology and our capacity to use it to make drugs that have an impact on patients,” Hart says. “Everyone I know in the field has acquiesced to this notion that it’s an unsolvable problem and we are going to go ahead and screen our way out of it.”

Creyon, which employs about 20 scientists and engineers, means to sidestep prolonged screening, beginning with the collection of data from in vivo, in vitro, and ex vivo experiments conducted at its headquarters in San Diego and at labs in Research Triangle Park, North Carolina, according to Swagatam Mukhopadhyay, chief scientific officer. Experiments are performed on compounds deemed optimal for building machine learning pharmacology models.

Creyon is currently doing preclinical research, “but we have no ambition to stop there,” Hart says. The company is in talks with potential drug discovery partners and intends to develop its own drug candidate pipeline, he says.

Hart emphasizes that Creyon’s platform is therapy-agnostic. “We are not focused on developing a single drug,” he says. “We are really focused on what is the right set of experiments, the right set of data to create in order to uncover the design rules and engineering principles of oligonucleotide-based medicines.”

Creyon is one of several startups that intend to use AI to fundamentally change drug research. Vesalius Therapeutics, for example, debuted earlier this month with an AI engine that works on genetic data and large clinical databases to segment people with complex diseases into therapeutically meaningful groups.

While no new drugs have been discovered and developed entirely by AI, the technology is widely used in drug research and has proven to expedite drug discovery, drug development, and clinical trials.

Sam Heaps, a specialist in machine learning and bioinformatics at BioTeam, a health-care informatics consultancy, says he likes Creyon’s approach of eliminating brute force screening to find oligonucleotide-based therapies. “From what I have read, they have a deep learning-backed platform that maps sequences and properties of oligonucleotide based-medicine to build a fairly accurate predictive model.”

Moreover, Creyon is using a rule-based design platform that draws from data that have proven successful in prior research. “This tells me they are in scope with the capabilities of their platform and not reaching too far, which seems to be a common problem in this field,” Heaps says.


This article has been sent to the following recipient:

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