Open science yields broad-spectrum coronavirus antiviral

A new broad-spectrum coronavirus antiviral, ASAP-0017445, has been nominated as a pre-clinical drug candidate by the Drugs for Neglected Diseases initiative (DNDi).

The candidate is the first coronavirus antiviral developed through crowdsourcing and open-science, and the first with its origins in artificial intelligence (AI). The candidate is a result of research from the COVID Moonshot initiative which originated at Diamond Light Source.

It is a main protease inhibitor that shows promising activity against SARS-CoV2 and other viruses of the same family, including other viruses of pandemic potential such as MERSCoV—hence its qualification as broad-spectrum. The compound was designed and developed by the open-science research initiative COVID Moonshot and its sister organization, the AI-driven Structure-enabled Antiviral Platform (ASAP) consortium.

“Our goal is to deliver an effective and affordable antiviral medicine that would be accessible to everyone if and when the next coronavirus pandemic strikes,” said Annette von Delft, Head of Anti-Infectives at the Center of Medicines Discovery, University of Oxford, and partner of the Moonshot initiative.

The COVID Moonshot was a spontaneous global collaboration that started in March 2020, triggered by data from Diamond’s XChem platform for fragment screening. Researchers around the world submitted more than 18,000 molecule designs to inhibit the main protease of SARS-CoV-2.

The detailed structures were made openly available, ensuring that every stage of the discovery process could be built upon by the global research community. This commitment to transparency and open data laid the foundation for the optimization of ASAP-0017445.

Diamond’s contribution highlights the importance of advanced research infrastructure in accelerating drug discovery. By combining crystallography, computational modeling, and global crowdsourced design, the Moonshot demonstrated how open collaboration can deliver a drug candidate intended to be direct-to-generic and globally accessible.

The structure of ASAP-0017445 was publicly disclosed in March 2025. All the data generated during its development, including the structure data of the more than 2,000 compounds submitted during the crowdsourcing phase, are publicly available. Other researchers can build on this unprecedented dataset for drug discovery for their own research.