B y mid-March, Alpha Lee was growing restless. A pioneer of AI-driven drug discovery, Lee leads a team of researchers at the University of Cambridge, but his lab had been closed amidst the government-initiated lockdowns spreading inexorably across Europe. Having spoken to his collaborators across the globe – many of whom were seeing their own experiments and research projects postponed indefinitely due to the pandemic – he noticed a similar sense of frustration and helplessness in the face of COVID-19. While colleagues talked about finding a novel treat- ment for the virus, Lee was well aware the process was likely to be long and laborious. Traditional methods of drug discovery risked suffering the same fate as the efforts to find a cure for SARS in the early 2000s, which took years and were ultimately abandoned long before a drug ever reached the market. To avoid such an outcome, Lee was convinced that global collaboration was required. Together with a collection of scientists in the U.K. and Canada, he launched the ‘COVID Moonshot’ – a project which encouraged chemists worldwide to share their ideas for potential drug designs. If the Moonshot proves suc- cessful, they hope it could serve as a future benchmark for finding new medicines for chronic diseases. Solving a Complex Jigsaw In February, ShanghaiTech University published the first detailed snapshots of the proteins of the SARS-CoV-2 coronavirus using a technique called X-ray crystallography. In particular, they revealed a high-resolution profile of the virus’s main protease – the part of its structure which enables it to replicate inside a host – and the main drug target. The images were tantalizing. “We could see all the tiny pieces sitting in the structure like pieces of a jigsaw,” said Lee. “All we needed was for someone to come up with the best idea of joining these pieces together with a drug. Then you’d be left with a strong molecule which sits in the protease, and stops it from working, killing the virus in the process.”
A computer simulation of one of the drug candidates that has emerged from the Moonshot project so far, showing how it sits in the COVID-19 virus structure.
Normally, ideas for how best to design such a drug would be kept as carefully guarded secrets within individual labs and companies due to their potential value. But as a result, the steady process of trial and error to reach an optimum design can take years to come to fruition. However, given the scale of the global emergency, Lee felt that the scientific community may be open to collective brainstorming on a mass scale. “Big Pharma usually wouldn’t necessarily do this but time is of the essence here,” he said. “It was a case of, ‘Let’s just rethink every drug discovery stage to see, ‘Ok, how can we go as fast as we can?’” On March 13, he launched the COVID Moonshot, calling for chemists around the globe to come up with the most creative ideas they could think of, on their laptops at home. No design was too weird or wacky to be considered, and crucially nothing would be patented. The entire project would be done on a not- for-profit basis, meaning that any drug which makes it to market will have been created simply for the good of humanity. It worked – within just two weeks more than 2,300 potential drug designs had been submitted. By the end of April, over 4,000 had been received from scientists around the globe.
110
Made with FlippingBook - Online magazine maker