An Oxford-based firm that uses artificial intelligence to develop new medicines has teamed up with a UK national science facility to screen more than 15,000 drugs for their effectiveness as a treatment for Covid-19. Exscientia, a spinoff company from the University of Dundee that is now based in Oxford science park, said it had gained access to a large collection of existing drugs held by the Scripps research institute in California and funded by the Bill and Melinda Gates Foundation. It will screen them in partnership with Diamond Light Source near Oxford, which works like a giant microscope and generates bright light that allows scientists to study viruses. Exscientia hopes to discover a drug that can be repurposed to treat coronavirus within the next six to 12 months, whereupon it would be tested on Covid-19 patients, Prof Andrew Hopkins, the firm’s chief executive, told the Guardian. Any potential treatment could be made available for compassionate use before clinical trials are completed, but this would depend on how much can be manufactured quickly. The drug collection, which comprises more than 15,000 compounds that have been approved and tested for human safety in clinical trials or pre-clinical studies, has been shipped from California to Oxford. Prof David Stuart, director of life sciences at Diamond and professor of structural biology at Oxford University, said: “The drugs we are testing have either been approved by the [US regulator] FDA for other diseases or have been extensively tested for human safety. By being able to repurpose existing molecules, we can save a lot of time in the drug discovery process, meaning a faster route to clinical trials, and potentially a treatment for patients.” Exscientia is using its biosensor technology to screen the drug molecules for effectiveness against Sars-CoV-2, the virus responsible for Covid-19. Work will focus on components for viral replication and the interaction between the virus’s spike protein and a human cell receptor that enables the virus’s entry to human cells. The project started in the last few days and the firm hopes to have full data sets within six to eight weeks. Any promising drug molecules will then be tested further to make sure they work as a treatment for Covid-19. Exscientia is collaborating with Diamond and Oxford University, which have been working together since January to develop methods for the production of viral proteins for drug screening and structural analysis at Diamond. This can provide an atomic level of detail in understanding how anti-viral drugs can work. Academics and companies around the world are racing to find a treatment for Covid-19. Efforts have focused on using existing drugs, including the antimalaria medicine chloroquine, the HIV treatment Kaletra, the Japanese anti-flu drug favipiravir and the Ebola drug remdesivir, which has emerged as a frontrunner. The first results from human trials of remdesivir in China and the US are due in April. The UK biotech firm Synairgen is trialling its lung-disease drug, an inhaler, in Covid-19 sufferers. Exscientia, which was founded in 2012, specialises in the use of computer algorithms and machine learning, known as artificial intelligence (AI), to discover novel drugs. The algorithms learn what molecular features make for effective drugs, and the firm will apply this during its search for a Covid-19 medicine. As well as repurposing existing drugs, it plans to use its research to develop a novel treatment for the virus. Exscientia recently claimed a world first when it announced that human tests of the first drug generated entirely by AI – for obsessive-compulsive disorder – would start in March. The project took less than 12 months, instead of the usual four to five years. Using AI to generate new medicines cuts nearly a third from the early-stage drug development cost, and big pharmaceutical firms such as GSK and AstraZeneca are investing heavily in AI. Massachusetts Institute of Technology announced in February that it had used AI to develop a powerful antibiotic that wipes out a range of drug-resistant bacteria. Source