US Edition
AI could be our best hope in fighting the next coronavirus
A breakthrough in E.coli research, driven by the use of AI, could point to how to stem the spread of viruses in the future
Artificial intelligence has often been depicted as a malign force against humans. Critics say the widespread use of AI and automation will ultimately trigger the loss of millions of jobs while technology pioneers like Elon Musk have warned of the risk the technology could spiral out of control.
However, the benefits of AI are also potentially enormous.
Supporters cite how the technology could improve road safety, enhance how we communicate, and free up workers to focus on more important and rewarding tasks.
News this week that AI played a central role in the development of a powerful new antibiotic to tackle the world’s most-dangerous bacteria demonstrates another benefit of the technology.
The breakthrough, made by MIT researchers, may be too late to tackle the ongoing coronavirus epidemic, but could it play a vital role in other similar future outbreaks?
MIT specifically trained AI to track down possible antibiotic molecules known for being effective against E.coli growth. The molecule, named halicin, was found by letting machine learning algorithms loose on the extensive libraries of more than 100 million chemical compounds over the course of a few days.
Halicin has proven to be effective against E.coli in initial trials.
The algorithms can dramatically improve the pace at which antibiotics can be discovered for complex bacteria that are resistant to traditional medicines.
Dr Aldo Faisal, director of the behaviour analytics lab at Imperial College said the breakthrough was off the back of “beautiful research”.
“Effectively it follows a pattern of using machine screening to very quickly and cheaply go through thousands of possible compounds and go back over compounds that are already licensed,” he says.
Dr Faisal said it follows similar studies conducted by Imperial where AI was used to search libraries of food compounds to tackle cancerous cells.
The associate professor also says that if AI was used to discover an antibiotic for an outbreak like the coronavirus, it will still need to go through rigorous clinical trials that could take months.
“These are chance discoveries, but it’s a very smart way of looking for gold on the beach,” he says.
“They’ll still need to test and verify that it actually works, it still needs a clinical trial and will still need to be treated with that specific compound.”
Dr Faisal says that AI could be used in other ways to stem an outbreak of a virus.
“Some countries choose draconian measures by effectively putting people on house arrest, but there are smarter ways of preventing the spread of disease,” he says.
One such way, he says, is the use of AI to create “virtual quarantines”. An example of which would be suggesting people from a certain part of London travel on a certain line at a certain time, thus reducing their interaction with the rest of the population more generally.
Elsewhere, Oxford start-up Exscientia is among those using AI to help in drug discovery.
Chief executive Andrew Hopkins says new treatments could go from conception to clinical testing in as little as 18 to 24 months within the next decade, thanks to the use of algorithms.
Exscientia has already designed a new compound for treating obsessive-compulsive disorder that’s ready to be tested in the lab after less than a year in the initial research phase. That’s about five times faster than average, according to the company.
Cambridge-based Healx has a similar approach, but it uses machine learning to find new uses for existing drugs. Both companies feed their algorithms with information - gleaned from sources such as journals, biomedical databases and clinical trials - to help suggest new treatments for diseases.
While MIT's breakthrough is at an early stage. It does demonstrate how important AI could be during the next epidemic to rock the world.
Additional reporting by Bloomberg