Outbreaks of diseases like the coronavirus often develop too quickly before scientists can find a cure. But in the future, artificial intelligence can help researchers.
While it is probably too late for new technology to play an important role in the current epidemic, there is hope for the next outbreaks. Artificial intelligence is efficient for searching large volumes of data and finding connections that make it easier to determine what types of treatments could work or which experiments should be followed.
The question is what Big Data will find when it receives little information about a new disease like the one caused by the Covid-19 virus, which appeared at the end of last year in China and infected more than 75,000 people in about two months.
The fact that researchers were able to produce the genetic sequencing of the new virus a few weeks after the first reported cases is promising, as it shows that there is now much more immediate data available when outbreaks occur.
Andrew Hopkins, chief executive of Oxford, England-based startup Exscientia, is among those working to help train artificial intelligence for drug discovery. He believes that new treatments can range from conception to clinical testing between 18 months to 24 months in the next decade, thanks to artificial intelligence.
Exscientia has developed a new compound for the treatment of obsessive-compulsive disorder that is ready to be tested in the laboratory after less than a year in the initial research phase. The time is about five times faster than the average, according to the company.
Cambridge-based Healx takes a similar approach, but uses machine learning to find new uses for existing drugs. Both companies feed their algorithms with information – obtained from sources such as scientific journals, biomedical databases and clinical trials – to help suggest new treatments for diseases.
The two companies have a team of human researchers to work with artificial intelligence, which helps guide the process. In Exscientia's approach, called the Centaur Chemist, drug developers help teach algorithm strategies for finding compounds. Healx makes AI predictions available to researchers, who analyze the results and decide what to look for.
Neil Thompson, scientific director of Healx, said the technique could be applied against an outbreak like that of coronavirus, as long as it has sufficient data on the new disease. Healx does not have a project to fight the coronavirus or adjust its technology for outbreaks, but it would not be impossible.
"We are very close," Thompson said in an interview. "We would not need to change the AI algorithms we use a lot. We try to combine the properties of the drug with the characteristics of the disease."
—With the collaboration of James Paton.