The artificial intelligence attack the coronavirus

The artificial intelligence will enable the design of new chemical inhibitors for the viral proteins and the proteins of human hosts in which the virus depends.

Share

July 18, 2020 10: 34

Share

The artificial intelligence attack the coronavirus

Jean-Benoit Legault

The canadian Press

MONTREAL — A research project camped at the intersection between genomics, artificial intelligence, and medicinal chemistry will attempt to expedite the identification of new molecules that could prove useful in the fight against the sars coronavirus.

The project comprises, in particular, Genome Quebec, the Institute for research in immunology and cancer (IRIC) of the Université de Montréal, University of Montreal, Mila – Institut québécois artificial intelligence and McMaster University.

“When we share nothing with the traditional methods, developing a new drug can take ten or fifteen years, said professor Michael Tyers, IRIC/University of Montreal. We hope that this approach will enable us to go somewhat more quickly.”

Each discipline will make a contribution of its own.

The use of the screening, genomics, in a first time, should lead to a better understanding of the genetic interactions between the virus and the cells of human hosts, and thus to the identification of new targets for drug discovery.

The artificial intelligence will enable the design of new chemical inhibitors for the viral proteins and the proteins of human hosts in which the virus depends.

Then, with the medicinal chemistry advanced, the team will be able to synthesize and test these inhibitors.

Artificial Intelligence

“Think of someone who walks in Canada and who seeks a gold mine, said the professor Yoshua Bengio, and Drama/University of Montreal, to illustrate the contribution of artificial intelligence.

“If we go there at random and digging holes to see if it works, it can be a long one. If we take samples in such a region, so field, it had worked in the past, we can explore and find places where it might be worthwhile to dig a hole and at this time to accelerate greatly the search.”

The artificial intelligence is able to learn from examples, he then recalled. Once the computer has identified a molecule that seems promising, he will be able to find others who are like him.

“If you already have something that looks no worse, we can explore variations around this with artificial intelligence that offers us attractive candidates, said the professor Bengio. And if in the computer it looks interesting, we will pass it to our friends, the chemists.”

Because despite all the progress made by computers and technology, the contribution of the human remains vital, highlighted by dr. Anne Marinier, IRIC/University of Montreal.

“The computer will be able to create all sorts of different structures, but we will not know (…) if these molecules are active, she said. It will have absolutely synthesize, and evaluate.”

Several of the predictions that will be made by the computer will come to nothing, recalled the professor Tyers, but you can still take advantage of these predictions to improve the whole process.

Shiny stiletto

The researchers leave in search of the proverbial needle in a haystack: identifying, among the hundreds of millions of molecules of the universe, those who would be able to bind a therapeutic target to combat the sars coronavirus.

Professor Bengio was hopeful that artificial intelligence would be able to simplify a little this quest.

“We can make it sparkle needle in the haystack”, he started laughing.
Mr. Bengio believes that artificial intelligence has the potential to revolutionize and especially to accelerate the development of the drug, which takes on a whole new significance in the context of the pandemic.

It is, however, used most recently in the development of drugs, and the pandemic of sars coronavirus gives him the opportunity to show what she is able to contribute.

“It’s going to be interesting to see what artificial intelligence can do,” said professor Seaman. If it works, it’s going to be essentially an additional tool in our box (…) for those of us who design new medicines. I think this is a great opportunity to develop the artificial intelligence in the field of drug discovery.”

Le Soleil