AI is dreaming up drugs that no one has ever seen before. Now we have to see if they work.

Today, on average, it takes more than 10 years and billions of dollars to develop a new drug. The vision is to use AI to make drug discovery faster and cheaper. By predicting how dangerous drugs will behave in the body and discarding dead compounds before they leave the computer, machine learning models reduce the need for tedious laboratory work.

There is always a need for new drugs, says Adityo Prakash, CEO of a California-based pharmaceutical company. He said.

New labs are now being built around the world. Last year Exscientia opened a new research center in Vienna; In February, Hong Kong-based drug discovery firm Ensilico Medicine opened a major new lab in Abu Dhabi. All told, nearly two dozen drugs (and counting) developed with the help of AI are now entering clinical trials.

“If someone could tell you they could predict exactly which drug molecule would pass through the gut… they might even have land to sell you on Mars.

Adityo Prakash, CEO of Verseon

“We’re seeing this in action and investment because increasing automation in the pharmaceutical industry is starting to produce enough chemical and biological data to train good machine-learning models,” said Sean McLain, founder and CEO of Vancouver-based ABC. Washington, which uses AI to search for billions of potential drug designs. “Now is the time,” McClain says. We are going to see a big change in this industry in the next five years.

But it’s still early days for AI medicine. “If someone could tell you that they could accurately predict which drug molecule would enter the gut or not be broken down by the liver,” Prakash said, there are many AI companies. You may also have land to sell on Mars.

And the technology is not a cure: experiments on cells and tissues in the laboratory and experiments on humans – the slowest and most expensive development processes – cannot be cut completely. “We are saving a lot of time. Louisa Salter-Side, chief scientific officer of Pioneering Medicines, a startup incubator in Cambridge, Massachusetts, is already doing many of the steps we’ve been doing in our hands. But the final confirmation must be done in the laboratory. Still, AI is changing how medicine is made. It may be a few years before the first drugs designed with the help of AI reach the market, but the technology is about to shake up the pharmaceutical industry, from the initial stages of drug design to the final approval process.

The basic steps involved in developing a new drug from scratch have not changed much. First, select a target in the body that the drug will interact with, such as a protein; Then design a molecule that does something to the target, such as changing how it works or shutting it down. Next, put that molecule in the lab and make sure it does exactly what it was designed to do (and it doesn’t). And finally test it in humans to see if it’s safe and effective.

For decades, chemists have screened candidate drugs by placing samples of a desired target into many small portions in the lab, adding different molecules, and watching for reactions. Then they repeat this process several times, changing the structure of the candidate drug molecules – change this atom for that – and so on. Automation has sped things up, but the underlying process of trial and error is inevitable.