New medications are developed on a regular basis, and many of them are also quite harmful. Law enforcement agencies are trying to find ways to regulate synthetic opioids, bath salts, and other drugs. But we have seen that many chemists, who are mostly unaccredited, produce and distribute drugs that have the ability to trigger psychoactive effects as conventional drugs. This is why researchers have trained computers to forecast what designer medications will emerge on the scene before they enter the market.
New research is already assisting law enforcement agencies throughout the world in reducing the time it takes to discover new versions of harmful psychoactive substances. An AI algorithm may quickly offer plausible predictions for the chemical structures of psychoactive “designer medicines.” The technology could hasten the development of lab tests to screen for use of drugs that have similar effects to cocaine and heroin but are undetectable by present assays.
With extremely addictive substances saturating communities across the United States, this program has the potential to save countless lives. However, it has the potential to open up a whole “dark matter” universe of unexplored psychotropic possibilities. It can take months to identify these so-called “legal highs” in seized pills or powders, during which thousands of people may have already tried a new designer drug.
“Our strategy could reduce the time it takes to find a new designer medication from weeks to months to just hours,” says Michael Skinnider of the University of British Columbia.
Dr. Skinnider and his colleagues trained an artificial intelligence algorithm on the structures of these compounds using a database of known psychoactive substances given by forensic laboratories throughout the world. The deep neural network algorithm they utilized was inspired by the structure and function of the human brain.
The researchers also utilized the AI to construct 1 billion different chemical structures in order to look at medications that could be developed in the future. Following that, the scientists gathered data for 194 novel designer pharmaceuticals and discovered that 176 of them were included in the AI-generated set. The model also reveals those specific molecules that have a higher probability of showing up on the black market.
You can see the research for yourself here.
