Early drug discovery is big business, so what research trends should we look for next year?
What is it about AI that draws such interest from chemists?
According to David Clark, Research Leader in Computer-aided Drug Design (CADD) at Charles River’s Harlow (UK) site, computers have contributed to drug design for decades. However, it was only recently that researchers have been able to rely on computers for realistic designs.
“The great challenge for the compounds that were invented in those days was that they were all incredibly difficult to synthesize,” said Clark. “It was really hard to interest a medicinal chemist to invest the time in turning your idea from the computer into an actual chemical compound that you could test.”
Back then, computers were a bit like the proverbial monkeys at a typewriter. Every once in a while they might produce a Hamlet, but for the most part their contributions were not useable. In the current Big Data era, where computers can be trained on huge data sets, AI is evolving at a promising rate. By teaching the computers how to build realistic compounds, and telling them which chemicals are known to interact with certain biological targets, the computers can make inferences and suggest possible compounds that have not been tried before.
“We need to [ground] this in reality, and we’re not at the point yet where a computer can design a drug that could just suddenly be synthesized and put immediately into humans,” Clark said. “That is a very, very long way off. But I think we are now at the point where we have these techniques that can produce compound designs that are interesting, that are potentially novel and that have the properties of interest.”
This is certainly the consensus of Charles River scientists Ian Waddell and Chris Hill, who are joining for this month’s podcast to discuss AI and biomarkers, the other hot trend to look for next year in drug discovery circles.
“If we are thinking very differently about the way that we interrogate biomarkers, the biomarkers that we develop in the very early stages of drug discovery could in some cases actually end up being used as patient selection biomarkers,” says Waddell, Executive Director of Biology at Charles River’s Saffron Walden site. “Thus, we could offer our clients the opportunity to develop companion diagnostics combined with biomarkers.”
To learn more about AI in drug discovery and other potential trends in 2020—and 2050—listen to this episode of Sounds of Science. And if you would like to learn more about how AI is influencing drug discovery, check out this Eureka article from our World Congress meeting, and our series on AI in drug discovery, which appeared this summer.