Medicinal drug development is a long and complicated process that often takes years before a new drug is ready for human trials. That’s far too long for some people to wait, so pharmaceutical firms are looking at using artificial intelligence technology to simulate much of the testing process in order to speed things up while saving a significant amount of money along the way.

To that end, GlaxoSmithKline Plc., one of the biggest pharmaceutical companies around, said it’s ready to throw up to $43 million at Scottish AI firm Exscientia Ltd. to use its deep learning smarts to create new drugs.

GSK said it will fund Exscientia’s research into AI-powered drug discovery, which involves identifying molecules that can treat up to 10 diseases in different areas. The idea is to leverage supercomputers and deep learning systems to predict the behavior of those molecules and their suitability to be developed into drugs. If it works, it could save GSK an enormous amount of time and money compared with traditional drug research methods.

The research involves identifying how different molecules interact with disease targets, with the aim of producing drugs based on those molecules that are specific for each target. By using AI, researchers should be able to test and assess different kinds of molecules more quickly and identify those that are suitable candidates for drug development.

Exscientia’s work involves using algorithms to test the affect of molecules on different types of cell, tissue and organisms, generating massive data sets that would take years to build using traditional methods. That data can then be used to accelerate drug discovery, the company said.

The approach could deliver potential drugs in “roughly one-quarter of the time, and at one-quarter of the cost of traditional approaches,” said Andrew Hopkins, Exscientia’s chief executive officer.

The company explained that its deep learning systems are designed to balance the potency and selectiveness of each new drug with its pharmacokinetics – which refers to the speed at which the human body can absorb, process and excrete the drug.

“By applying a rapid design-make-test cycle, the Exscientia AI system actively learns from the preceding experimental results and rapidly evolves compounds towards the desired candidate criteria,” Exscientia said in its press release.

Image: NASA’s Marshall Space Flight Center/Flickr