Hiding behind the current COVID-19 pandemic, another serious public health threat is looming – the rise of antibiotic-resistant “superbugs.” New antibiotics are needed to help turn the tide, but developing them takes time. Now, IBM Research has put AI to work on the task, producing two promising new drug candidates very quickly.
The discovery of penicillin was one of the most important scientific breakthroughs of the 20th century, as previously deadly infections became easily treatable. But decades on, those benefits are beginning to falter.
Like all organisms, bacteria evolve in response to their environment – so when we pump their environments (ie, our bodies) with drugs, it’s only a matter of time before some of them figure out how to defend themselves. Given enough time and antibiotic use, the only microbes remaining will be those that are genetically immune to the drugs.
That’s the situation we’re increasingly finding ourselves in. We’re now down to our last line of defense – and worryingly, even those are starting to fail. Without new antibiotics or other treatments, scientists predict that once-minor infections could claim up to 10 million lives a year by 2050.
Worse still, developing new drugs takes years and involves a huge amount of trial and error, with potential molecules being made up of countless possible chemical combinations. Thankfully, that’s just the kind of work that artificial intelligence excels at, so IBM has developed a new system to sift through the numbers for us.
The IBM Research team created an AI system that’s much faster at exploring the entire possibility space for molecular configurations. First, the researchers started with a model called a deep generative autoencoder, which essentially examines a range of peptide sequences, captures important information about their function and the molecules that make them up, and looks for similarities to other peptides.
Next, a system called Controlled Latent attribute Space Sampling (CLaSS) is applied. This system uses the data gathered and generates new peptide molecules with specific, desired properties. In this case, that’s antimicrobial effectiveness.
But of course, the ability to kill bacteria isn’t the only requirement for an antibiotic – it also needs to be safe for human use, and ideally work across a range of classes of bacteria. So the AI-generated molecules are then run through deep learning classifiers to weed out ineffective or toxic combinations.
Over the course of 48 days, the AI system identified, synthesized and experimented with 20 new antibiotic peptide candidates. Two of them in particular turned out to be particularly promising – they were highly potent against a range of bacteria from the two main classes (Gram-positive and Gram-negative), by punching holes in the bugs’ outer membranes. In cell cultures and mouse tests, they also had low toxicity, and seemed very unlikely to lead to further drug resistance in E. coli.
The two new antibiotic candidates are exciting enough by themselves, but the process through which they were discovered is the real breakthrough. Being able to develop and test new antibiotics quickly and more efficiently could help prevent the nightmare scenario of returning to a time before antibiotics.
The research was published in the journal Nature.
Source: IBM Research
