AI creates a killer drug

Researchers in Canada and the United States have used deep learning to derive an antibiotic capable of attacking a resistant microbe, Acinetobacter baumannii, which can infect wounds and cause pneumonia. According to the BBC, an article in Nature Chemical Biology describes how the researchers used training data that measured the action of known drugs on tough bacteria. The learning algorithm then projected the effect of 6,680 compounds without data on their effectiveness against the germ.

In an hour and a half, the program narrowed the list down to 240 promising candidates. Laboratory tests revealed that nine of them were effective and one, now called abaucin, was extremely potent. Although testing 240 compounds in the lab seems like a lot of work, it's better than testing nearly 6,700.

Interestingly, the new antibiotic seems to only be effective against the target microbe, which is a plus. It's not available to people yet and may not be for some time - drug testing being what it is. However, it remains a great example of how machine learning can augment the human brain, allowing scientists and others to focus on what really matters.

The WHO has identified Acinetobacter baumannii as one of the world's leading superbugs, so a weapon against it would be welcome. You can hope that this technique will significantly reduce the time needed to develop new drugs. It also makes you wonder if there are other areas where AI techniques could quickly weed out alternatives, allowing humans to focus on the most promising candidates.

Want to catch up on machine learning algorithms? Google can help you. Or dive into an even longer course.