Halicin is the fashion antibiotic and was discovered thanks to an intelligence neural network that has just identified it and has defined it as a very promising substance to attack bacteria that are difficult to kill.
But that’s not the best of all. The best thing is that with the discovery of halicin, the use of deep learning approaches applied to the world of antibiotics begins to show its true potential. A potential that comes at the best of times: in the midst of a profound drug crisis that threatens the near future of modern medicine.
Halicin faces superbacteria
The last time a human being said “we have discovered a new class of antibiotics” was in the 1980s and many of those reading this article were not even born by then.
In essence, all the antibiotics that have been released over these three decades are variations of drugs that had been found before. The reason? finding a new drug is equivalent to trying to find a needle in a haystack that is tens of acres wide.
The Wellcome Trust estimates that the process needed to find a new antibiotic that holds promise in the fight against superbacteria, is something that would take no less than 15 years and cost no less than a billion euros. Money that is invested without knowing whether everything will end in a successful process or not.
The truth of the matter is that our ability to find new molecules to help defend ourselves against bacteria has slowed down almost to the same speed as concerns about antibiotic resistance. Fortunately, we have (what we think is) a secret weapon: artificial intelligence.
Deep learning vs. multi-resistant bacteria
While artificial intelligence has been emerging for years as a solution to humanity’s knowledge bottleneck, the truth is that until now, there has been no opportunity to demonstrate it.
Using compounds known to suppress E. coli growth, Collins’ team trained a machine learning neuronal network to identify potential antibiotics capable of targeting this bacterium. Once ready, the researchers used it to examine thousands of molecules recorded in numerous existing chemical libraries and tried to predict their effectiveness. The researchers found that nearly 50% of the compounds identified by the network were effective in vitro in killing E. coli.
A Very Hopeful Start
This, despite the fact that we are still talking about a 50% effectiveness rate, is excellent news because, although these enormous libraries of molecules have been available for years, the teams of researchers do not have an efficient way of selecting those that are most likely to have antibiotic properties. If, as Collins’s team at Cell suggests, neural networks can be used to identify good candidates that could save us a lot of time and resources, they could well be used for other research activities as well.
But we should not fool ourselves. Research on these neural networks is in its infancy and there is still a lot of work to be done, but it is comforting to see how these kinds of approaches leave the world of possibilities and materialize in this reality in the form of interesting projects in favor of human beings.
Although there are still many discoveries to be made in this field, it is encouraging to think that at last, artificial intelligence is beginning to find its own space for the benefit of humanity, providing new solutions that serve to make this world a better place to live, far from the deadly diseases that claim hundreds of lives around the world every day, without medical science being able, until now, to offer any greater hope.