Bacteria are becoming more and more resistant to drugs. However, artificial intelligence appears to be a huge asset in the search for organisms capable of thwarting antibiotic resistance. They could lead to new drugs against dangerous bacterial infections.
Antibiotic resistance is one of the greatest public health challenges, killing approximately 1.3 million people each year. Researchers even estimate that antibiotic resistance could cause ten million deaths per year by 2050.
Microbiome
Research into this resistance may have led to something to counter it. They received help from artificial intelligence (AI) which focused on the ‘global microbiome’.
The microbiome includes all the bacteria, fungi and other micro-mechanisms present in our intestines, but also in plants, corals, soil and animals. Researchers say nature has always been a good place to research new drugs, including antibiotics. Bacteria are omnipresent on our planet and have themselves developed antibacterial defense mechanisms, often in the form of mini-proteins (peptides).
Machine learning
For the study in Cell, researchers used artificial intelligence on the global microbiome. They used machine learning to search for antibiotics in a dataset containing tens of thousands of bacteria and other primitive organisms. This gave rise to almost a million potential ‘antibiotic linkages’, dozens of which showed promising activity against bacteria.
‘During the first test, 63 candidates out of a hundred killed at least one of the bacteria tested and sometimes even several,’ explains co-author César de la Fuente. ‘In some cases, these molecules were effective even at very low doses. Research that once took years can now be done in a few hours thanks to computers.’
