AI Has Found the Solution for Drug-Resistant Bacteria 

Artificial intelligence examined millions of chemical compounds, leading to the discovery of new classes of antibiotics.

Artificial intelligence examined millions of chemical compounds, leading to the discovery of a class of antibiotics that can eradicate two types of drug-resistant bacteria. 

A new category of antibiotics capable of treating infections caused by drug-resistant bacteria has been identified, partly thanks to artificial intelligence. In 2019, over 1.2 million people died due to antibiotic resistance, and this number is expected to increase in the coming decades. This discovery could be a significant tool in the fight against it. 

The new antibiotic compounds showed potential in treating vancomycin-resistant Enterococcus, a bacterium that has developed resistance to the drug often used to treat MRSA infections, and Methicillin-resistant Staphylococcus aureus (MRSA), based on tests conducted on mice. 

Felix Wong of the Broad Institute of MIT and Harvard in Massachusetts states, “Our [AI] models tell us not only which compounds have selective antibiotic activity, but also why, in terms of their chemical structure.” 

The objective of Wong and his colleagues’ work was to show that AI-guided drug development could predict the biological effects of entire categories of drug-like compounds, rather than simply identifying specific receptors for drug particles to bind to. 

They studied the effects of over 39,000 compounds on three different types of human cells from the liver, skeletal muscle, and lungs, and on Staphylococcus aureus. The results served as training data for AI models, which identified patterns in the atoms and bonds comprising each compound. This enabled the AI systems to predict the antibacterial activity of these substances and any potential toxicity to human cells. 

Then, by analyzing 12 million compounds using simulated data, trained AI models identified 3,646 compounds with ideal drug-like qualities. Further calculations determined the chemical substructures responsible for the properties of each compound. 

The researchers found two non-toxic compounds capable of eliminating vancomycin-resistant Enterococci and MRSA by comparing similar substructures in various compounds. This approach enabled them to identify additional categories of potential antibiotics. 

The researchers successfully demonstrated the effectiveness of these drugs in treating skin and thigh infections caused by MRSA in mice. 

According to James Collins of the Broad Institute, a co-author of the study, resistance has been increasing against the few novel classes of antibiotics found to be effective against vancomycin-resistant Enterococci and MRSA, including drugs like lipopeptides and oxazolidinones. 

He stated, “Our work identifies a new class of antibiotics that complements these other antibiotics, one of the few in 60 years.” 

Leveraging this AI-guided methodology, researchers have started to develop entirely new antibiotics and discover new drug classes. This includes substances that specifically target aging, damaged cells implicated in diseases such as cancer and osteoarthritis. 

Inside Telecom provides you with an extensive list of content covering all aspects of the Tech industry. Keep an eye on our Medtech section to stay informed and updated with our daily articles.