AI Can Help Prevent Deaths From Antibiotic-Resistant Infections
MIT researchers have discovered a potent new antibiotic that can kill 35 powerful bacteria, using AI algorithm, shared The Financial Times on February 20.
Newly discovered antibiotic halicin can treat such diseases, as tuberculosis. It was discovered by machine learning algorithm developed by computer scientist Regina Barzilay.
In the paper published in the journal Cell, scientists detailed that machine learning algorithm was trained on 2,500 molecules, than used the library of 100,000,000 molecules to predict their effectiveness against pathogens and to create new molecules for the most resistant ones.
This discovery is especially important with the raise in pathogens, that do not respond to current antibiotics, shared MIT biological engineer James Collins, who predicts that “annual deaths due to antibiotic-resistant infections will grow to 10 million” in next 30 years.
Earlier MIT lab has discovered eight other promising molecules, like halicin, but “the key gaps are economic”, shared healthcare lawyer Kevin Outterson. Therefore, new drug discovery can only work only if scientists find a way to drastically reduce the cost of clinical trials for new antibiotics.