The discovery was made using a machine-learning algorithm. This technology enabled scientists to discover a powerful new antibiotic compound.
The importance of the antibiotic has been shown through various tests, where the chemical was challenged against several disease-causing bacteria. Among the microbial cohort were some organisms previously identified to be resistant to mot antibiotics. Further studies were undertaken using mice, yielding similarly successful results.
READ MORE: Genetic testing can identify antibiotic resistance
The reason why there is strong scientific interest in finding new antibiotics is due to the phenomenon of antimicrobial resistance. This is a significant global health issue since the pace at which bacteria are becoming resistant to common antibiotic treatments is increasing. This means that infections that were once easy to treat are no longer certain of being tackled through existing medications. The consequence is that routine operations or transplants now present additional risks.
With the new discovery from MIT, the algorithm processed millions of chemical compounds, processing this vast data set in just a few days. This approach also avoided the necessity of running thousands of experiments; only those compounds selected by the machine learning program as having strong potential need be tested.
The use of computer models for drug screening and other applications is captured by the term “in silico.”
According to lead researcher Professor James Collins: “Our approach revealed this amazing molecule which is arguably one of the more powerful antibiotics that has been discovered.”
He molecule selected has been named halicin (with a reference to the computer in 2001: A Space Odyssey). The drug was shown to be effective against Escherichia coli, as part of the tests. Further bacterial killing effects were demonstrated using other organisms of concern, such as Clostridium difficile, Acinetobacter baumannii, and Mycobacterium tuberculosis.
The bacterial killing properties of halicin arise from the compound’s ability to disrupt the electrochemical gradient across bacterial cell membranes, which triggers cell death.
ALSO READ: Antibiotic use may lead to heart problems
The research has been published in the journal Cell. The research paper is titled “A Deep Learning Approach to Antibiotic Discovery.”
