AI Helps Scientists Discover The First New Antibiotics In Nearly 60 Years
Dive inside a new era of medicine as artificial intelligence directs scientists to the discovery of ground-breaking antibiotics, a watershed moment in the fight against drug-resistant germs.
A groundbreaking class of antibiotics designed to combat drug-resistant Staphylococcus aureus (MRSA) bacteria has emerged through the utilization of more transparent deep-learning models.
Artificial intelligence (AI) has become a revolutionary force in the field of medicine, enabling scientists to unlock the first novel antibiotics in six decades.
The identification of a novel compound capable of eradicating a drug-resistant bacterium responsible for thousands of global fatalities annually may signify a pivotal moment in the battle against antibiotic resistance.
“The insight here was that we could see what was being learned by the models to make their predictions that certain molecules would make for good antibiotics,” shared James Collins, professor of Medical Engineering and Science at the Massachusetts Institute of Technology (MIT) and a study author.
“Our work provides a framework that is time-efficient, resource-efficient, and mechanistically insightful, from a chemical-structure standpoint, in ways that we haven’t had to date.”
Published in Nature, the results were collaboratively authored by a team of 21 researchers.
The study’s goal was to “unlock the black box.”
The project team harnessed a deep-learning model to forecast the activity and toxicity of the innovative compound.
Deep learning, employing artificial neural networks to autonomously learn and represent features from data without explicit programming, is increasingly integral in drug discovery, expediting the identification of potential drug candidates, predicting their properties, and refining the drug development process.
In this instance, researchers directed their attention to methicillin-resistant Staphylococcus aureus (MRSA).
MRSA infections span from mild skin ailments to severe, potentially life-threatening conditions such as pneumonia and bloodstream infections.
The European Centre for Disease Prevention and Control (ECDC) reports nearly 150,000 MRSA infections annually in the European Union, resulting in almost 35,000 annual deaths due to antimicrobial-resistant infections.
The MIT research team cultivated an extensively enlarged deep learning model utilizing expanded datasets.
To construct the training data, around 39,000 compounds underwent assessment for their antibiotic activity against MRSA. Subsequently, the resulting data and details on the chemical structures of the compounds were fed into the model.
“What we set out to do in this study was to ‘open the black box.’ These models consist of very large numbers of calculations that mimic neural connections, and no one really knows what’s going on underneath the hood,” explained Felix Wong, a postdoc at MIT and Harvard, and one of the study’s lead authors.
Unveiling a novel compound
To enhance the curation of potential medications, our team harnessed the power of three additional deep-learning models. These models underwent training to evaluate the toxicity of compounds across three distinct types of human cells.
By amalgamating these toxicity forecasts with the previously established antimicrobial prowess, we precisely identified compounds adept at combatting microbes while inflicting minimal harm to the human body.
Leveraging this suite of models, we meticulously screened approximately 12 million commercially available compounds.
The models pinpointed compounds from five diverse classes, sorted based on particular chemical substructures within the molecules, showcasing anticipated activity against MRSA.
Following this, we procured around 280 of these compounds and subjected them to rigorous testing against MRSA in a laboratory environment. This methodology led us to unearth two promising antibiotic candidates within the same class.
In experiments featuring two mouse models — one for MRSA skin infection and another for MRSA systemic infection — each of these compounds significantly reduced the MRSA population by a factor of 10.
Thanks for reading the entire article! We aim to deliver the most up-to-date and trending news and updates. If you have any further questions about any problem feel free to ask us at aatar@aatardigitalmedia.com.
