The MIT team has made a remarkable breakthrough in the fight against drug-resistant bacteria! After 7 years of hard work and dedication, they have discovered the first new class of antibiotics in decades, using deep learning models. Their AI model was able to learn which chemical structures were most likely to kill the bacteria while avoiding adverse effects on healthy cells. This is an incredible achievement, and will no doubt save thousands of lives.
Alexander Fleming was aware of the dangers associated with antibiotics, as he said in his Nobel Prize acceptance speech in 1945, “Then there is the danger that the ignorant man may easily underdose himself and by exposing his microbes to non-lethal quantities of the drug, make them resistant.” Unfortunately, this is exactly what happened with newer antibiotics. Misuse and incorrect prescriptions led to the bacteria they targeted gradually becoming resistant.
The team which Dr. Jim Collins heads at The Collins Lab at MIT started the Antibiotics-AI Project back in 2020 to address this. The project has a seven-year plan to develop seven new classes of antibiotics to treat seven of the world’s deadliest bacterial pathogens. By using machine learning, the team has already achieved one of their goals, having discovered a new class of antibiotics that can kill methicillin-resistant Staphylococcus aureus (MRSA) bacteria. This is a huge success, as drug-resistant MRSA infections kill up to 10,000 people every year in the US alone.
The challenge in creating a new antibiotic is that there are near-infinite molecular arrangements and it’s hard to know which of those will kill a specific bacteria. To make matters more difficult, researchers also need to make sure that the new compound doesn’t kill healthy cells. To tackle this, the MIT team created a database of 39,000 different compounds and their effects on Staphylococcus aureus. They also incorporated the cytotoxicity of the compounds by tracking the effects they have on human liver, skeletal muscle, and lung cells. With this, they used the deep learning AI model to screen 12 million commercially available compounds, narrowing down the list to 280 compounds to test against MRSA bacteria grown in a lab dish. Their tests led to the discovery that two of the compounds reduced MRSA by a factor of 10!
It’s understandable why pharmaceutical companies would not direct much of their research into antibiotics; it is not as lucrative as other medicines, and it has become increasingly difficult to discover new antibiotics. Luckily, AI has made it a lot easier. The research team even used an adapted Monte Carlo tree search algorithm to get an insight into their deep learning model’s decision-making process. This allows us to understand what led the model to select the compounds it did, giving researchers a better idea of where to look for more effective drugs.
The Collins Lab’s incredible achievement has been made possible with contributions from MIT, the Broad Institute, Integrated Biosciences, the Wyss Institute for Biologically Inspired Engineering, and the Leibniz Institute of Polymer Research in Dresden, Germany. AI sifting through millions of potential chemical arrangements has had a huge impact on drug discovery, and it looks like the Collins Lab’s seven-year project may have time to spare.
We should all be thankful for this incredible breakthrough in the fight against drug-resistant bacteria, and the team at MIT who have dedicated their time and effort to this project! The world is forever in their debt, and we can only hope that the success of their work will continue.