A generative AI tool for molecular design has created a new antibiotic that has shown promising preclinical results against methicillin-resistant Staphylococcus aureus (MRSA).
The SyntheMol-RL generative model, described in Molecular Systems Biology, could speed drug discovery and help in the fight against antibiotic resistance.
The algorithm uses reinforcement learning to rapidly design easily synthesizable small-molecule drug candidates from a massive chemical space of 46 billion compounds.
It created a compound that the researchers named synthecin, which was effective against MRSA wound infection in a mouse model, showing its utility for real-world drug discovery.
“We used our model to design new antibiotics, but it’s capable of so much more,” said researcher Jon Stokes, PhD, from McMaster University.
“We built it to be disease agnostic, meaning it could just as easily generate novel drug candidates for diabetes or cancer or other indications.”
The rapid spread of antibiotic resistance is a critical challenge facing modern medicine. In 2019, just under five million deaths were linked with drug-resistant bacteria and this number is expected to more than double by 2050 if the emergence of antimicrobial resistance continues to outpace the creation of new antibiotics.
Stokes and team examined whether SyntheMol-RL could identify potential antibiotics for MRSA, an infection listed by the World Health Organization as a high priority for new antibiotics.
It replaces SyntheMol, a previous incarnation that was not as effective for exploring the chemical space and was not able to optimize more than one molecular property, which is a necessity in real-world drug discovery.
The second-generation model uses reinforcement learning, which enables it to rapidly explore massive combinatorial chemical spaces with tens of billions of molecules for promising compounds that are easy to synthesize.
The researchers deployed SyntheMol-RL to identify compounds that simultaneously possessed the multiple drug-like properties of antibacterial activity against MRSA and aqueous solubility.
Next, they synthesized and experimentally tested 79 compounds designed by two variants of SyntheMol-RL and found a corresponding two and 11 potent hits.
One of these compounds, which they named synthecin, was able to fully arrest the growth of MRSA in a murine wound infection model.
“These results demonstrate that SyntheMol-RL is an effective and flexible framework for drug design applications,” the authors maintained.
They added: “SyntheMol-RL is compatible with any property predictor and combinatorial chemical space, it can be readily extended to a wide variety of drug discovery and molecular design problems.”
The post AI Tool Creates Designer Antibiotics appeared first on Inside Precision Medicine.