The bacterium Streptococcus pneumoniae triggers severe illnesses such as pneumonia and meningitis, which are typically addressed through antibiotics. However, the pathogen’s evolving resistance is steadily diminishing the efficacy of these medications over time.
This process, called drug repurposing, saves time and cost while reducing risk. The researchers’ goal was to find new treatments for S. pneumoniae infections.
Researchers have employed artificial intelligence to identify existing drugs that could be repurposed to combat S. pneumoniae, a pathogen prioritized by the World Health Organization for its growing antimicrobial resistance. Building on previous AI applications for drug repurposing against other pathogens, the team utilized advanced machine learning techniques and multiple models to screen a drug library specifically for S. pneumoniae.
The study involved training three distinct machine learning algorithms using two datasets: one containing molecules active against S. pneumoniae and another with molecules inactive against a different drug-resistant bacterium. These algorithms were then applied to evaluate nearly 7,000 potential drugs for their ability to inhibit S. pneumoniae.
From the 7,000 candidates, 11 were chosen for further testing. Nine of these compounds successfully inhibited S. pneumoniae growth. Notably, one of the two most effective repurposed drugs demonstrated efficacy even against drug-resistant strains of the pathogen.
The researchers highlighted the cooperation of using three different AI models, which collectively enhanced the selection of candidate drugs, outperforming any single model. As the World Health Organization updates its priorities, AI-driven drug repurposing emerges as a cost-effective and efficient approach with proven success in addressing antimicrobial resistance.
Senior author Pedro J. Ballester said, “Artificial intelligence-guided drug repurposing has become a powerful strategy for combating antimicrobial resistance, particularly for pathogens for which some active molecules are already known and can be leveraged as a training or fine-tuning dataset.”
These findings highlight the benefits of using artificial intelligence models to identify novel drugs and rationally select drugs with the highest potential for repurposing, which could lead to more effective treatments for infections caused by S. pneumoniae. Given the urgent need for effective compounds in this field, artificial intelligence-driven approaches may offer a valuable solution.
Leave a Reply