AI Models Identify Drug Candidates Against Pneumonia Bacteria
Researchers used three machine learning models to screen nearly 7,000 approved drugs for activity against Streptococcus pneumoniae.
Updated on Nov. 11, 2026 in Artificial Intelligence

An international study published in Advanced Science used three AI model types to screen nearly 7,000 approved drugs against Streptococcus pneumoniae, a bacterium that can cause pneumonia and meningitis. Nine of the eleven candidates selected for laboratory testing inhibited the bacterium's growth.
Why it matters
Antibiotic resistance is making S. pneumoniae infections harder to treat as the bacterium acquires changes that reduce existing medicines' effectiveness. Repurposing approved drugs, whose safety profiles are already known, could shorten the search for new treatments.
The screen covered nearly 7,000 drugs, and 9 of 11 selected candidates inhibited S. pneumoniae growth in laboratory tests. Whether these results translate to effective treatment in patients has not been established.
The details
Researchers used drug repurposing, testing already approved drugs for new uses, and trained three machine learning model types, decision tree ensembles, graph neural networks, and sequence-based transformer models, to analyze the drug set. One of the two repurposed drugs with the greatest efficacy also acted against medicine-resistant strains of the bacterium.
Timeline
Article on the study findings published November 11, 2026.
The Tech Race
The study extends computational drug repurposing, in which machine learning screens already-approved compounds for new uses, toward a resistance-prone respiratory pathogen. Its finding that combining three model types outperformed any single model points to ensemble approaches as a practical direction for this kind of screening.
The findings do not yet change how pneumonia or meningitis is treated, since the candidate drugs still require experimental validation before clinical use. Readers should expect any resulting therapies to take considerable further testing before they reach patients.
The takeaway
AI-supported drug repurposing may reduce the time and cost of finding new treatments for resistant infections, but each candidate must still prove itself in further experiments. The approach offers a faster starting point rather than a finished cure.
Further reading
For more on how computational methods are reshaping research, explore our Artificial Intelligence coverage.
Source note: This article includes information reported by PlusNews.







