St. John's University's Bukhari Lab has secured $562,000 in funding from the National Institutes of Health (NIH) to develop artificial intelligence systems that can identify existing medicines with potential new applications for treating serious respiratory diseases. The three-year initiative, called RESPOND, represents a growing shift in drug discovery away from developing entirely new medications and toward repurposing drugs already approved for other conditions.

Why Is Drug Repurposing Such a Big Deal in Medicine?

Developing a brand-new pharmaceutical drug typically requires years of laboratory research, clinical testing, regulatory review, and billions of dollars in investment. Drug repurposing offers a faster, more cost-effective alternative by examining whether medicines already developed and approved for one condition might treat another. A real-world example illustrates this approach: GLP-1 receptor agonists were initially created to treat Type 2 diabetes, but researchers later discovered that several of these medicines could produce significant weight loss, leading to their use as obesity treatments.

Artificial intelligence can dramatically accelerate this process by narrowing an enormous pharmaceutical search space into a focused set of drug candidates that researchers can then evaluate in the laboratory. Rather than manually reviewing thousands of potential drug-disease combinations, AI systems can process vast amounts of biomedical data to identify the most promising leads.

How Does the RESPOND Project Actually Work?

The RESPOND initiative combines two complementary AI approaches: machine learning, which finds patterns in data, and neuro-symbolic AI, which integrates human-readable knowledge and reasoning. This hybrid approach is crucial because biomedical researchers need more than just predictions. They need to understand the scientific evidence behind those predictions and how different pieces of knowledge connect.

"Biomedical researchers need more than a prediction. They need to understand the evidence behind that prediction, how different pieces of scientific knowledge are connected, and why a particular candidate should be examined more closely. Trust, transparency, and scientific reasoning are central to the RESPOND project," said Syed Ahmad Chan Bukhari, Associate Professor in the Division of Computer Science, Mathematics, and Science at St. John's University.

Syed Ahmad Chan Bukhari, Associate Professor, Division of Computer Science, Mathematics, and Science, St. John's University

The project will examine relationships among drugs, diseases, genes, biological pathways, and other forms of scientific evidence. By making AI reasoning transparent and scientifically grounded, the system helps researchers understand not just which drugs might work, but why they might work.

How to Build AI Systems That Researchers Can Actually Trust

• Explainability: AI systems must show their reasoning and evidence, not just deliver a final answer, so researchers can evaluate whether the logic is sound.
• Scientific Grounding: Predictions must be anchored in reliable biomedical knowledge and evidence, not just statistical correlations in data.
• Auditability: The decision-making process must be transparent enough for researchers to trace how the AI arrived at its conclusions and verify each step.

The Bukhari Lab prioritizes these principles across all its work in health care and biomedical research. According to Dr. Bukhari, the real value of artificial intelligence in science is not simply its ability to process more information, but rather its capacity to help scientists ask better questions, connect evidence more effectively, and make more informed research decisions.

Dr. Bukhari

Who Will Be Involved in This Research?

The NIH grant will be distributed annually from July 2026 through June 2029. During this period, Dr. Bukhari's team will include machine-learning engineers and both undergraduate and graduate students from St. John's University studying computer science, pharmacy, and related health disciplines. These students will gain hands-on experience in AI model development, biomedical data analysis, computational drug discovery, and validation studies with external collaborators.

This educational component reflects a broader institutional commitment to training the next generation of researchers at the intersection of artificial intelligence and biomedical science. Students will contribute directly to an NIH-funded effort while learning skills that are increasingly essential in modern drug discovery.

The RESPOND project builds on the Bukhari Lab's track record of securing competitive federal funding. In 2024, the lab received a $550,000 grant from the US National Science Foundation to develop an AI solution for the time-consuming process of medical coding and health care billing, demonstrating the lab's expertise in applying AI to complex health care challenges.

"The value of artificial intelligence is not simply that it can process more information. Its real value comes from helping scientists ask better questions, connect evidence more effectively, and make more informed research decisions. We hope RESPOND will contribute to a faster, more transparent, and more thoughtful approach to discovering new treatment possibilities," noted Dr. Bukhari.

Syed Ahmad Chan Bukhari, Associate Professor, Division of Computer Science, Mathematics, and Science, St. John's University

The RESPOND initiative demonstrates the growing importance of interdisciplinary collaboration in addressing complex health care challenges. By bringing together expertise in computer science, biomedical research, health informatics, and translational medicine, the project aims to help scientists navigate vast amounts of biomedical knowledge and identify treatment opportunities that might otherwise remain undiscovered.