�鶹��Ƶ������ȫ

*This article is a part of the series “Artificial Intelligence and Its Applications: Perspectives From Across �鶹��Ƶ������ȫ,” highlighting the applications of AI in different fields and including insights from students and faculty. Stay tuned for future articles covering topics such as translation and design, or check out a previous article about law enforcement and manufacturing.  

In the wake of the COVID-19 pandemic, Qiang Guan, Ph.D., an associate professor in �鶹��Ƶ������ȫ University’s Department of Computer Science, redirected his research to focus on drug discovery and began partnering with Cleveland Clinic to merge biology and computing for better health outcomes. 

Drug discovery is the process of examining protein structures and identifying potential docking points for a molecule, such as a drug candidate, to bind with the protein. 

The collaboration started during the COVID-19 pandemic. Guan, who said he was originally a pure computer scientist focused only on improving hardware without paying mind to what issues were being solved, had a change of perspective during this time. When his mother-in-law passed away from COVID-19, he began to shift the focus of his research.  

“During the pandemic, I realized this computing can save people’s lives,” Guan said.  

Guan was inspired by a drug repurposing study from Cleveland Clinic during the pandemic that he saw in the news. The Cleveland Clinic scientist he spoke to said they were eagerly seeking support from the computer science community to form a complementary relationship of solving biology problems with computing solutions.  

Guan’s team consists of 10 doctoral students, two master’s students and seven undergraduate students. He believes that it’s especially important for undergraduate students to get practical hands-on training with AI. 

Traditional drug discovery requires highly experienced technicians who spend several weeks preparing protein structures. AI solutions such as Google DeepMind’s award-winning are drastically accelerating this process.  

AlphaFold uses deep learning technology to make predictions about the structure of proteins and can gather digital representations of them within seconds. The technology allows scientists to test thousands of proteins each day and quickly narrow down candidates. Guan said saving time in the drug discovery phase is helpful because clinical trials and Food and Drug Administration approval are also lengthy processes.  

“This can definitely accelerate the drug discovery procedure,” Guan said. “People want this fast development phase.” 

Guan has already published two papers in this area with Cleveland Clinic. The papers explain how AI can predict the functions of proteins and potential docking opportunities based on known structures, databases and proteins.  

Guan said the limitations of this technology are that AI can make mistakes and AI models are limited to what they know. It accelerates the procedure, but the recommended candidates still need to be validated manually. Guan and his team are working to build a quantum computing-based solution to further enhance drug discovery.  

“Quantum computing is a natural solution for such a problem,” Guan said. “It is faster and more accurate.”  

Quantum computing differs from classical computing because it’s not based on ones and zeroes; it operates from a physics-based perspective. According to Guan, this makes it perfect for addressing very specific questions with existing algorithms.  

Using the , Guan and his team have run applications that show promising results for predicting protein structures. They have found the predictions to be 20-25% more accurate than AlphaFold 3, the latest version. Guan said the work is currently being submitted to journals and the next step is to look for a quantum solution to the docking problem.  

Guan hopes that his research will continue to improve healthcare and make a real impact. 

“This is not just making life better, this is changing lives and saving lives,” Guan said. “We need that, and that’s why I started to focus my research on applying the systems we build to these specific problems.” 

This series of articles only scratches the surface of the applications of AI—we’d love to hear from you! Fill out the following survey to share how you use AI in your work:  

###