PhD student presents AI tool at BMES conference, publishes paper in Bioinformatics
block
Bioinformatics PhD student Mehrdad Zandigohar presented research at the 2025 Biomedical Engineering Society’s annual meeting in San Diego in October.
Zandigohar’s presentation focused on his tool, scRegulate, a model that infers how genes are regulated in single cells.
Working under Richard and Loan Hill Department of Biomedical Engineering Professor Yang Dai, the associate director of graduate studies, and interim director of the Center for Bioinformatics and Quantitative Biology, his research focuses on creating AI tools that transform large amounts of genomic data into clear, easily comprehensible insights. He also works to develop new ways to explore large language models (LLMs) to connect scientific predictions with existing knowledge in the literature..
“The challenge was to make powerful AI models meaningful for biology,” Zandigohar said. “Black-box results are hard to act on, so I set out to design a model whose internal parts map to real biology. The goal is to pair GPU-accelerated deep learning with explanations that scientists can use.”
Zandigohar added attending the BME conference was an incredible learning experience.
“I gained new perspectives on how others are applying deep learning and large language models to biological questions,” he said. “Talking with peers and experts helped me see the broader efforts in systems biology and biomedical AI. It also gave me valuable feedback for future directions of my work, including scRegulate.”
He added that conferences help students grow beyond the lab. Conferences help students gain confidence, connections, and fresh perspectives. For UIC, having students and faculty attend conferences and meetings is about visibility: showing the world that innovative, impactful research is happening here in the College.
The paper on scRegulate was recently published in Bioinformatics, a leading journal in computational biology and bioinformatics. Jalees Rehman, the Benjamin Goldberg Professor of the Department of Biochemistry and Molecular Genetics, is also the co-author.
Zandigohar said science isn’t only about data and algorithms; it’s really about people and impact. Conferences like BMES remind researchers that research is a community effort, and progress happens when ideas are shared, inspire each other, and build connections across different fields.
Zandigohar plans to defend his thesis in the spring and hopes his work will help improve human health.
“My strength is combining computer science and biology to answer therapeutic questions,” he said. “I plan to keep building AI tools that make complex data understandable and useful for precision medicine, from uncovering disease mechanisms to informing treatment.”