AI, Vaccines, and Climate-Resilient Crops: Pakistani Scientist on the Future of Computational Biology
From accelerating vaccine design to protecting global food security, computational biology is reshaping how science responds to real-world challenges. In an interview, Prof. Muhammad Tahir Ul Qamar of Government College University in Faisalabad, Pakistan, shared his insights on the limitations of current epitope prediction algorithms, the promise of AI in rapid vaccine development, and the role of pan-genomics in building climate-resilient crops.
MSTF Media reports:
As vaccine design increasingly relies on computational pipelines, epitope prediction has become a crucial step in identifying potential vaccine targets. But as Prof. Ul Qamar explained, predicted epitopes often fail to trigger strong immune responses in real-world conditions.
“The main issue is that many databases and algorithms are not regularly updated,” he said. “The data are there, but they’re not efficiently linked. Sometimes we get the same epitopes from multiple targets. They look good in theory, but in the translation stage, they don’t work.”
He emphasized the need to combine traditional prediction tools with artificial intelligence and feature-rich models to improve accuracy. “Right now, most algorithms rely too much on similarity scores,” he added. “That’s not enough anymore. We need to extract and use all available features.”
Reflecting on lessons from the COVID-19 pandemic, Prof. Ul Qamar also highlighted how immunoinformatics has transformed vaccine development.
“In the past, researchers spent years testing data and methods to identify vaccine targets,” he explained. “Now, with computational tools, we can quickly pinpoint target proteins and streamline the production process. This means faster and broader vaccine strategies during outbreaks.”
Beyond medical research, Prof. Ul Qamar’s work also extends to pan-genomics which “will shape the future of agricultural research.”
“Pan-genomics covers the genetic diversity of entire species,” he said. “Through this, we can modify and improve crops, develop molecular markers, and create varieties that are more resilient to diseases.”
When asked about the major computational challenges in integrating diverse biological datasets for drug discovery, he explained: “Many institutions don’t have access to high-end servers and the quality of the data determines the quality of the results. If the data are weak, the results are biased. I believe AI will help overcome these limitations and give us more reliable outcomes.”
Prof. Muhammad Tahir Ul Qamar, a Professor at Government College University, Faisalabad in Pakistan, was a participant of the 10th Science and Technology Exchange Program (STEP) which was held at Amirkabir University of Technology on the sidelines of the 2025 Mustafa(pbuh) Week. His research spans computational biology, immunoinformatics, pan-genomics, and data integration for drug discovery.
