Nazreen Pallikkavaliyaveetil is a postdoctoral research fellow in the Department of Biomedical Engineering at the University of Michigan. Her research focuses on developing machine learning models that can learn structured, interpretable representations of complex biological systems to support therapeutic discovery and systems-level reasoning.
Currently, she is building machine learning models, including graph-based and transformer-based architectures, to analyze complex biological systems and predict how drugs might interact. By integrating multiple types of biomedical data, her work aims to uncover patterns in drug response and improve strategies for combination therapies. This research is part of her broader goal to create AI tools that are both scientifically grounded and practically useful for biomedical decision-making.
Nazreen’s long-term interests lie in designing multimodal models that integrate diverse biological datasets and apply this knowledge across a variety of biomedical tasks. These models aim to generate versatile representations of cells or patients that can support applications such as drug response prediction, biological data imputation, and disease outcome forecasting. A key emphasis of her work is interpretability, ensuring models highlight meaningful molecular patterns and provide human-readable explanations.
Nazreen is particularly interested in applying these tools to niche or underexplored diseases and tissues, where data is limited and traditional methods often fall short. As a MIDAS Postdoctoral Affiliate, she looks forward to collaborating across disciplines and participating in professional development programs that support leadership in AI-driven biomedical research. She is committed to developing models that are biologically grounded, robust in low-data settings, and useful for real-world translational applications.