This project integrates high-content cellular imaging and single-cell RNA sequencing to investigate adipocyte differentiation and cardiometabolic disease risk. Linking cell morphology with gene regulation remains a major gap in understanding how genetic and environmental factors influence adipocyte function. Data from adipose-derived mesenchymal stem cells across 120 human donors, profiled under multiple stimulations, include features from LipocyteProfiler and single-cell transcriptomes. Modality-specific embeddings are aligned in a shared latent space across genetic variation, CRISPR perturbations, and compound treatments. This integration, enabled by ML-based embedding models, reveals how risk factors and therapeutic interventions alter adipocyte biology. It also establishes a scalable framework that not only advances research in adipocyte differentiation and cardiometabolic disease but also extends to other biological systems and diseases, enabling cross-modal insights with therapeutic and translational relevance.

I am excited about my SuperUROP as it offers me the unique opportunity to conduct collaborative and impactful research at the intersection of computer science and biology. I have always been passionate about applying computational and machine learning approaches to solve complex biological problems, and I am grateful to be able to build my technical, professional, and communication skills while contributing to research that improves human health.