2. Computational Methods for Multimodal Spatiotemporal Integration

A core focus of our research is the development of sophisticated computational methods for integrating diverse spatiotemporal data types. We create advanced algorithms and tools to combine multiple layers of biological information, including spatial transcriptomics, proteomics, and imaging data, while preserving their spatial and temporal contexts. Our methods address the unique challenges of data alignment, normalization, and integration across different modalities and scales. We aim to construct comprehensive representations of tissue microenvironments that capture both spatial organization and temporal dynamics. These integrated approaches provide a more complete understanding of cellular interactions and tissue architecture than any single data type alone.

3.​ Mining Spatiotemporal Data for Biological Insights

T he third pillar of our research focuses on extracting meaningful biological insights and biomedical applications from integrated spatiotemporal data. We develop and apply advanced machine learning and statistical methods to analyze complex spatial patterns, cellular interactions, and temporal dynamics within tissues. Our approaches help identify key biological mechanisms, disease markers, and therapeutic targets. We particularly emphasize the translation of our findings into practical biomedical applications, such as improving disease diagnosis, understanding development and aging processes, and identifying new therapeutic strategies. This work bridges the gap between computational analysis and clinical applications, contributing to advances in precision medicine and therapeutic development.