Abstract
The spatially resolved transcriptomics (SRT) field has revolutionized our ability to comprehensively leverage image and molecular profiles to elucidate spatial organization of cellular microenvironments. Current clustering analysis of SRT data primarily relies on molecular information and fails to fully exploit the morphological features present in histology images, leading to compromised accuracy and interpretability. To overcome these limitations, we have developed a multi-stage statistical method called iIMPACT. It includes a finite mixture model to identify and define histology-based spatial domains based on AI-reconstructed histology images and spatial context of gene expression measurements, and a negative binomial regression model to detect domain-specific spatially variable genes. Through multiple case studies, we demonstrate iIMPACT outperformed existing methods, confirmed by ground truth biological knowledge. These findings underscore the accuracy and interpretability of iIMPACT as a new clustering approach, providing valuable insights into the cellular spatial organization and landscape of functional genes within spatial transcriptomics data.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Conflict of Interests: The authors declare no potential conflicts of interest.