A1070
Title: MisTIC: Missegmented transcript inference correction for improved spatial transcriptomics analysis
Authors: Yuqiu Yang - UT Southwestern Medical Center (United States) [presenting]
Erica DePasquale - Cincinnati Children Hospital (United States)
Yunguan Wang - Cincinnati Children Hospital (United States)
Abstract: Imaging-based spatial transcriptomics technologies such as 10X Xenium and MERSCOPE have revolutionized the ability to study gene expression in the spatial context of tissues. However, their data analysis pipelines rely heavily on cell segmentation algorithms, which often produce imperfect boundaries leading to transcript misassignment. This misassignment can significantly impact downstream analyses, including cell type identification, differential expression analysis, and cell-cell communication. The aim is to present MisTIC (missegmented transcript inference correction), a computational method designed to correct transcript misassignment errors resulting from imperfect cell segmentation. Through extensive simulation studies, it is demonstrated that MisTIC successfully pinpoints misassigned transcripts from negative controls. Application to real Xenium and MERSCOPE datasets reveals that MisTIC enhances the detection of cell type-specific markers and improves the resolution of cell-cell communication patterns. The method is computationally efficient and can be seamlessly integrated into existing spatial transcriptomics analysis pipelines. MisTIC represents a significant advancement in addressing a critical limitation of imaging-based spatial transcriptomics technologies, enabling more accurate biological insights from these powerful platforms.
