Reconstructing biologically coherent cellular profiles from imaging-based spatial transcriptomics

This seminar was delivered on 15 June 2026. For more information: https://spds.sydney.edu.au/ Speaker: Long Yuan, Johns Hopkins University Abstract: In imaging-based spatial transcriptomics, transcript-to-cell assignment shapes downstream biological interpretation, including cell typing, ligand–receptor inference, and niche characterization. However, two-dimensional segmentation of volumetric tissue often yields mixed cellular profiles, while cells without detected nuclei may be missed entirely, affecting downstream analyses. We present TRACER, a framework that refines cellular representations in imaging-based spatial transcriptomics by leveraging gene–gene coherence and spatial co-localization of transcripts observed directly in the data, without requiring external annotations or reference atlases. TRACER resolves mixed cellular profiles and reconstructs partial cells whose nuclei are not detected, enabling more complete representation of cells within tissue sections. We also introduce coherence-based metrics that quantify transcriptional purity and conflict, enabling platform-agnostic benchmarking of segmentation quality. Across diverse platforms, tissues, and segmentation methodologies, TRACER consistently improves the coherence of cellular profiles and the quality of downstream analyses. About the speaker: Long Yuan is a PhD candidate in Immunology and an M.S.E. candidate in Computer Science at Johns Hopkins University. His research focuses on developing scalable machine learning and statistical methods for spatial and single-cell omics, with applications in cancer biology and immune-metabolic diseases. His work spans spatial multi-omics, graph-based learning, and multimodal data integration. As a member of the Break Through Cancer GBM and Data Science TeamLab, he develops computational approaches for integrating and analyzing large-scale spatial omics datasets.