A method for resolving coupling of spatial transcriptomes with cell lineage tracing
By constructing a bimodal graph topology network and a graph attention network, and combining the topological relationships of single-cell lineage trees, the problems of signal loss and distribution distortion in the coupled analysis of spatial transcriptomics and cell lineage tracing were solved, achieving high-precision reconstruction of single-cell developmental trajectories and improving biological analysis capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2026-04-13
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for coupled analysis of spatial transcriptomics and cell lineage tracing suffer from problems such as high in-situ capture loss rate, severe lineage contamination, and disruption of cell developmental kinship, leading to inaccurate signal interpolation and distorted abundance allocation.
By constructing a bimodal graph topology network based on physical spatial distance and gene expression similarity, using graph attention network for phylogenetic barcode interpolation, and combining the topological relationship of single-cell phylogenetic trees to construct a joint optimization objective function, spatial speckle deconvolution is performed to reconstruct the single-cell developmental trajectory map.
It achieves high-precision phylogenetic signal interpolation and abundance allocation, reconstructs four-dimensional spatiotemporal differentiation and migration trajectories at the single-cell level, improves biological fidelity, and provides a high-scientific-value algorithm architecture for tumor microenvironment analysis and embryonic development.
Smart Images
Figure CN122417153A_ABST