Integrative Inference of Spatially Resolved Cell Lineage Trees using LineageMap.

Area of Science:

• Developmental Biology
• Computational Biology
• Genomics

Background:

• Understanding tissue growth and cell differentiation is crucial in biology.
• Spatially resolved lineage tracing offers multi-modal data (lineage, gene expression, spatial location) for studying these processes.
• Existing computational methods lack the ability to fully integrate these three data types for lineage reconstruction.

Purpose of the Study:

• To develop a computational method that integrates gene expression, lineage barcodes, and spatial location data for accurate cell lineage reconstruction.
• To infer high-resolution cell lineage trees and ancestral cell states and locations.
• To provide a scalable and interpretable tool for analyzing spatio-temporal cellular dynamics.

Main Methods:

• Introduced LineageMap, a hybrid lineage inference algorithm.
• Integrated distance-based and likelihood-based methods within a unified probabilistic framework.
• Utilized spatially resolved lineage tracing data with three modalities: lineage barcodes, gene expression profiles, and spatial locations.

Main Results:

• LineageMap accurately reconstructs cell lineage trees and ancestral cell locations from tri-modality single-cell data.
• The algorithm demonstrates superior accuracy compared to existing methods on simulated and experimental datasets.
• Revealed biologically coherent spatiotemporal trajectories, advancing the understanding of dynamic cellular ancestries.

Conclusions:

• LineageMap effectively bridges molecular lineage tracing with spatial and transcriptomic information.
• The framework enables advanced computational reconstruction of cellular ancestries in both time and space.
• Provides a scalable and interpretable tool for high-resolution lineage inference.