Neighbor-seq

Neighbor-seq reconstructs physical cell interaction networks from massively parallel single-cell sequencing data to identify and annotate direct cell–cell interactions and ligand–receptor signaling within undissociated cell fractions without prior knowledge of sample cell types or multiplets.

Key Features:

• Network reconstruction: Reconstructs physical cell interaction networks directly from single-cell sequencing data.
• Interaction annotation: Identifies and annotates the architecture of direct cell–cell interactions.
• Ligand–receptor focus: Detects and highlights relevant ligand–receptor signaling within undissociated cell fractions.
• Sequencing modality: Leverages massively parallel single-cell sequencing data.
• Microanatomical delineation: Delineates microanatomical features across tissues including the small intestinal epithelium, terminal respiratory tract, and splenic white pulp.
• Complex topology capture: Captures complex topologies of cell communication networks, including cancer–immune–stromal interactions.
• Tumor applications: Applied to characterize interactions in pancreatic and skin tumors.
• Spatial concordance: Produces findings consistent with spatial transcriptomic data patterns.
• Performance: Operates with fast and scalable computational performance for large datasets.
• Minimal input requirements: Infers cell interactions from routine single-cell datasets without additional sample preparation or prior cell-type identification.
• Integrative scope: Bridges single-cell and spatial transcriptomics to study the organ-level cellular interactome.

Scientific Applications:

• Microanatomical mapping: Mapping cellular organization in the small intestinal epithelium, terminal respiratory tract, and splenic white pulp.
• Cancer microenvironment analysis: Characterizing cancer–immune–stromal communication networks in pancreatic and skin tumors.
• Spatial validation: Comparing and validating interaction patterns against spatial transcriptomic data.
• Organ-level interactomics: Studying organ-level cellular interactomes across health and disease states using routine single-cell datasets.

Methodology:

Reconstructs interaction networks and annotates direct cell–cell interactions by analyzing massively parallel single-cell sequencing data, focusing on ligand–receptor signaling within undissociated cell fractions and inferring contacts without prior cell-type or multiplet annotations.