What is Region-Aware Matching?

It is a methodology used in bioinformatics to improve the integration of single-cell RNA sequencing and spatial proteomics data by considering tissue architecture, such as T cell zones or B follicles, during the matching process.

Can I customize the matching constraints?

Yes, the skill supports soft priors rather than hard filters, allowing you to define compatible, neutral, and incompatible weights for specific cell-type/region pairs.

Which clustering algorithms are used for region detection?

The skill utilizes DBSCAN (Density-Based Spatial Clustering of Applications with Noise) because it effectively captures irregular tissue region shapes compared to standard k-means.

How does this improve standard MaxFuse workflows?

Standard MaxFuse treats tissue as homogeneous. This skill adds spatial priors and neighborhood features to ensure scRNAseq cells are matched to biologically appropriate regions in the spatial data.

Region-Aware Spatial Cell Matching

Name: Region-Aware Spatial Cell Matching
Author: smith6jt-cop

bysmith6jt-cop

0•

データサイエンスとML

Optimizes CODEX and scRNAseq data integration by incorporating tissue heterogeneity into spatial cell matching workflows.

This skill enhances the MaxFuse framework by introducing spatial context into the cell-matching process, specifically designed for complex tissue environments like the spleen. By accounting for distinct tissue regions—such as B cell follicles, T cell zones, and red pulp—it prevents biologically implausible matches where cells from scRNAseq are incorrectly paired with incompatible CODEX regions. It utilizes a three-pronged approach involving prior-weighted distance interpolation, neighborhood-augmented features, and DBSCAN-based region detection to ensure high-fidelity multi-omic integration while maintaining the spatial architectural integrity of the sample.

主な機能

01Neighborhood-augmented feature extraction for enhanced spatial context

020 GitHub stars

03Support for Python-based spatial toolkits including MaxFuse, Scanpy, and Scikit-learn

04Prior-weighted distance interpolation to enforce soft biological constraints

05Automated tissue region detection using marker-based DBSCAN clustering

06Flexible configuration for cell-type-to-region compatibility weights

ユースケース

01Integrating scRNAseq datasets with CODEX spatial proteomics in heterogeneous tissues

02Refining cell-type mapping to maintain tissue architecture integrity

03Automating region labeling based on dominant marker gene expression and spatial coherence

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add smith6jt-cop/skills_registry region-aware-matching

For use in Claude.ai and ChatGPT

Download Skill