SpatialCPie

SpatialCPie evaluates clusters in spatial transcriptomics datasets using a multi-resolution clustering and visualization approach to aid interpretation of spatially resolved gene expression.

Key Features:

• Implementation: Implemented as an R/Bioconductor package.
• Input data: Operates on spatial transcriptomics and spatially resolved gene expression profiles.
• Multi-resolution clustering: Offers a multi-resolution approach to clustering to explore data at different levels of granularity.
• Pie-chart visualizations: Generates pie charts that depict the similarity between spatial regions and identified clusters.
• Cluster graph: Produces a cluster graph that illustrates relationships among clusters across different resolutions.
• Cluster evaluation: Supports evaluation and selection of cluster number by exposing interrelationships among clusters.
• Demonstration: Has been demonstrated on publicly available datasets.

Scientific Applications:

• Cluster evaluation in spatial transcriptomics: Evaluating and comparing clustering solutions in spatial transcriptomics data.
• Determining cluster granularity: Determining an appropriate number of clusters and assessing cluster granularity.
• Interpreting spatial gene expression: Interpreting spatial patterns of gene expression across spatial regions.
• Supporting downstream analyses: Informing downstream analyses and biological interpretation of spatially resolved transcriptomic studies.

Methodology:

Uses a multi-resolution clustering strategy together with pie-chart visualizations of region–cluster similarity and a cluster-graph representation of cluster relationships across resolutions.