Voyager

Voyager implements exploratory spatial data analysis (ESDA) methods for spatial single-cell genomics data stored in the SpatialFeatureExperiment (SFE) S4 class, adapting geospatial statistical techniques to spatial -omics analysis.

Key Features:

• Univariate Spatial ESDA: Computes Moran's I with permutation testing and correlograms for global spatial autocorrelation analysis.
• Bivariate Spatial ESDA: Implements Lee's L and cross variogram for analyzing spatial relationships between two variables.
• Multivariate Spatial ESDA: Includes MULTISPATI PCA and multivariate local Geary's C for analyzing spatial patterns across multiple variables simultaneously.
• Data Structure Integration: Uses SpatialFeatureExperiment (SFE), which combines Simple Feature with SingleCellExperiment and AnnData to represent geometries bundled with gene expression data.
• Visualization: Provides plotting functions for SFE data and ESDA results.

Scientific Applications:

• Spatial -omics Analysis: Enables application of geospatial statistical methods to spatial single-cell genomics datasets, including detection of biologically relevant negative spatial autocorrelation.
• Cross-Platform Reproducibility: Supports consistent analysis workflows across R/Bioconductor and Python implementations with compatibility testing.

Methodology:

Voyager systematically adapts geospatial ESDA methods including local, bivariate, and multivariate spatial statistics to spatial -omics data analysis through the SpatialFeatureExperiment data structure.