SpatialDE2

SpatialDE2 identifies spatially variable genes and segments tissue into expression-based zones to analyze spatial transcriptomics data.

Key Features:

• Integrated analysis: Unifies identification of tissue zones with detection of spatially variable genes for consistent, combined analyses.
• Bayesian statistical framework: Implements a Bayesian model that explicitly accounts for Poisson count noise in transcriptomic measurements.
• Computational efficiency: Achieves superior processing speeds compared to previous approaches, enabling analysis of large-scale spatial transcriptomics datasets.
• Validation: Performance and robustness have been validated using simulated data.

Scientific Applications:

• Mouse brain analysis: Applied to spatial transcriptomics profiles from mouse brain to characterize spatial gene expression patterns.
• Human endometrium analysis: Applied to spatial transcriptomics profiles from human endometrium to characterize spatial gene expression patterns.

Methodology:

Combines tissue zone identification and spatially variable gene detection within a Bayesian statistical framework that models Poisson count noise; methods were validated on simulated data and optimized for computational speed.