sepal

sepal identifies transcriptional profiles with spatial patterns by simulating diffusion-based behavior of individual transcripts to analyze spatial transcriptomics data.

Key Features:

• Diffusion-Based Modeling: Simulates diffusion of individual transcripts to extract genes with spatial patterns and provides an alternative to traditional statistical hypothesis-testing frameworks.
• Quantification of Spatial Structure: Provides a metric that quantifies spatial structure to distinguish patterned profiles from background noise.
• Performance and Efficiency: Demonstrates superior time performance on multiple cores and reduced dependence on gene expression levels compared with existing methods.

Scientific Applications:

• Identification of Spatially Patterned Genes: Identifies genes with distinct spatial profiles that are associated with key biological processes or cell-type characteristics in real datasets.
• Versatility Across Omics Fields: Applicable across various omics fields where collection and analysis of spatial signals are performed.

Methodology:

Simulating the diffusion process of individual transcripts for spatial transcriptomics analysis; validated on synthetic data; implemented in Python.