scHOT

scHOT implements higher-order interaction testing as an R package for single-cell gene expression and spatial transcriptomics data to detect changes in variability and correlation across developmental pseudotime and spatial contexts.

Key Features:

• Higher-Order Interaction Analysis: Detects changes in higher-order structures such as variability and correlation among genes that are not captured by first-order differential expression tests.
• Flexible Application Contexts: Applies along continuous trajectories (developmental pseudotime), across discrete groups, and in spatial orientations for spatial transcriptomics data.
• Statistical Robustness: Employs a statistically robust framework to identify changes in higher-order interactions reliably.
• Modular Design: Provides a general, modular implementation accommodating various higher-order measurements (e.g., variability, correlation).

Scientific Applications:

• Developmental Biology: Identifies coordinated changes in gene interactions over time, demonstrated in embryonic mouse liver studies.
• Spatial Transcriptomics: Detects subtle changes in gene-gene correlations across regions of the mouse olfactory bulb.

Methodology:

Implemented as an R package that performs higher-order interaction testing by comparing variability and correlation measures across continuous trajectories (developmental pseudotime), discrete groups, or spatial regions, extending beyond first-order differential expression analysis.