waddR

waddR applies statistical testing based on the 2-Wasserstein distance to detect and characterize complex distributional differences in gene expression data, including single-cell RNA sequencing (scRNA-seq).

Key Features:

• 2-Wasserstein-based statistical testing: Implements statistical tests that quantify distributional differences using the 2-Wasserstein distance to identify differences between conditions.
• Differential distribution testing: Provides a procedure for differential distribution testing that captures complex changes beyond mean shifts detected by standard differential expression methods.
• Decomposition of 2-Wasserstein distance: Decomposes the 2-Wasserstein distance into components reflecting mean, variance, and shape to interpret contributions to distributional differences.
• Application beyond scRNA-seq: Applies the mathematical framework to any setting that requires comparison of probability distributions beyond single-cell RNA sequencing.

Scientific Applications:

• Single-cell RNA sequencing: Detects and characterizes complex distributional differences in scRNA-seq gene expression across conditions or cell populations.
• Broad disciplinary distribution analysis: Enables comparison and interpretation of distributional differences in other fields that analyze probability distributions.

Methodology:

Performs differential distribution testing by computing the 2-Wasserstein distance between distributions and decomposing that distance into mean, variance, and shape components.