SplineDV

SplineDV identifies differentially variable genes across two experimental conditions in single-cell RNA sequencing data by constructing 3D splines from mean expression, coefficient of variance, and dropout rate statistics.

Key Features:

• 3D Spline Construction: Constructs a 3D spline from three key gene statistics: mean expression, coefficient of variance, and dropout rate, for each experimental condition.
• Variability Measurement: Measures gene variability by calculating the distance of observed mean, coefficient of variance, and dropout rate from the expected 3D spline.
• Differential Variability Detection: Compares variabilities between two conditions to identify genes with significantly increased or decreased expression variability.

Scientific Applications:

• Obesity Research: Identifies differentially variable genes relevant to cellular conditions in obesity studies.
• Fibrosis Research: Detects variability changes in genes functionally relevant to fibrotic conditions.
• Cancer Research: Reveals differentially variable genes associated with cancer cellular states and heterogeneity.

Methodology:

SplineDV constructs a 3D spline from mean expression, coefficient of variance, and dropout rate for each condition, then measures the distance of observed statistics from the expected spline to quantify variability, and finally compares variabilities between conditions to identify differentially variable genes.