Researchers at Texas A&M University have developed a new method called spline-DV to analyze variability in gene expression using single-cell RNA sequencing (scRNA-seq) data. Traditional RNA sequencing studies focus on the average expression levels of genes across different conditions, but this approach ignores cell-to-cell variability, which plays a crucial role in biological processes like disease progression and treatment response.

The new spline-DV method goes beyond standard differential expression (DE) analysis by identifying genes that exhibit increased or decreased variability between two experimental conditions. By capturing these changes in variability rather than just shifts in average expression, researchers can uncover previously hidden insights into how cellular systems respond to different biological states.

Illustration of the spline-DV method proposed in this study

A Two spline-fit curves are created independently with scRNA-seq data of two conditions, blue for condition 1 and red for condition 2. These spline curves represent the “expected” expression profile for genes based on their statistical properties in each condition. B Two spline-fit curves are projected into the same 3D space for comparison. For a given gene such as Gene A, its position is compared with respect to the condition 1 spline-fit curve. The vector v→1 extends from the closest point on the blue curve (spline-fit) to the blue point (Gene A in condition 1). Similarly, for condition 2 (red point), the vector v→2 extends from the closest point on the red curve to the red point (Gene A in condition 2). The L2-norm of v→1 (or v→2) quantifies the deviation of the same Gene A from its expectation in condition 1 (or condition 2). C Given that the deviation is the shortest Euclidean distance from the blue (or red) point to the blue (or red) spline-fit curve, representing the level of expression variability of the gene in condition 1 (or condition 2), dv→=v→2−v→1, is used to measure the level of differential variability. The L2-norm of dv→ is called DV score. Note that, in B, v→1 and v→2 have their own origin with respect to their respective conditions; in C, v→1 and v→2 share the same origin to compute dv→.

To demonstrate its effectiveness, the team applied spline-DV to case studies in obesity, fibrosis, and cancer, showing that genes with altered variability are functionally relevant to these conditions. This means that analyzing gene expression variability, rather than just the mean, could improve our ability to detect key molecular changes driving disease.

By integrating this new approach into single-cell RNA sequencing studies, scientists can gain a more complete understanding of cellular behavior, which may lead to better diagnostic tools and targeted therapies for complex diseases.

Availability – Code implementing the spline-DV method in both R and MATLAB is available at https://github.com/cailab-tamu/Spline-DV along with example data sets for testing.