CSOA

CSOA calculates per-cell gene signature scores in single-cell RNA sequencing (scRNA-seq) data by computing, ranking, filtering, and scoring overlaps between cell sets defined by highly expressing cells for each gene in a signature.

Key Features:

• Cell Set Construction: For each gene in the signature, CSOA constructs a set consisting of cells that highly express that gene.
• Pairwise Overlap Analysis: All overlaps of pairs of cell sets are computed, ranked, filtered, and scored.
• Per-Cell Scoring: The CSOA per-cell score is calculated by summing all products of overlap scores and min-max-normalized expression of the two involved genes.
• Multiple Input Formats: CSOA can run on a Seurat object, a SingleCellExperiment object, a matrix, and a dgCMatrix.

Scientific Applications:

• Gene Signature Scoring in scRNA-seq: CSOA enables calculation of per-cell gene signature scores to assess signature activity at single-cell resolution.

Methodology:

CSOA constructs cell sets for each signature gene based on high expression, computes all pairwise overlaps between these sets, ranks and filters the overlaps, scores them, and derives per-cell scores by summing products of overlap scores with min-max-normalized gene expression values.