Polar Gini Curve

Polar Gini Curve identifies single-cell biomarkers by integrating gene expression with 2D embedded spatial information through geometric analysis of cell-point distributions.

Key Features:

• Integration of Data Types: PGC combines two-dimensional embedded spatial information with gene expression profiles to relate spatial position to molecular state.
• Characterization of Cell Distribution: PGC uses Polar Gini Curves to describe the shape and evenness of cell-point distributions within clusters.
• Marker Identification Framework: PGC constructs and compares two Polar Gini Curves—one for cells expressing a specific gene and one for all cells in a cluster—and uses the proximity between curves as an indicator of marker potential.

Scientific Applications:

• Simulation case studies: Applied in simulations to demonstrate utility in detecting biomarkers.
• Neonatal mouse heart single-cell analysis: Identified potential biomarkers that may define novel subtypes of cardiac muscle cells in neonatal mouse heart data.
• Research impact areas: Enables study of cellular heterogeneity with relevance to developmental biology, oncology, and regenerative medicine.

Methodology:

For each gene, construct two Polar Gini Curves—one from cells expressing the gene and one from all cells in the cluster—and assess the similarity between these curves to determine the likelihood of the gene being a cluster-specific biomarker.