oneSENSE

oneSENSE projects high-dimensional mass cytometry and single-cell datasets into a one-dimensional space per predefined biological category using a modified t-distributed stochastic neighbor embedding (t-SNE) to enable categorical dimensionality reduction and biological interpretation.

Key Features:

• Categorical Dimensionality Reduction: Groups measured parameters into predefined biological categories and projects cells onto one dimension per category for category-specific representation.
• Modified t-SNE: Uses an adaptation of t-distributed stochastic neighbor embedding (t-SNE) to perform one-dimensional projections tailored to categorical analysis.
• Biological Annotation via Binned Heat Plots: Links each one-dimensional category axis to binned heat plots to annotate and visualize marker expression patterns.
• Enhanced Cellular Discrimination: Improves separation and interpretability of complex cellular phenotypes compared to conventional multi-dimensional t-SNE plots.

Scientific Applications:

• Human T cell phenotype analysis: Explores relationships between categories of human T cell phenotypes by organizing markers into category-specific one-dimensional axes.
• Mass cytometry data interpretation: Interprets high-dimensional mass cytometry datasets by reducing dimensionality per biological category while preserving marker-specific information.
• Single-cell high-dimensional datasets: Applied generally to single-cell datasets to dissect cellular heterogeneity across predefined marker categories.

Methodology:

Grouping measured parameters into predefined biological categories; applying a modified t-distributed stochastic neighbor embedding (t-SNE) algorithm to project cells onto a one-dimensional space per category; and using binned heat plots to annotate each dimension's biological relevance.