CytoNorm

CytoNorm normalizes high-dimensional flow cytometry data to remove batch effects and technical variation while preserving biologically relevant signals for accurate single-cell phenotyping.

Key Features:

• Batch Effect Correction: Uses shared control samples across batches to learn and apply batch-specific transformations that correct technical variation.
• Population-Specific Transformations: Clusters overall cellular distributions to identify subsets and computes subset-specific transformations that account for cell-type–dependent variation.
• Quantile Distribution Alignment: Computes quantile distributions from control samples for each subset and aligns these distributions using spline functions to derive normalization transformations.

Scientific Applications:

• Deep Phenotyping in Clinical Studies: Provides consistent preprocessing for deep phenotyping of cellular systems at clinical scale by reducing batch-induced variability.
• Single-Cell Immune Profiling: Facilitates accurate characterization of the human immune system at single-cell resolution by preserving biological signals across batches.

Methodology:

Clusters cellular distributions to define subsets, computes subset-specific quantile distributions from control samples, aligns these distributions with spline functions to derive transformations, and applies the derived transformations to all clinical samples within each batch.