ADTnorm

ADTnorm normalizes and integrates antibody-derived tag (ADT) abundance data across CITE-seq datasets to remove batch effects and align cell populations for improved cell-type identification and atlas-level analyses.

Key Features:

• Population Alignment: Accurately aligns cell populations with negative- and positive-expression of surface protein markers across datasets with varying experimental designs.
• Batch Effect Removal: Effectively removes technical variation across batches while preserving biological variation in ADT expression.
• Automated Threshold-Gating: Provides built-in utilities for automated threshold-gating of surface protein marker expression.
• Antibody Quality Assessment: Includes utilities for assessment of antibody staining quality to aid in titration optimization and antibody panel selection.
• Cross-Study Integration: Enables efficient integration of public CITE-seq datasets for atlas-level analyses.

Scientific Applications:

• Cell-Type Identification: Improves separation and identification of cell types by normalizing surface protein expression data across batches.
• Disease Biomarker Discovery: Identifies previously undetected disease-associated markers, as demonstrated in COVID-19 CITE-seq data analysis.
• Atlas-Level Immunophenotyping: Facilitates large-scale integration of CITE-seq datasets for comprehensive immune cell characterization across studies.
• Antibody Panel Optimization: Supports titration optimization and selection of antibody panels for CITE-seq experiments through staining quality assessment.

Methodology:

ADTnorm was benchmarked against 14 existing scaling and normalization methods across 13 public CITE-seq datasets.