Data processing workflow for large-scale immune monitoring studies by mass cytometry

Area of Science:

• Immunology
• Computational Biology
• Data Science

Background:

• Mass cytometry enables deep immune monitoring but is susceptible to operator and instrument variability.
• Experimental and analytical design are crucial for maximizing data quality in mass cytometry.
• Artifacts and batch effects can compromise data integrity even in well-controlled experiments.

Purpose of the Study:

• To present a robust data processing pipeline for mass cytometry.
• To minimize experimental artifacts and batch effects in immune monitoring data.
• To enhance overall data quality for large-scale studies.

Main Methods:

• Utilized an R pipeline with packages: CATALYST (normalization, debarcoding), flowAI/flowCut (anomaly cleaning), AOF (file QC), flowClean/flowDensity (gating).
• Implemented CytoNorm for batch normalization and FlowSOM/UMAP for data exploration.
• Included a standardized sample processing protocol to complement the analysis pipeline.

Main Results:

• The pipeline effectively minimizes experimental artifacts and batch effects.
• Demonstrated improved data quality suitable for complex immune monitoring.
• The workflow is scalable for large-scale, multicenter, and multibatch studies.

Conclusions:

• The presented data processing pipeline enhances mass cytometry data quality.
• This scalable workflow is ideal for large-scale, multicenter, and retrospective immune monitoring studies.
• Integration of experimental and analytical protocols ensures reliable data for deep immune analysis.