Automated cell type annotation and exploration of single-cell signaling dynamics using mass cytometry

Area of Science:

• Biotechnology
• Bioinformatics
• Computational Biology

Background:

• Mass cytometry by time-of-flight (CyTOF) enables deep cellular heterogeneity analysis.
• High-dimensional CyTOF data analysis presents significant computational challenges.

Purpose of the Study:

• To develop a bioinformatics framework for automated cell population annotation and patient stratification using CyTOF data.
• To address the challenges of high-dimensional data analysis in CyTOF.

Main Methods:

• Deployment of a bioinformatics framework for automated cell annotation using reference datasets.
• Application of single-cell data for patient stratification.
• Utilized the Scaffold approach for cell annotation and XGBoost for survival prediction in a leukemia cohort.

Main Results:

• The Scaffold approach demonstrated a favorable balance between sensitivity and specificity in automated cell type annotation.
• Identified key signaling protein interactions predictive of short-term survival in leukemia patients.
• The framework provides automated and versatile CyTOF data analysis.

Conclusions:

• The developed framework offers an automated and versatile solution for CyTOF data analysis.
• This approach has significant potential for precision medicine applications, particularly in cancer research.
• Facilitates deeper insights into cellular heterogeneity and patient outcomes.