cyCombine allows for robust integration of single-cell cytometry datasets within and across technologies

Area of Science:

• Single-cell biology
• Computational biology
• Immunology

Background:

• Combining single-cell cytometry datasets enhances analytical power but is hindered by technical batch effects.
• Existing methods often struggle with integrating data from different experimental batches or techniques.

Purpose of the Study:

• To develop a robust method for integrating diverse single-cell cytometry datasets.
• To minimize technical variance while preserving biological signals across experiments and techniques.

Main Methods:

• Introduced cyCombine, a novel computational method for cytometry data integration.
• Demonstrated cyCombine's ability to merge data from flow cytometry, mass cytometry, and CITE-seq.
• Validated the method's performance without requiring technical replicates across datasets.

Main Results:

• cyCombine effectively minimizes technical variance between datasets.
• Biological variance and data structure are successfully maintained post-integration.
• The method exhibits linear scaling with cell number, facilitating large-scale dataset integration.

Conclusions:

• cyCombine provides a robust, accurate, and scalable solution for cytometry data integration.
• Enables comprehensive multi-dataset analyses and validation using public resources.
• Facilitates deeper insights into biological systems by leveraging combined cytometry data.