Abstract
Advancements in cytometry technologies have enabled quantification of up to 50 proteins across millions of cells at single cell resolution. Analysis of cytometry data routinely involves tasks such as data integration, clustering, and dimensionality reduction. While numerous tools exist, many require extensive run times when processing large cytometry data containing millions of cells. Existing solutions, such as random subsampling, are inadequate as they risk excluding rare cell subsets. To address this, we propose SuperCellCyto, an R package that builds on the SuperCell tool which groups highly similar cells into supercells. SuperCellCyto is available on GitHub (https://github.com/phipsonlab/SuperCellCyto).
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Emails: GHP: putri.g{at}wehi.edu.au, GH: george.howitt{at}petermac.org, FMW: felix.marsh-wakefield{at}sydney.edu.au, TMA: thomas.ashhurst{at}sydney.edu.au, BP: phipson.b{at}wehi.edu.au
-Extra analysis on the b cell data where cell type identification at the supercell level vs subsampled data is compared, and found the latter yielded poorer outcome (only able to detect 7 out of 10 subsets). - Additional analysis (tables and figures) that better demonstrate the superiority of purity scores of supercells. - Imprved batch effect correction section by incorporating 3 metrics that account for preservation of biological signals. - Improved the label transfer section by including weighted accuracy metric of the cell type label transfer and stressing the method is performant but should only serve as a complement to manual annotation rather than a straight replacement. - Improved the vignettes and instructions for the package and built a website to host it https://phipsonlab.github.io/SuperCellCyto/index.html - Supplemental files updated
