summarizeSweep: summarizeSweep

In chris-mcginnis-ucsf/DoubletFinder: DoubletFinder is a suite of tools for identifying doublets in single-cell RNA sequencing data.

summarizeSweep

Description

Summarizes results from doubletFinder_ParamSweep, computing the bimodality coefficient across pN and pK parameter space. If ground-truth doublet classifications are available, then ROC analysis is performed, enabling optimal DoubletFinder parameter selection.

Usage

sweep.stats <- summarizeSweep(sweep.list, GT = FALSE)
sweep.stats <- summarizeSweep(sweep.list, GT = TRUE, GT.calls = classifications)

Arguments

sweep.list	List of pANN vectors across pN-pK space, as produced by doubletFinder_ParamSweep.
GT	Logical set to TRUE when ground-truth doublet classifications are available for ROC analysis. Default set to FALSE.
GT.calls	An nCell-length character vector of ground-truth doublet classifications (e.g., "Singlet" or "Doublet") used to gauge performance of logistic regression models trained using pANN vectors during ROC analysis.

Value

Dataframe with bimodality coefficient values at each pN-pK parameter set. If GT = TRUE, dataframe also includes AUC for each pN-pK parameter set computed during ROC analysis.

Examples

sweep.list <- paramSweep(seu)
sweep.stats <- summarizeSweep(sweep.list, GT = FALSE)
bcmvn <- find.pK(sweep.stats)

chris-mcginnis-ucsf/DoubletFinder documentation built on April 12, 2024, 5:44 p.m.