A wrapper function for bcds. Annotate
doublets/multiplets using a binary classification approach to discriminate
artificial doublets from original data. Generate a doublet
score for each cell. Infer doublets if estNdbl
is TRUE
.
runBcds( inSCE, sample = NULL, seed = 12345, ntop = 500, srat = 1, verb = FALSE, retRes = FALSE, nmax = "tune", varImp = FALSE, estNdbl = FALSE, useAssay = "counts" )
inSCE | A SingleCellExperiment object.
Needs |
---|---|
sample | Character vector. Indicates which sample each cell belongs to. bcds will be run on cells from each sample separately. If NULL, then all cells will be processed together. Default NULL. |
seed | Seed for the random number generator. Default 12345. |
ntop | See bcds for more information. Default |
srat | See bcds for more information. Default |
verb | See bcds for more information. Default |
retRes | See bcds for more information. Default |
nmax | See bcds for more information. Default |
varImp | See bcds for more information. Default |
estNdbl | See bcds for more information. Default |
useAssay | A string specifying which assay in the SCE to use. |
A SingleCellExperiment object with bcds output appended to the colData slot. The columns include bcds_score and optionally bcds_call. Please refer to the documentation of bcds for details.
data(scExample, package = "singleCellTK") sce <- subsetSCECols(sce, colData = "type != 'EmptyDroplet'") sce <- runBcds(sce)#>#> [15:46:33] WARNING: amalgamation/../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.