filterCells: Filter Cells

Description Usage Arguments Value Author(s) See Also Examples

Description

Apply gene detection, novelty score, and mitochondrial abundance cutoffs to cellular barcodes. By default we recommend applying the same filtering cutoff to all samples. The filtering parameters now support per-sample cutoffs, defined using a named numeric vector. When matching per sample, be sure to use the sampleNames() return values (i.e. the sampleName column in sampleData()).

Usage

1
2
3
4
5
6
filterCells(object, ...)

## S4 method for signature 'SingleCellExperiment'
filterCells(object, minUMIs = 0L,
  maxUMIs = Inf, minGenes = 0L, maxGenes = Inf, minNovelty = 0L,
  maxMitoRatio = 1L, minCellsPerGene = 10L)

Arguments

object

Object.

...

Additional arguments.

minUMIs

Minimum number of UMI disambiguated counts per cell.

maxUMIs

Maximum number of UMI disambiguated counts per cell.

minGenes

Minimum number of genes detected.

maxGenes

Maximum number of genes detected.

minNovelty

Minimum novelty score (log10 genes per UMI).

maxMitoRatio

Maximum relative mitochondrial abundance (0-1 scale).

minCellsPerGene

Include genes with non-zero expression in at least this many cells.

Value

bcbioSingleCell, with filtering information slotted into metadata() as filterCells and filterParams.

Author(s)

Michael Steinbaugh

See Also

Seurat::CreateSeuratObject().

Other Quality Control Functions: barcodeRanksPerSample, metrics, plotCellCounts, plotGenesPerCell, plotMitoRatio, plotMitoVsCoding, plotNovelty, plotQC, plotReadsPerCell, plotUMIsPerCell, plotZerosVsDepth

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
# SingleCellExperiment ====
object <- cellranger_small
show(object)

x <- filterCells(object, minNovelty = 0L)
show(x)
metadata(x)$filterParams

# Per sample cutoffs
sampleNames(object)
x <- filterCells(
    object = object,
    minUMIs = c(pbmc4k = 100)
)
show(x)
metadata(x)$filterParams

roryk/bcbioSinglecell documentation built on May 27, 2019, 10:44 p.m.