poolCells: Pool cells for pre-processing

In MarioniLab/cydar: Using Mass Cytometry for Differential Abundance Analyses

Pool cells for pre-processing

Description

Construct a flowFrame object by pooling cells from multiple (barcoded) samples, for use in common transformation and gating.

Usage

poolCells(x, equalize=TRUE, n=NULL)

Arguments

x	A named list of numeric matrices, where each matrix corresponds to a sample and contains expression intensities for each cell (row) and each marker (column). Alternatively, a ncdfFlowSet object containing the same information.
equalize	A logical scalar specifying whether the same number of cells should be taken from each sample for pooling. If FALSE, all cells are used from all samples.
n	A numeric scalar specifying the number of cells to be used from each sample if equalize=TRUE. If NULL, this is set to the number of cells in the smallest sample.

Details

The idea is to use the pooled set of cells to estimate common parameters such as transformation values and gating thresholds. Otherwise, if these parameters were estimated separately for each sample, they may distort the comparisons between samples. This function is typically used to generate an object for use in estimateLogicle or in various gating functions like outlierGate. This yields parameter values that can be applied to the full set of cells in the original x object.

Value

A flowFrame object containing cells pooled from all samples.

Author(s)

Aaron Lun

See Also

flowFrame, outlierGate, estimateLogicle

Examples

### Mocking up some data: ###
set.seed(100)
nmarkers <- 40
marker.names <- paste0("X", seq_len(nmarkers))
nsamples <- 10
sample.names <- paste0("Y", seq_len(nsamples))

x <- list()
for (i in sample.names) {
    ex <- matrix(rexp(nmarkers*1000, 0.01), ncol=nmarkers, nrow=1000)
    colnames(ex) <- marker.names
    x[[i]] <- ex
}

### Running the function: ###
ff <- poolCells(x)
ff

### Using for estimation: ###
library(flowCore)
trans <- estimateLogicle(ff, colnames(ff))
ff <- transform(ff, trans) # or, apply to original data.

MarioniLab/cydar documentation built on Sept. 7, 2024, 6:24 a.m.