BASiCS_Filter: Filter for input datasets
In BASiCS: Bayesian Analysis of Single-Cell Sequencing data

Description Usage Arguments Value Author(s) Examples

BASiCS_Filter indicates which transcripts and cells pass a pre-defined inclusion criteria. The output of this function used to generate a SingleCellExperiment object required to run BASiCS. For more systematic tools for quality control, please refer to the scater Bioconductor package.

BASiCS_Filter(
  Counts,
  Tech = rep(FALSE, nrow(Counts)),
  SpikeInput = NULL,
  BatchInfo = NULL,
  MinTotalCountsPerCell = 2,
  MinTotalCountsPerGene = 2,
  MinCellsWithExpression = 2,
  MinAvCountsPerCellsWithExpression = 2
)

`Counts`	Matrix of dimensions `q` times `n` whose elements corresponds to the simulated expression counts. First `q.bio` rows correspond to biological genes. Last `q-q.bio` rows correspond to technical spike-in genes.
`Tech`	Logical vector of length `q`. If `Tech = FALSE` the gene is biological; otherwise the gene is spike-in.
`SpikeInput`	Vector of length `q-q.bio` whose elements indicate the simulated input concentrations for the spike-in genes.
`BatchInfo`	Vector of length `n` whose elements indicate batch information. Not required if a single batch is present on the data. Default: `BatchInfo = NULL`.
`MinTotalCountsPerCell`	Minimum value of total expression counts required per cell (biological and technical). Default: `MinTotalCountsPerCell = 2`.
`MinTotalCountsPerGene`	Minimum value of total expression counts required per transcript (biological and technical). Default: `MinTotalCountsPerGene = 2`.
`MinCellsWithExpression`	Minimum number of cells where expression must be detected (positive count). Criteria applied to each transcript. Default: `MinCellsWithExpression = 2`.
`MinAvCountsPerCellsWithExpression`	Minimum average number of counts per cells where expression is detected. Criteria applied to each transcript. Default value: `MinAvCountsPerCellsWithExpression = 2`.

A list of 2 elements

Counts: Filtered matrix of expression counts
Tech: Filtered vector of spike-in indicators
SpikeInput: Filtered vector of spike-in genes input molecules
BatchInfo: Filtered vector of the 'BatchInfo' argument
IncludeGenes: Inclusion indicators for transcripts
IncludeCells: Inclusion indicators for cells

Catalina A. Vallejos cnvallej@uc.cl

set.seed(1)
Counts <- matrix(rpois(50*10, 2), ncol = 10)
rownames(Counts) <- c(paste0('Gene', 1:40), paste0('Spike', 1:10))
Tech <- c(rep(FALSE,40),rep(TRUE,10))
set.seed(2)
SpikeInput <- rgamma(10,1,1)
SpikeInfo <- data.frame('SpikeID' = paste0('Spike', 1:10),
                        'SpikeInput' = SpikeInput)

Filter <- BASiCS_Filter(Counts, Tech, SpikeInput,
                        MinTotalCountsPerCell = 2,
                        MinTotalCountsPerGene = 2,
                        MinCellsWithExpression = 2,
                        MinAvCountsPerCellsWithExpression = 2)
SpikeInfoFilter <- SpikeInfo[SpikeInfo$SpikeID %in% rownames(Filter$Counts),]