fluidigmSC_identifyExpOutliers: An outlier detection function

In JinmiaoChenLab/uSORT: uSORT: A self-refining ordering pipeline for gene selection

Description

An outlier detection function identifies cells with median expression below that of the bulk, reproduced from FluidigmSC package.

Usage

fluidigmSC_identifyExpOutliers(log2ex_data, expression_data_raw, threshold,
  step, fine_step, num_fine_test, pct_goodsample_threshold = 0.5,
  quantile_threshold = 0.95, low_quantile_threshold = 0.25,
  min_gene_number = 25, lod)

Arguments

log2ex_data	A data frame containing log2 tranformed expression values, with rownames of genes and colnames of cells.
expression_data_raw	A data frame containing raw expression values, with rownames of genes and colnames of cells.
threshold	A value in raw expression used as the starting threshold value.
step	An integer number indicating the increment of threshold value at each iteration.
fine_step	An integer number indicating the increment of threshold value at each iteration, at the refining stage.
num_fine_test	An integer number indicating the number of iteration of the refining stage.
pct_goodsample_threshold	A fraction value indicating the minimum percentage of samples on which the representative genes are detectable.
quantile_threshold	A probability of gene detection rate above which a sample is considered as good sample.
low_quantile_threshold	A probability of average gene expression value below which a sample is taken as an outlier.
min_gene_number	An integer indicating the minimum size of representative genes.
lod	A value of limit of detection in the unit of TPM/CPM/RPKM.

Value

A vector of character stating the IDs of outlier cells.

JinmiaoChenLab/uSORT documentation built on May 7, 2019, 10:53 a.m.