findDropouts: Identify dropout-suspected zeros from all zero entries

In zky0708/2DImpute: 2DImpute: Imputing scRNA-seq data from correlations in both dimensions

Description

This function distinguishes zeros that are likely to be dropouts from those biologically true zeros based on pairwise Jaccard Index between cells.

Usage

findDropouts(
  exprs,
  t = 0.2,
  genes = NULL,
  ncores = 1,
  return_J = FALSE,
  verbose = TRUE
)

Arguments

exprs	An input log-transformed data matrix. The rows and columns correspond to genes and cells, respectively.
t	Threshold (between 0 and 1) for determining similar cells. Default is 0.2.
genes	A vector containing gene symbols that will get imputed. Default is NULL, in which case all available genes in the matrix exprs will be imputed.
ncores	Number of cores to use. Default is 1.
return_J	Whether to return the calculated pairwise Jaccard matrix between cells. Default is FALSE.
verbose	Whether to show the progress of dropout identification. Default is TRUE.

Value

A list with the following components:

dropout_ind	A matrix in which each row contains the indices of a dropout
J	(optional) Calculated pairwise Jaccard Index between cells

Examples

data(ge_10x_sample)
results1 <- findDropouts(ge_10x_sample, ncores = 2, return_J = TRUE)
# pairwise Jaccard Index between cells
J <- results1$J
# indices of identified dropouts
dropout_ind <- results1$dropout_ind

zky0708/2DImpute documentation built on March 12, 2020, 3:13 a.m.