filterTrajFeaturesByCOV: Filter features by Coefficient of Variation (COV)

Description Usage Arguments Details Value Author(s) See Also Examples

Description

Filters trajectory features by their coefficient of variation.

Usage

1
filterTrajFeaturesByCOV(sce, threshold, design = NULL, show_plot = TRUE)

Arguments

sce

An SingleCellExperiment object

threshold

Minimum coefficient of variation; numeric value between 0 and 1

design

A numeric matrix describing the factors that should be blocked

show_plot

Indicates if plot should be shown (default: TRUE)

Details

For each trajectory feature x listed in the SingleCellExperiment object the coefficient of variation is computed by CoV(x) = sd(x) / mean(x). Features with a CoV(x) greater than threshold remain labeled as trajectory feature in the SingleCellExperiment object, otherwise they are not considered for dimensionality reduction, clustering and trajectory reconstruction. Please note that spike-in controls are ignored and are not listed as trajectory features.

To account for systematic bias in the expression data (e.g., cell cycle effects), a design matrix can be provided for the learning process. It should list the factors that should be blocked and their values per sample. It is suggested to construct a design matrix with model.matrix.

Value

A character vector

Author(s)

Daniel C. Ellwanger

See Also

trajFeatureNames model.matrix

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
# Simulate example data
set.seed(1101)
dat <- simulate_exprs(n_features=15000, n_samples=100)

# Create container
alist <- list(logcounts=dat)
sce <- SingleCellExperiment(assays=alist)

# Filter incrementally
trajFeatureNames(sce) <- filterTrajFeaturesByDL(sce, threshold=2)
trajFeatureNames(sce) <- filterTrajFeaturesByCOV(sce, threshold=0.5)

# Number of features
length(trajFeatureNames(sce)) #filtered
nrow(sce) #total

elldc/CellTrails documentation built on May 16, 2020, 4:40 a.m.