runNMDS: Perform non-metric MDS on sample-level data

In FelixErnst/mia: Microbiome analysis

Perform non-metric MDS on sample-level data

Description

Perform non-metric multi-dimensional scaling (nMDS) on samples, based on the data in a SingleCellExperiment object.

Usage

getNMDS(x, ...)

## S4 method for signature 'ANY'
getNMDS(
  x,
  FUN = vegdist,
  nmds.fun = nmdsFUN,
  nmdsFUN = c("isoMDS", "monoMDS"),
  ncomponents = 2,
  ntop = 500,
  subset.row = subset_row,
  subset_row = NULL,
  scale = FALSE,
  transposed = FALSE,
  keep.dist = keep_dist,
  keep_dist = FALSE,
  ...
)

## S4 method for signature 'SummarizedExperiment'
getNMDS(
  x,
  ...,
  assay.type = assay_name,
  assay_name = exprs_values,
  exprs_values = "counts",
  FUN = vegdist
)

## S4 method for signature 'SingleCellExperiment'
getNMDS(
  x,
  ...,
  assay.type = assay_name,
  assay_name = exprs_values,
  exprs_values = "counts",
  dimred = NULL,
  ndimred = n_dimred,
  n_dimred = NULL,
  FUN = vegdist
)

calculateNMDS(x, ...)

addNMDS(x, ..., altexp = NULL, name = "NMDS")

runNMDS(x, ...)

Arguments

x	TreeSummarizedExperiment.
...	additional arguments to pass to FUN and nmds.fun.
FUN	Function or Character scalar. A value with a function name returning a dist object
nmds.fun	Character scalar. A value to choose the scaling implementation, either “isoMDS” for MASS::isoMDS or “monoMDS” for vegan::monoMDS
nmdsFUN	Deprecated. Use nmds.fun instead.
ncomponents	Numeric scalar. Indicates the number of DPCoA dimensions to obtain. (Default: 2)
ntop	Numeric scalar. Specifies the number of features with the highest variances to use for dimensionality reduction. Alternatively NULL, if all features should be used. (Default: NULL)
subset.row	Character Vector. Specifies the subset of features to use for dimensionality reduction. This can be a character vector of row names, an integer vector of row indices or a logical vector. (Default: NULL)
subset_row	Deprecated. Use subset.row instead.
scale	Logical scalar. Should the expression values be standardized? (Default: FALSE)
transposed	Logical scalar. Specifies if x is transposed with cells in rows. (Default: FALSE)
keep.dist	Logical scalar. Indicates whether the dist object calculated by FUN should be stored as ‘dist’ attribute of the matrix returned/stored by getNMDS/ addNMDS. (Default: FALSE)
keep_dist	Deprecated. Use keep.dist instead.
assay.type	Character scalar. Specifies which assay to use for calculation. (Default: "counts")
assay_name	Deprecated. Use assay.type instead.
exprs_values	Deprecated. Use assay.type instead.
dimred	Character scalar or integer scalar. Specifies the existing dimensionality reduction results to use.
ndimred	integer vector. Specifies the dimensions to use if dimred is specified.
n_dimred	Deprecated. Use ndimred instead.
altexp	Character scalar or integer scalar. Specifies an alternative experiment containing the input data. (Default: NULL)
name	Character scalar. A name for the column of the colData where results will be stored. (Default: "NMDS")

Details

For addNMDS a SingleCellExperiment

Either MASS::isoMDS or vegan::monoMDS are used internally to compute the NMDS components. If you supply a custom FUN, make sure that the arguments of FUN and nmds.fun do not collide.

Value

For getNMDS, a matrix is returned containing the MDS coordinates for each sample (row) and dimension (column).

See Also

MASS::isoMDS, vegan::monoMDS for NMDS component calculation.

plotMDS, to quickly visualize the results.

Examples

# generate some example data
mat <- matrix(1:60, nrow = 6)
df <- DataFrame(n = c(1:6))
tse <- TreeSummarizedExperiment(assays = list(counts = mat),
                                rowData = df)
#
getNMDS(tse)

#
data(esophagus)
esophagus <- addNMDS(esophagus, FUN = vegan::vegdist, name = "BC")
esophagus <- addNMDS(esophagus, FUN = vegan::vegdist, name = "euclidean",
                     method = "euclidean")
reducedDims(esophagus)

FelixErnst/mia documentation built on Nov. 18, 2024, 5:02 a.m.