The relevance map is cached insided of the DiffusionMap.

gene_relevance(coords, exprs, ..., k = 20L, dims = 1:2,
  distance = NULL, smooth = TRUE, verbose = FALSE)

# S4 method for DiffusionMap,missing
gene_relevance(coords, exprs, ...,
  k = 20L, dims = 1:2, distance = NULL, smooth = TRUE,
  verbose = FALSE)

# S4 method for matrix,matrix
gene_relevance(coords, exprs, ..., k = 20L,
  dims = 1:2, distance = NULL, smooth = TRUE, verbose = FALSE)

Arguments

coords

A DiffusionMap object or a cells \(\times\) dims matrix.

exprs

An cells \(\times\) genes matrix. Only provide if coords is no DiffusionMap.

...

If no DiffusionMap is provided, a vector of weights (of the same length as dims) can be provided.

k

Number of nearest neighbors to use

dims

Index into columns of coord

distance

Distance measure to use for the nearest neighbor search.

smooth

Smoothing parameters c(window, alpha) (see smth.gaussian). Alternatively TRUE to use the smoother defaults or FALSE to skip smoothing,

verbose

If TRUE, log additional info to the console

Value

A GeneRelevance object:

Slots

coords

A cells \(\times\) dims matrix or sparseMatrix of coordinates (e.g. diffusion components), reduced to the dimensions passed as dims

exprs

A cells \(\times\) genes matrix of expressions

partials

Array of partial derivatives wrt to considered dimensions in reduced space (genes \(\times\) cells \(\times\) dimensions)

partials_norm

Matrix with norm of aforementioned derivatives. (n\_genes \(\times\) cells)

nn_index

Matrix of k nearest neighbor indices. (cells \(\times\) k)

dims

Column index for plotted dimensions. Can character, numeric or logical

distance

Distance measure used in the nearest neighbor search. See find_knn

smooth_window

Smoothing window used (see smth.gaussian)

smooth_alpha

Smoothing kernel width used (see smth.gaussian)

See also

Gene Relevance methods, Gene Relevance plotting: plot_differential_map/plot_gene_relevance

Examples

data(guo_norm) dm <- DiffusionMap(guo_norm) gr <- gene_relevance(dm) m <- t(Biobase::exprs(guo_norm)) gr_pca <- gene_relevance(prcomp(m)$x, m) # now plot them!