An R6 Class to represent a ZIPLNfit in a standard, general framework, with sparse inverse residual covariance

Super class

ZIPLNfit -> ZIPLNfit_sparse

Active bindings

penalty: the global level of sparsity in the current model
penalty_weights: a matrix of weights controlling the amount of penalty element-wise.
n_edges: number of edges if the network (non null coefficient of the sparse precision matrix)
nb_param_pln: number of parameters in the PLN part of the current model
vcov_model: character: the model used for the residual covariance
pen_loglik: variational lower bound of the l1-penalized loglikelihood
ebic_gamma: the tuning parameter gamma of the EBIC, between 0 and 1. Zero gives back the BIC; the default 0.5 is the value recommended by Foygel and Drton (2010). Assign to it to change the EBIC of this fit.
EBIC: variational lower bound of the EBIC of Foygel and Drton (2010), that is the BIC with the additional penalty 2 gamma |E| log(p) on the edge set
density: proportion of non-null edges in the network
criteria: a vector with loglik, penalized loglik, BIC, EBIC, ICL, R_squared, number of parameters, number of edges and graph density

Methods

Inherited methods

`ZIPLNfit_sparse$new()`

Initialize a ZIPLNfit_fixed model

Usage

ZIPLNfit_sparse$new(data, control)

Arguments

data: a named list used internally to carry the data matrices
control: a list for controlling the optimization. See details.

`ZIPLNfit_sparse$optimize()`

Call to the optimizer and update of the relevant fields. Reports non-convergence of the graphical Lasso on top of ZIPLNfit's method.

Usage

ZIPLNfit_sparse$optimize(data, control)

Arguments

data: a named list used internally to carry the data matrices
control: a list for controlling the optimization. See details.

`ZIPLNfit_sparse$latent_network()`

Extract interaction network in the latent space

Usage

ZIPLNfit_sparse$latent_network(type = c("partial_cor", "support", "precision"))

Arguments

type: edge value in the network. Can be "support" (binary edges), "precision" (coefficient of the precision matrix) or "partial_cor" (partial correlation between species)

Returns

a square matrix of size ZIPLNfit_sparse$n

`ZIPLNfit_sparse$plot_network()`

plot the latent network.

Usage

ZIPLNfit_sparse$plot_network(
  type = c("partial_cor", "support"),
  output = c("igraph", "corrplot"),
  edge.color = c("#F8766D", "#00BFC4"),
  remove.isolated = FALSE,
  node.labels = NULL,
  layout = layout_in_circle,
  edge.alpha = 0.2,
  plot = TRUE
)

Arguments

type: edge value in the network. Either "precision" (coefficient of the precision matrix) or "partial_cor" (partial correlation between species).
output: Output type. Either igraph (for the network) or corrplot (for the adjacency matrix)
edge.color: Length 2 color vector. Color for positive/negative edges. Default is c("#F8766D", "#00BFC4"). Only relevant for igraph output.
remove.isolated: if TRUE, isolated node are remove before plotting. Only relevant for igraph output.
node.labels: vector of character. The labels of the nodes. The default will use the column names ot the response matrix.
layout: an optional igraph layout. Only relevant for igraph output.
edge.alpha: opacity of the weakest edge, the strongest one being fully opaque, so that the strength of an edge can be read off a dense network. Default is 0.2. Set it to 1 for uniformly opaque edges. Only relevant for igraph output with type = "partial_cor".
plot: logical. Should the final network be displayed or only sent back to the user. Default is TRUE.

`ZIPLNfit_sparse$clone()`

The objects of this class are cloneable with this method.

Usage

ZIPLNfit_sparse$clone(deep = FALSE)

Arguments

deep: Whether to make a deep clone.

Examples

if (FALSE) { # \dontrun{
# See other examples in function ZIPLN
data(trichoptera)
trichoptera <- prepare_data(trichoptera$Abundance, trichoptera$Covariate)
myPLN <- ZIPLN(Abundance ~ 1, data = trichoptera, control=  ZIPLN_param(penalty = 1))
class(myPLN)
print(myPLN)
plot(myPLN)
} # }