get_pcor: Extract a 'coglasso' partial correlation matrix

In coglasso: Collaborative Graphical Lasso - Multi-Omics Network Reconstruction

Extract a coglasso partial correlation matrix

Description

get_pcor() extracts the selected partial correlation matrix from a select_coglasso object, or a different specific one from either a select_coglasso or a coglasso object when specifying the optional parameters.

Usage

get_pcor(sel_cg_obj, index_c = NULL, index_lw = NULL, index_lb = NULL)

Arguments

sel_cg_obj	The object of S3 class select_coglasso or of S3 class coglasso.
index_c	The index of the c value different from the one selected by model selection. To set only if the desired partial correlation matrix is not the selected one.
index_lw	The index of the \lambda_w value of the chosen non-optimal partial correlation matrix. To set only if the desired partial correlation matrix is not the selected one.
index_lb	The index of the \lambda_b value of the chosen non-optimal partial correlation matrix. To set only if the desired partial correlation matrix is not the selected one.

Details

If the input is a coglasso object, it is necessary to specify all the indexes to extract the chosen partial correlation matrix.
If the input is a select_coglasso object, it extracts by default the selected partial correlation matrix. If the selection method was "ebic", and you want to extract a different partial correlation matrix than the selected one, specify all indexes. Otherwise, if the objective is to extract the optimal partial correlation matrix for a specific c value different than the selected one, set index_c to your chosen one. Also here it is possible to extract a specific non-optimal partial correlation matrix by setting all the indexes to the chosen ones.

Value

get_pcor() returns the selected partial correlation matrix.

Examples

sel_cg <- bs(multi_omics_sd_micro, p = c(4, 2), nlambda_w = 3, nlambda_b = 3,
                 nc = 3, verbose = FALSE)
sel_pcor <- get_pcor(sel_cg)
print(sel_pcor)

coglasso documentation built on July 30, 2026, 5:11 p.m.