Get pairwise correlations from PLS run — diablo_get_pairwise_pls

Computes the correlation matrix between datasets based on the first component of a PLS run.

Usage

diablo_get_pairwise_pls_corr(pairwise_pls_result)

Arguments

pairwise_pls_result: List containing the results from the pairwise PLS runs, computed by the run_pairwise_pls function.

Value

A matrix of correlation coefficients between datasets.

Details

The correlation coefficient between two datasets is computed as the correlation coefficient between the first component of each dataset obtained from a Projection to Latent Structure (PLS) run, from the mixOmics package.

Examples

if (FALSE) { # \dontrun{
## get mixomics input from MultiDataSet object
mixomics_data <- get_input_mixomics_supervised(mo_set, "outcome_group")

## Get the list of dataset names
datasets <- setdiff(names(mixomics_data), "Y")

## Get all possible pairwise combinations of dataset names
ds_pairs <- utils::combn(datasets, 2)

## run PLS for each pair of datasets
pls_res_list <- lapply(1:ncol(ds_pairs), function(i) {
  run_pairwise_pls(mo_set, ds_pairs[, i])
})

## extract the pairwise correlation matrix
diablo_get_pairwise_pls_corr(pls_res_list)
} # }