This functions create the signature centroids estimated from the GalgoR output and the expression matrix of the training sets.

create_centroids(
  output,
  solution.names,
  train.set,
  distancetype = "pearson",
  usegpu = FALSE
)

Arguments

output

@param output An object of class galgo.Obj

solution.names

A character vector with the names of the solutions for which the centroids are to be calculated

train.set

a matrix or data.frame. Must be an expression matrix with features in rows and samples in columns

distancetype

a character that can be either 'pearson', 'uncentered', 'spearman' or 'euclidean'

usegpu

logical TRUE or FALSE

Value

Returns a list with the centroid matrix for each of the solutions in solution.names, where each column represents the prototypic centroid of a subtype and each row the constituents features of the solution signature

Examples

if (FALSE) { # Load data rna_luad <- use_rna_luad() TCGA_expr <- rna_luad$TCGA$expression_matrix TCGA_clinic <- rna_luad$TCGA$pheno_data OS <- survival::Surv(time = TCGA_clinic$time, event = TCGA_clinic$status) # Run galgo output <- galgoR::galgo(generations = 10, population = 30, prob_matrix = TCGA_expr, OS = OS) RESULTS <- non_dominated_summary( output = output, OS = OS, prob_matrix = TCGA_expr, distancetype = "pearson", usegpu = FALSE ) CentroidsList <- create_centroids(output, RESULTS$solution, train.set = TCGA_expr) }