sparse_hclust: Run a Sparse Hierarchical Clustering

In korkutlab/reflect: REcurrent Features LEveraged for Combination Therapy

Description

Given a matrix of expression/alteration and the wbound value for L1 regularization, it performs a sparse hierarchical clustering.

Usage

sparse_hclust(
  mat_value,
  wbound,
  dissimilarity = c("squared.distance", "absolute.value")
)

Arguments

mat_value	A matrix of expression/alteration with samples as rows and features as columns.
wbound	A real number as the wbound used in sparse hierarchial clustering.
dissimilarity	A string for the type of dissimilarity, either "squared.distance" or "absolute.value". Default "squared.distance".

Value

hc	An object of class hclust which describes the tree produced by the clustering process. See detail in function hclust from stats.
weight	The weights of features used in sparse hierachical clustering.
features_nonzero	Features that have nonzero weights.
mat_value_clustered	A matrix of values that are clustered by the sparse hierarchical clustering.

Examples

library(reflect)
mat_value <- egfr_data$mat_value
wbound <- 2.0
shc <- sparse_hclust(mat_value, wbound)

korkutlab/reflect documentation built on July 5, 2021, 7:38 a.m.