k.means.spatial: Clusters spatial omics data with spatial regularization

In jeliason/spatialkmeans: Cluster Spatial Omics Data Via Spatially Regularized K-Means

Description

Clusters spatial omics data with spatial regularization

Usage

k.means.spatial(
  X,
  cell.dists,
  nclust = 3,
  lambda = 0.1,
  metric = matrixStats::colMins,
  max.iter = 100
)

Arguments

X	An n x p matrix with n cells and p features
cell.dists	An n x n matrix with pairwise distances between all cells
nclust	A scalar indicating the number of clusters
lambda	The weight that spatial penalty takes
metric	A column-wise quantile function to determine distance to cluster (eg. colMins, colMedians, colMaxs from matrixStats)
max.iter	Maximum number of iterations before stopping clustering

Value

a list containing mus (the centroid of each cluster), est (cluster membership for each cell), num_it (number of iterations taken to converge)

jeliason/spatialkmeans documentation built on Dec. 20, 2021, 11 p.m.