Run PHATE (Potential of Heat-diffusion for Affinity-based Trajectory Embedding)

Usage

RunPHATE(object, ...)

# S3 method for Seurat
RunPHATE(
  object,
  reduction = "pca",
  dims = NULL,
  features = NULL,
  assay = NULL,
  slot = "data",
  n_components = 2,
  knn = 5,
  decay = 40,
  n_landmark = 2000,
  t = "auto",
  gamma = 1,
  n_pca = 100,
  knn_dist = "euclidean",
  knn_max = NULL,
  t_max = 100,
  do_cluster = FALSE,
  n_clusters = "auto",
  max_clusters = 100,
  reduction.name = "phate",
  reduction.key = "PHATE_",
  verbose = TRUE,
  seed.use = 11L,
  ...
)

# S3 method for default
RunPHATE(
  object,
  assay = NULL,
  n_components = 2,
  knn = 5,
  decay = 40,
  n_landmark = 2000,
  t = "auto",
  gamma = 1,
  n_pca = 100,
  knn_dist = "euclidean",
  knn_max = NULL,
  t_max = 100,
  do_cluster = FALSE,
  n_clusters = "auto",
  max_clusters = 100,
  reduction.key = "PHATE_",
  verbose = TRUE,
  seed.use = 11L,
  ...
)

Arguments

object: An object. This can be a Seurat object or a matrix-like object.
...: Additional arguments to be passed to the phate.PHATE function.
reduction: A character string specifying the reduction to be used. Default is "pca".
dims: An integer vector specifying the dimensions to be used. Default is NULL.
features: A character vector specifying the features to be used. Default is NULL.
assay: A character string specifying the assay to be used. Default is NULL.
slot: A character string specifying the slot name to be used. Default is "data".
n_components: An integer specifying the number of PHATE components. Default is 2.
knn: An integer specifying the number of nearest neighbors on which to build kernel. Default is 5.
decay: An integer specifying the sets decay rate of kernel tails. Default is 40.
n_landmark: An integer specifying the number of landmarks to use in fast PHATE. Default is 2000.
t: A character string specifying the power to which the diffusion operator is powered. This sets the level of diffusion. If ‘auto’, t is selected according to the knee point in the Von Neumann Entropy of the diffusion operator. Default is "auto".
gamma: A numeric value specifying the informational distance constant between -1 and 1. gamma=1 gives the PHATE log potential, gamma=0 gives a square root potential. Default is 1.
n_pca: An integer specifying the number of principal components to use for calculating neighborhoods. For extremely large datasets, using n_pca < 20 allows neighborhoods to be calculated in roughly log(n_samples) time. Default is 100.
knn_dist: A character string specifying the distance metric for k-nearest neighbors. Recommended values: "euclidean, "cosine, "precomputed". Default is "euclidean".
knn_max: An integer specifying the maximum number of neighbors for which alpha decaying kernel is computed for each point. For very large datasets, setting knn_max to a small multiple of knn can speed up computation significantly. Default is NULL.
t_max: An integer specifying the maximum t to test. Default is 100.
do_cluster: A logical value indicating whether to perform clustering on the PHATE embeddings. Default is FALSE.
n_clusters: An integer specifying the number of clusters to be identified. Default is "auto".
max_clusters: An integer specifying the maximum number of clusters to test. Default is 100.
reduction.name: A character string specifying the name of the reduction to be stored in the Seurat object. Default is "phate".
reduction.key: A character string specifying the prefix for the column names of the PHATE embeddings. Default is "PHATE_".
verbose: A logical value indicating whether to print verbose output. Default is TRUE.
seed.use: An integer specifying the random seed to be used. Default is 11.

Examples

pancreas_sub <- Seurat::FindVariableFeatures(pancreas_sub)
pancreas_sub <- RunPHATE(object = pancreas_sub, features = Seurat::VariableFeatures(pancreas_sub))
CellDimPlot(pancreas_sub, group.by = "CellType", reduction = "phate")