ppt.tree: Reconstruction of the tree

In hms-dbmi/crestree: Analysis of branching branching trajectories from single-cell data

Description

Using SimplePPT approach to model principal tree (pptree) of the data

Usage

ppt.tree(X, W = NA, emb = NA, M, init = NULL, plot = TRUE,
  output = TRUE, lambda = 10, sigma = 0.1, seed = NULL,
  n.steps = 50, err.cut = 0.05, metrics = "cosine", p.power = 1,
  knn = NULL, ...)

Arguments

X	gene (row) vs cell (columns) expression matrix
emb	embdedding to visalize cells and principal tree together
M	number of principal points to use (more than zero, no more than number of cells)
init	matrix of initial gene coordinates of principal points
plot	plot or not intermediate trees
lambda	penalty for the tree length, as used in SimplePPT
sigma	parameter as used in SimplePPT
seed	used to make initial assignment of principal points to a subset of cells
n.steps	number of iteraions
err.cut	stop algorithm if proximity of principal points between iterations less than err.cut
metrics	metrics used to calculated distances between cells or principal points. "euclidean" or "cosine"
p.power	if cosine metrics used, option p.power allows to use (1-cor)^p.power (p.power=1 by default)

Value

pptree object

hms-dbmi/crestree documentation built on July 11, 2019, 8:04 p.m.