getPatterns: Discover lineage dependent temporal patterns

In minzheguo/SLICE: Determining Cell Differentiation and Lineage based on Single Cell Entropy

Discover lineage dependent temporal patterns

Description

Given the M1 predicted expression profiles of differentially expressed genes in a transition, the function uses PAM clustering to divide differentially expressed genes into clusters (patterns)

Usage

getPatterns(
  expr,
  do.zscore = T,
  c.method = "pam",
  k = NULL,
  d.method = function(x) as.dist((1 - cor(t(x)))/2),
  do.plot = T,
  plot.filename = NULL
)

Arguments

expr	The M1 predicted expression profiles of differentially expressed genes in a transition
do.zscore	If TRUE, perform gene based zscore transformation prior to clustering
c.method	The algorithm to find pattern clusters, currently support "pam" and "kmeans"
k	The number of patterns to be discovered; if NULL, gap statistic (cluster::clusGap) will be used to determine an optimal k; parameters of cluster::clusGap are set as follows: k.max=10, B=50
d.method	The distance method
do.plot	Whether to plot discovered patterns
plot.filename	The file name of the plot; if NULL, plot to the screen

Value

An object of class "pam" or "kmeans" representing the clustering results.

minzheguo/SLICE documentation built on June 5, 2023, 11:56 a.m.