rpart.tree: Recursive partitioning tree

In jgarces02/FlowCT: A semi-automated workflow for deconvolution of immunophenotypic data and objective reporting on large datasets

Recursive partitioning tree

Description

This function performs a recursive partitioning tree using the percentage (or raw counts) of identified cell populations. Depending if time.var is present or absent, final tree were a cassification or regression tree, respectively.

Usage

rpart.tree(
  fcs.SCE,
  cell.clusters,
  variables,
  value = "percentage",
  time.var,
  event.var,
  xerror,
  rpart.control,
  return.data = T
)

Arguments

fcs.SCE	A fcs.SCE object generated through FlowCT::fcs.SCE().
cell.clusters	Name of column containing clusters identified through FlowCT::clustering.flow().
variables	Vector with variables for calculating the cutoff. If nothing is detailed (NULL, default), all immune populations from cell.clusters will be considered.
value	String specifying if final resuls should be proportions ("percentage", default) or raw counts ("counts").
time.var	Survival time variable.
event.var	Variable with event censoring.
xerror	Numeric value with cross-validation error for pruning the tree. If missing, the minimal error will be automatically selected.
rpart.control	List with parameters for controling rpart. See ?rpart::rpart.control for more information and defaults values (caution because, in FlowCT there are some changes: cp = 0 and minsplit = 10).
return.data	Original, pruned and used metadata should be returned?. Default = TRUE.

Examples

## Not run: 
tr <- rpart.tree(fcs.SCE = fcs, cell.clusters = "clusters_named", 
   time.var = "PFS", event.var = "PFS_c")

## End(Not run)

jgarces02/FlowCT documentation built on March 28, 2023, 12:42 p.m.