BoRT: Bootstrap-based Test Regression Tree

In hk1785/MiVT: Microbiome Virtual Twins

Bootstrap-based Test Regression Tree

Description

This function implements BoRT.

Usage

BoRT(Z, tax.prop, tax.rank = c("Phylum", "Class", "Order", "Family", "Genus", "Species"), minsplit = 10, minbucket = 5, cp = 0.01, n.boot = 20000)

Arguments

Z	Predicted treatment effects using dML (see dML()).
tax.prop	A list of tables for the proportions of microbial taxa on each taxonomic rank (Phylum, Class, Order, Family, Genus, Species) (see biom.clean()).
tax.rank	A taxonomic rank to be used for the subgroup identification (tax.rank = c("Phylum", "Class", "Order", "Family", "Genus", "Species")).
minsplit	The minimum number of observations that must exist in a node in order for a split to be attempted (Default = 10).
minbucket	The minimum number of observations in any terminal node (Default = 5).
cp	The complexity parameter of the decision tree (Default = 0.01).
n.boot	The number of bootstrap samples (Default = 20000).

Value

$Sel.Taxa: Short taxonomic IDs and full taxonomic names. $BoRT.out: The output table of BoRT. Columns are the identified subgroups that correspond with the terminal nodes from left to right. N is the sample size for each subgroup. Overall TE represents the overall treatment effect, and Subgroup TE represents the subgroup treatment effect.

Author(s)

Hyunwook Koh

References

Koh, H. Subgroup identification using virtual twins for human microbiome studies. (Under review).

Foster, J. C., Taylor, J. M. & Ruberg, S. J. Subgroup identification from randomized clinical trial data. Stat. Med. 30(24), 2867-2880 (2011).

Examples

data(fit)
data(tree)
data(tax.tab)

prop <- fit$pi
disp <- fit$theta

sim.biom <- gen.syn.dat(tree = tree, tax.tab = tax.tab, prop = prop, disp = disp)
sim.biom

qc.out <- biom.qc(biom = sim.biom)

dml.out <- dML(y = qc.out$sam.dat$y, Tr = qc.out$sam.dat$Tr, X = qc.out$otu.tab, tree = qc.out$tree)

bort.out <- BoRT(Z = dml.out$Z, tax.prop = qc.out$tax.prop, tax.rank = "Genus")
bort.out

hk1785/MiVT documentation built on Dec. 21, 2024, 1:08 p.m.