MGLMdtmTune: Finds the tuning parameter value that yields the smallest BIC...

In liudoubletian/phyloMDA: An R package for phylogeny-aware microbiome data analysis

Finds the tuning parameter value that yields the smallest BIC for the Dirichlet-tree multinomial (DTM) sparse regression.

Description

Finds the tuning parameter value that yields the smallest BIC for the Dirichlet-tree multinomial (DTM) sparse regression.

Usage

MGLMdtmTune(otu.tab, X, tree, penalty = "sweep", ngridpt = 20)

Arguments

otu.tab	a data frame or matrix containing the count data. Rows of the matrix represent observations and columns are the taxa.
X	a data frame or matrix containing the covariates. Rows of the matrix represent observations.
tree	the phylogenetic tree.
penalty	penalty type for the regularization term. Can be chosen from 'sweep', 'group', or 'nuclear'. See Details for the description of each penalty type of MGLM.
ngridpt	an optional numeric variable specifying the number of grid points to tune. see MGLMtune.

Value

Returns a list of MGLM objects ('MGLMtune') corresponding to internal nodes of the tree.

References

Wang, T., Zhao, H.: A Dirichlet-tree multinomial regression model for associating dietary nutrients with gut microorganisms. Biometrics 73(3), 792–801 (2017).

Examples

library(phyloseq)
library(magrittr)
otu.tab <- t(otu_table(combo.phyloseq.obj))
metadata <- sample_data(combo.phyloseq.obj)
tree <- phy_tree(combo.phyloseq.obj)
X <- as.matrix(metadata)[,18:37] %>% apply(., 2, as.numeric)
# fit <- MGLMdtmTune(otu.tab, X, tree, penalty = "sweep")
# Extract_logL(fit)

liudoubletian/phyloMDA documentation built on April 28, 2022, 6:51 p.m.