A thin wrapper that implements the main data-adaptive multiple hypothesis testing strategy for data structures commonly found in computational biology experiments, using the popular SummarizedExperiment container class.

bioadaptest(data_in, var_int, cntrl_set = NULL, n_top = 25,
  n_fold = 10, parameter_wrapper = rank_DE,
  learning_library = c("SL.mean", "SL.glm"), absolute = FALSE,
  negative = FALSE, p_cutoff = 0.05, q_cutoff = 0.05)

Arguments

data_in

An object of class SummarizedExperiment, a common container class for computational biology and bioinformatics. This object is used to construct the output object of class adaptmle.

var_int

A numeric vector of binary treatment assignment whose effect on the biological units is to be assessed. The data-adpative target parameter approach finds any biological sites strongly impacted by this quantity across the observed experimental units (subjects).

cntrl_set

A matrix of discrete variables representing baseline covariates that are controlled for in the estimation of the data-adaptive target parameter via targeted maximum likelihood estimation. If NULL, an identity vector is generated internally.

n_top

(integer vector) - value for the number of candidate covariates to generate using the data-adaptive estimation algorithm.

n_fold

(integer vector) - number of cross-validation folds.

parameter_wrapper

(function) - user-defined function that takes input (Y, A, W, absolute, negative) and outputs a (integer vector) containing ranks of biomarkers (outcome variables). For detail, please refer to the documentation for rank_DE.

learning_library

(character vector) - library of learning algorithms to be used in fitting the "Q" and "g" step of the standard TMLE procedure.

absolute

(logical) - whether or not to test for absolute effect size. If FALSE, test for directional effect. This overrides argument negative.

negative

(logical) - whether or not to test for negative effect size. If FALSE = test for positive effect size. This is effective only when absolute = FALSE.

p_cutoff

The minimum p-value required to evaluate a given biological unit (e.g., gene) as statistically significant.

q_cutoff

The minimum p-value required to evaluate a given biological unit (e.g., gene) as statistically significant after applying a correction for multiple hypothesis testing.

Value

An object of class adaptmle, sub-classed from the popular container class SummarizedExperiment, containing information about the experiment being analyzed as well as results from applying the TMLE for the data-adaptive target parameter as produced by adpatest.

Examples

library(SummarizedExperiment) library(airway) set.seed(5678) data(airway) genes_sub <- order(sample(seq_len(100))) air_reduced <- airway[genes_sub, ] simple_air <- cbind(air_reduced, air_reduced) dex_var = as.numeric(as.matrix(colData(simple_air))[, 3] - 1) airway_out <- bioadaptest(data_in = simple_air, var_int = dex_var, cntrl_set = NULL, n_top = 5, n_fold = 2, parameter_wrapper = rank_DE)