methyvim_tmle: Differential Methylation with Classical Target Parameters
In methyvim: Targeted, Robust, and Model-free Differential Methylation Analysis

Computes the Targeted Minimum Loss-Based Estimate of the Average Treatment Effect (ATE) or the Risk Ratio, treating DNA methylation as an outcome (Y) and the indicated variable of interest (which ought to be binarized) as a treatment/exposure (A), using the neighbors of a given CpG site as the adjustment set (W). INTERNAL USE ONLY.

methyvim_tmle(
  target_site,
  methytmle_screened,
  var_of_interest,
  type = c("Beta", "Mval"),
  corr,
  obs_per_covar,
  target_param = c("ate", "rr"),
  g_lib = c("SL.mean", "SL.glm", "SL.glm.interaction"),
  Q_lib = c("SL.mean", "SL.glm", "SL.gam", "SL.earth"),
  cv_folds = 5,
  ...,
  return_ic = FALSE
)

`target_site`	Numeric indicating the column containing the screened CpG site indices that will be looped over in TMLE procedure.
`methytmle_screened`	An object of class `methytmle` with clustered sites based on genomic windows containing the screened CpG site indices.
`var_of_interest`	Numeric indicating the column index of the binarized variable of interest, treated as an exposure.
`type`	Character indicating the particular measure of DNA methylation to be used as the observed data in the estimation procedure, either Beta values or M-values. The data are accessed via `getBeta` or `getM`.
`corr`	Numeric indicating the maximum correlation that a neighboring site can have with the target site.
`obs_per_covar`	Numeric indicating the number of observations needed for for covariate included in W for downstream analysis. This ensures the data is sufficient to control for the covariates.
`target_param`	Character indicating the target causal parameter for which an estimator will be constructed and computed via targeted minimum loss estimation. Currently, this is limited to the Average Treatment Effect (ATE) and the Risk Ratio (RR), with routines from the tmle package being used for the computation.
`g_lib`	Character or vector of characters indicating the algorithms to be implemented in SuperLearner if `tmle_type` is set to "glm".
`Q_lib`	Character or vector of characters indicating the algorithms to be implemented in SuperLearner if `tmle_type` is set to "sl".
`cv_folds`	A `numeric` scalar indicating how many folds to use in performing targeted minimum loss estimation. Cross-validated estimates are more robust, allowing relaxing of theoretical conditions and construction of conservative variance estimates.
`...`	Additional arguments passed to `tmle` in fitting the targeted minimum loss estimator.
`return_ic`	Logical indicating whether an influence curve estimate should be returned for each site that passed through the filter.

An data.frame containing the results of the Targeted Minimum Loss Estimation (TMLE) procedure for the target parameter of interest for a single CpG site, computed via the tmle package.

methyvim documentation built on Nov. 8, 2020, 11:11 p.m.