glmutli_pq {MiscMetabar}

Automated model selection and multimodel inference with (G)LMs for phyloseq

Description

Usage

glmutli_pq(
  physeq,
  formula,
  fitfunction = "lm",
  hill_scales = c(0, 1, 2),
  aic_step = 2,
  confsetsize = 100,
  plotty = FALSE,
  level = 1,
  method = "h",
  crit = "aicc",
  ...
)

Arguments

physeq

(required): a phyloseq-class object obtained using the phyloseq package.

formula

(required) a formula for glmulti::glmulti() Variables must be present in the physeq@sam_data slot or be one of hill number defined in hill_scales or the variable Abundance which refer to the number of sequences per sample.

fitfunction

(default "lm")

hill_scales

(a vector of integer) The list of q values to compute the hill number H^q. If Null, no hill number are computed. Default value compute the Hill number 0 (Species richness), the Hill number 1 (exponential of Shannon Index) and the Hill number 2 (inverse of Simpson Index).

aic_step

The value between AIC scores to cut for.

confsetsize

The number of models to be looked for, i.e. the size of the returned confidence set.

plotty

(logical) Whether to plot the progress of the IC profile when running.

level

If 1, only main effects (terms of order 1) are used to build the candidate set. If 2, pairwise interactions are also used (higher order interactions are currently ignored)

method

The method to be used to explore the candidate set of models. If "h" (default) an exhaustive screening is undertaken. If "g" the genetic algorithm is employed (recommended for large candidate sets). If "l", a very fast exhaustive branch-and-bound algorithm is used. Package leaps must then be loaded, and this can only be applied to linear models with covariates and no interactions. If "d", a simple summary of the candidate set is printed, including the number of candidate models.

crit

The Information Criterion to be used. Default is the small-sample corrected AIC (aicc). This should be a function that accepts a fitted model as first argument. Other provided functions are the classic AIC, the Bayes IC (bic), and QAIC/QAICc (qaic and qaicc).

...

Others arguments passed on to glmulti::glmulti() function

Details

This function is mainly a wrapper of the work of others. Please make a reference to glmulti::glmulti() if you use this function.

Value

A data.frame summarizing the glmulti results with columns

-estimates -unconditional_interval -nb_model" -importance -alpha

See Also

glmulti::glmulti()

Examples

if (requireNamespace("glmulti")) {
  res_glmulti <-
    glmutli_pq(data_fungi, "Hill_0 ~ Hill_1 + Abundance + Time + Height", level = 1)
  res_glmulti
  res_glmulti_interaction <-
    glmutli_pq(data_fungi, "Hill_0 ~ Abundance + Time + Height", level = 2)
  res_glmulti
}