PermuteMarginalModel: PermuteMarginalModel - This permutation function that will...

In DCAN-Labs/MarginalModelCIFTI: MarginalModelCifti constructs and tests marginal models for surface and volume data.

Description

This function wraps all of the other functions in the MarginalModelCifti package. The output is a text file containing the permutation tests.

Usage

PermuteMarginalModel(
  external_df,
  concfile,
  structtype,
  structfile,
  matlab_path,
  surf_command,
  wave = NULL,
  notation,
  zcor = NULL,
  corstr = NULL,
  family_dist = NULL,
  dist_type = "radenbacher",
  z_thresh = 2.3,
  nboot = 1000,
  p_thresh = 0.5,
  sigtype = "cluster",
  id_subjects = "subid",
  output_directory = "~/",
  fastSwE = TRUE,
  adjustment = NULL,
  norm_external_data = TRUE,
  norm_internal_data = TRUE,
  marginal_outputs = FALSE,
  marginal_matrix = NULL,
  enrichment_path = NULL,
  modules = NULL,
  wb_command = "wb_command",
  subsetfile = NULL,
  output_permfile = "~/permfile.txt"
)

Arguments

external_df	A data frame comprising non-brain measures to model. Can be specified as a string to a csv file with appropriate headers.
concfile	A character string denoting a single column text file that lists the dscalars in the same order as the external_df and wave frames.
structtype	A character string denoting whether the map is volumetric ('volume'), surface-based ('surface'), a pconn ('pconn'), or a niiconn ('niiconn').
structfile	A character string denoting the structural map file, used for cluster detection on surfaces only.
matlab_path	A character string denoting the path to the matlab compiler. Please use v91.
surf_command	A character string denoting the path to the SurfConnectivity command.
wave	A data frame comprising the order of the waves (longitudinal measures) for each independent case.
notation	A character string representing the model formula. Uses Wilkinson notation.
zcor	A matrix defining the correlation structure set by the user, can be used to specify custom dependencies (e.g. site or kinship matrices).
corstr	A character string that defines the correlation structure in a predetermined way. "Independence" is recommend for most cases.
family_dist	A character string denoting the assumed underlying distribution of all input data.
dist_type	A character string denoting the distribution to use for the wild bootstrap.
z_thresh	A numeric that represents the threshold for Z statistics when performing cluster detection.
nboot	A numeric that represents the number of wild bootstraps to perform.
p_thresh	A numeric that represents the p value threshold for assessing significance.
sigtype	A character string denoting cluster ('cluster'), point ('point'), or enrichment ('enrichment') comparisons.
id_subjects	A character string denoting the column for the subject ID. Only needed when FastSwE is FALSE
output_directory	A character string denoting the path to the output permutation tests
adjustment	A character string denoting the small sample size adjustment to use when fastSwE is set to TRUE. Is NULL by default.
norm_external_data	A boolean. If set to true, external data will be normed prior to analysis.
norm_internal_data	A boolean. If set to true, MRI data will be normed per datapoint prior to analysis.
marginal_outputs	A boolean. If set to true, marginal values will be output as statistical maps
marginal_matrix	A numeric matrix depicting how to draw the map. Only needed if marginal_outputs is set to TRUE
enrichment_path	A string depicting the path to the enrichment code. Used for enrichment analysis only (i.e. when sigtype is set to 'enrichment').
modules	A csv file or array that depicts the modules for the enrichment analysis.
wb_command	A character string denoting the path to the wb_command file
subsetfile	A character string denoting the path to a subset selection file, to select only a subset of the participants
fastSWE	A boolean that determine the sandwhich estimator approach. If set to FALSE, will use standard R package geeglm. If set to TRUE will use custom-built estimator using rfast.

Examples

max_cc <- ComputeMM_WB(cifti_map,zscore_map,resid_map,fit_map,type,external_df,
notation,family_dist,structtype,thresh,structfile,matlab_path,surf_command,correctiontype)

DCAN-Labs/MarginalModelCIFTI documentation built on Nov. 30, 2021, 3:40 p.m.