Function reference • RFSurrogates

Random Forest Surrogates

The basic object used in most further analysis functions.

RandomForestSurrogates(): Create a random forest with surrogates.

Surrogate Minimal Depth

Surrogate Minimal Depth (SMD) enables the identification of multiple features from the same metabolites and reveals meaningful biological relations, proving its high potential for the comprehensive utilization of high-dimensional metabolomics data. Reference: 10.3390/metabo12010005

SurrogateMinimalDepth(): Variable selection with Surrogate Minimal Depth (SMD)

MinimalDepth(): Variable selection with Minimal Depth (MD).

MeanAdjustedAgreement(): Investigate variable relations of a specific variable with mean adjusted agreement

Mutual Impact of Features

Mutual forest impact (MFI) is a relation parameter that evaluates the mutual association of the featurs to the outcome and, hence, goes beyond the analysis of correlation coefficients. Mutual impurity reduction (MIR) is an importance measure that combines this relation parameter with the importance of the individual features. Reference: 10.48550/ARXIV.2304.02490

MutualForestImpact(): Mutual Forest Impact (Corrected Mean Adjusted Agreement).

MFI(): Mutual Forest Impact shortcut function (recommended).

MutualImpurityReduction(): Mutual Impurity Reduction (MIR)

MutualForestImpactVariableSelection(): Variable selection for MutualForestImpact.

MutualImpurityReductionVariableSelection(): Variable selection for Mutual Impurity Reduction.

Data

Example data sets shipped with this package.

SMD_example_data: Example data set for the package SurrogateMinimalDepth

Version 0.3.x functions

The original functions from prior versions remain available for backward compatability.

var.relations(): Investigate variable relations of a specific variable with mean adjusted agreement

var.relations.mfi(): Investigate variable relations of a specific variable with mutual forest impact (corrected mean adjusted agreement).

var.select.md(): Variable selection with Minimal Depth (MD)

var.select.smd(): Variable selection with Surrogate Minimal Depth (SMD) (MAIN FUNCTION)

var.select.mir(): Variable selection with mutual impurity reduction (MIR).

Other functions

Additional functions published as part of version 0.3.x and earlier versions. These will remain available for backward compatability.

addLayer(): Add layer information to a forest that was created by getTreeranger

addSurrogates(): Add surrogate information to a tree list.

build.clusters(): Apply cluster analysis to build variable groups

count.surrogates(): Count surrogate variables

getTreeranger(): Get a list of structured trees from a ranger object.

meanAdjAgree(): Calculate mean adjusted agreement to investigate variables relations

mindep(): Execute minimal depth variable importance

reduce.surrogates(): Reduce surrogate variables in a random forest.

surrmindep(): Execute surrogate minimal depth variable importance