fitnessForOneComponent

Compute best pattern-based fitness for a single Simplivariate Component

Identifies constant, additive, multiplicative, and user-defined
simplivariate components in numeric data matrices using a genetic algorithm.
Supports flexible pattern definitions and provides visualization for
general biclustering applications across diverse domains. The method builds
on simplivariate models as introduced in Hageman et al. (2008)
<doi:10.1371/journal.pone.0003259> and is related to biclustering
frameworks as reviewed by Madeira and Oliveira (2004) <doi:10.1109/TCBB.2004.2>.

Jos Hageman

SIMPLICA

Biclustering via Simplivariate Component Analysis

fitnessForOneComponent function

<dl><dt>mat</dt>
<dd>A numeric matrix (the component)</dd>
<dt>dfMean</dt>
<dd>Overall mean of the full data matrix</dd>
<dt>patternFunctions</dt>
<dd>Named list of functions for structure types</dd>
<dt>penalty</dt>
<dd>Named numeric vector of penalties per pattern type</dd></dl>

Arguments

Compute best pattern-based fitness for a single Simplivariate Component — fitnessForOneComponent

<dl>

<dt>mat</dt>
<dd>A numeric matrix (the component)</dd>


<dt>dfMean</dt>
<dd>Overall mean of the full data matrix</dd>


<dt>patternFunctions</dt>
<dd>Named list of functions for structure types</dd>


<dt>penalty</dt>
<dd>Named numeric vector of penalties per pattern type</dd>

</dl>

fitnessForOneComponent: Compute best pattern-based fitness for a single Simplivariate Component

Description

Usage

Value

Arguments

Examples