{FuzzyGene} R# Documentation

FuzzyGene


require(GCModeller);

#' FuzzyGene: the neuro-fuzzy (Wilson/Takagi-Sugeno ANFIS) gene regulatory network toolkit.
imports "FuzzyGene" from "biosystem";

FuzzyGene: the neuro-fuzzy (Wilson/Takagi-Sugeno ANFIS) gene regulatory network toolkit.

This R# package module wraps the FuzzyGene sub-system. Every target gene is described by its own first-order Takagi-Sugeno fuzzy model whose inputs are 3-6 candidate regulators; the rules of that model can be read as biological statements ("IF TF-A is High AND TF-B is Low THEN gene-G is activated"), and the analytic partial derivatives of the model give a signed, weighted regulatory edge table.

The workflow exposed by this module is:

(FuzzyGeneConfig, registered as the fuzzygene_config R# type);

prior regulatory network;

when the matrix carries multiple time points);

signed regulatory edge table;

A trained FuzzyGeneModel implements the InsilicoPerturbationExperiment interface, so the knockouts, knockdown and overexpress api of the bnlearn package can be applied to it directly for in silico perturbation experiments (partial knockdown is natively supported because the fuzzy membership functions describe graded regulation). The generic writeBin handler of this module persists a model into a zip based binary stream, and readBin.fuzzygene restores it back to a fully usable model object.

.NET clr type export
fuzzygene: FuzzyGeneModel

神经模糊(ANFIS/TSK)基因调控网络模型的门面。

fuzzygene_config: FuzzyGeneConfig

神经模糊基因调控网络(ANFIS/TSK)的建模配置。



.NET clr function exports
fuzzygene_config

Create a FuzzyGene modelling configuration object.

build_fuzzygene

Build an untrained neuro-fuzzy gene regulatory network model.

train

Train the neuro-fuzzy model: one Takagi-Sugeno network per target gene.

fuzzygene_rules

Export the IF-THEN rules of every fitted target gene.

fuzzygene_edges

Export the signed regulatory edge table of a trained model.

fuzzygene_regulation_matrix

Export the signed regulation weight matrix of a trained model.

fuzzygene_confidence_matrix

Export the confidence matrix of a trained model (same shape as the weight matrix).

fuzzygene_context

Inject a transcriptome context as the initial condition of the virtual perturbation.

fuzzygene_reset_context

Drop the injected context and restore the training baseline.

fuzzygene_model_config

The configuration attached to a model (identical to config.json inside the package).

fuzzygene_context_upstream

Inject the result of an upstream perturbation as the control baseline (iterative perturbation).

fuzzygene_report

Human readable training report of a fitted model.

save_fuzzygene

Save a trained model into a *.fuzzygene.zip package.

load_fuzzygene

Load a model from a *.fuzzygene.zip package.


[Document Index]