learn the gene regulatory bayesian network from the gene expression data

bnlearn {bnlearn}

R Documentation

learn the gene regulatory bayesian network from the gene expression data

Description

Usage

bnlearn(exprData,
    priorNet = NULL,
    max.itrs = 500,
    strict = NULL);

Arguments

exprData

the gene expression matrix object, could be load from csv file via geneExpression::load.expr. [as Matrix]

priorNet

a collection of the prior knowledge regulatory edge data (RegulatoryEdge), which is used as the whitelist of the network structure learning: only the regulation relation that is described in this prior network will be considered in the structure learning.

max.itrs

[as integer]

strict

the strict option of the in silico perturbation experiment: if this parameter is TRUE, then an error will be thrown when the target gene of the perturbation is missing from the learned network; if this parameter is FALSE, then a warning message will be printed and the wildtype expression data will be returned as the perturbation result with the Undefined flag marked as TRUE. . [as boolean]

env

[as Environment]




        Details

        this function runs the network learning in two steps:



the structure learning: the MMHC algorithm with the whitelist prior network;
the parameter learning: the maximum likelihood estimation(MLE) of the Gaussian bayesian network.


        Authors

        biosystem

        Value

        a BNLearnWorkflow object that contains the learned bayesian network model, which implements the InsilicoPerturbationExperiment interface, so that it can be used by the knockouts, overexpress and knockdown api for run the in silico gene perturbation experiment;

this function returns a R# error message object if the given prior network data can not be cast to a collection of the RegulatoryEdge data.
clr value class
BNLearnWorkflow

        Examples

        

        
        [Package bnlearn version 1.0.0.0 Index]