GeneNet wrapper function

Share:

Description

Default wrapper function for the GeneNet network inference algorithm

Usage

1

Arguments

data

Numeric matrix with the microarray dataset to infer the network. Columns contain variables and rows contain samples.

Details

GeneNEt uses an heuristic for learning statistically a causal network. It relies on a conversion of a network inferred through correlation into a partial correlation graph. Then, a partial ordering of the nodes is assigned by means of a multiple testing of the log-ratio of standardized partial variances. This allows identifying a directed acyclic causal network as a sub-graph of the partial correlation network.

Value

GeneNet.wrap The function returns a matrix which is the weighted adjacency matrix of the network inferred by GeneNet algorithm. The shrinkage method used to estimate the partial correlation matrix is "static". - see ggm.estimate.pcor. The probability threshold is set to 0.8. - see ggm.estimate.pcor.

References

Opgen-Rhein, Rainer, and Korbinian Strimmer. "Inferring gene dependency networks from genomic longitudinal data: a functional data approach." RevStat 4.1 (2006): 53-65.

Opgen-Rhein, Rainer, and Korbinian Strimmer. "Using regularized dynamic correlation to infer gene dependency networks from time-series microarray data." Proceedings of the 4th International Workshop on Computational Systems Biology (WCSB 2006), Tampere. Vol. 4. 2006.

Sch\"afer, Juliane, and Korbinian Strimmer. "A shrinkage approach to large-scale covariance matrix estimation and implications for functional genomics." Statistical applications in genetics and molecular biology 4.1 (2005): 32.

See Also

netbenchmark, evaluate, GeneNet-package

Examples

1
2
3
4
5
    # Data
    data <- runif(100)
    dim(data) <- c(10,10)
    # Inference
    net <- GeneNet.wrap(as.data.frame(data))