ganet.getperformance: Performance results for Overlapping Significance

In altayg/ganet: Global Assessment of NETwork Inference Algorithms Based on the Available Literature of Gene/Protein Interactions

Description

Determines a statistical p-value that shows the significance of the overlap between two networks.

Usage

  ganet.getperformance(PredictedNet, allgenes, literature)

Arguments

PredictedNet	The predicted network as two colums matrix.
allgenes	All the genes that are available in the begingin of the prediction analysis as unuique vector.
literature	Optional. If not entered then all the literatyre combined automatically. Or you can input the database of interest as literature database as unique matrix of two columns.

Details

This may be used to practically get performance metrics, precision and p-value and others for the overlapping analysis.

Value

It results a vector that contains al the performance metrics.

performances

All the statistical results in a vector of the following outputs: "precision", "nOverlap (TP)", "nPredicted", "nFocusedSet", "nUniverse", "p.value" from hyper-geometric test.

Author(s)

Gokmen Altay

References

1. G. Altay, J Zapardel, B Peters, "Determining preprocessing combinations and sample size for Gene Network Inference over ...", bioArxiv, 2017.

See Also

ganet.ComLinks, ganet.hyperg, ganet.combine, ganet.UniqNetSimp

Examples

# data(ganet.ex.net)  # assume this is the predicted protein interactions

# Validated <- ganet.ComLinks(netlist=as.matrix(ganet.ex.net), netdata=as.matrix(hprd))
  
# data(ganet.ex.genes)  # load the example unique genes of the example 

#  res <- ganet.getperformance(PredictedNet=ganet.ex.net, allgenes=ganet.ex.genes

#performs overlap analysis over all the literature interactions  
# res

altayg/ganet documentation built on May 11, 2019, 11:29 p.m.