estimateCCMErrorRates {ppiStats}R Documentation

Estimate false positive and false negative error probabilities

Description

Estimate false positive and false negative error probabilities for complex comembership edges using a protein complex interactome gold standard

Usage

estimateCCMErrorRates(m,GS,filterSystematic=TRUE,       
         obsPropThresh=1,SystematicpThresh=.01)

Arguments

m The bait to prey data adjacency matrix. Baits index the rows and prey index the columns.
GS A gold standard protein complex interaction incidence matrix. Proteins index the rows and protein complexes index the columns.
filterSystematic A logical. If TRUE, all baits with with highly uneven directed degree will be filtered out of the data.
obsPropThresh A numeric between 0 and 1. The proportion of tested proteins found within a protein complex needed to keep that protein complex within the gold standard set.
SystematicpThresh A numeric between 0 and 1. The p-value threshold by which systematic errors are filtered.

Details

The model is described in the manuscript Estimating node degree in bait-prey graphs. by D. Scholtens et al.

Value

A list:

globalpTP A numeric between 0 and 1. Estimate of pTP.
globalpTPSE A numeric. Estimate of standard error of globalpTP estimate.
globalpFP A numeric between 0 and 1. Estimate of pFP.
pTP95CI A vector of length 2. 95 percent confidence interval upper and lower bounds for globalpTP estimate.
pFP95CI A vector of length 2. 95 percent confidence interval upper and lower bounds for globalpFP estimate.
nEligComplexes A numeric. Number of complexes from GS that met obsPropThresh criteria.
nEligBaits A numeric. Total number of eligible baits in GS set.
nEligEdges A numeric. Total number of eligible edges in GS set.
nBaitsInComplexes A vector. Number of baits in each eligible complex.
complexSizes A vector. Size of each complex in GS set.

Author(s)

T. Chiang and D. Scholtens

References

Scholtens D, Chiang T, Huber W, Gentleman R. Estimating node degree in bait-prey graphs. Bioinformatics. To appear.

Examples

data(Ho2002BPGraph)
data(ScISIC)
Ho2002mat = as(Ho2002BPGraph,"matrix")
estimateCCMErrorRates(Ho2002mat,ScISIC)$globalpTP

[Package ppiStats version 1.8.0 Index]