estErrorRates: Estimate Error Rates

In eLNNpaired: Model-Based Gene Clustering for Genomics Data from Paired/Matched Designs

Description

Estimate error rates: false discovery rate (FDR), false non-discovery rate (FNDR), false positve rate (FPR), and false negative rate (FNR) for data with known cluster membership for gene probes.

Usage

estErrorRates(memGenes2.est, memGenes2.true)

Arguments

memGenes2.est	estimated cluster membership for gene probes. memGenes2.est must take only two values 1 and 0.
memGenes2.true	true cluster membership for gene probes. memGenes2.true must take only two values 1 and 0.

Details

FDR is the proportion of detected differentially expressed (DE) gene probes that are truly non-differentially expressed (NE) gene probes among all detected DE gene probes.

FNDR is the proportion of detected NE gene probes that are truly DE gene probes among all detected NE gene probes.

FPR is the proportion of detected DE gene probes that are truly NE gene probes among all truly NE gene probes.

FNR is the proportion of detected NE gene probes that are truly DE gene probes among all truly DE gene probes.

Value

A vector of 4 elements: FDR, FNDR, FPR, and FNR.

Author(s)

Yunfeng Li <[e-mail]> and Weiliang Qiu <[e-mail]>

References

Li Y, Morrow J, Raby B, Tantisira K, Weiss ST, Huang W, Qiu W. (2017), <doi:10.1371/journal.pone.0174602>

Examples

  memGenes2.est = c(1, 1, 1, 0, 0, 0)
  memGenes2.true = c(1, 0, 1, 0, 0, 0)

  err = estErrorRates(memGenes2.est = memGenes2.est, 
    memGenes2.true = memGenes2.true)
  print(err)

eLNNpaired documentation built on May 29, 2017, 12:04 p.m.