perfDSC {maPredictDSC}R Documentation

Area Under the Precision-Recall Curve (AUPR), Belief Confusion Metric (BCM) and Correct Class Enrichment Metric (CCEM).

Description

This function implements the three metrics used in the IMPROVER Diagnostic Signature Challenge.

Usage

perfDSC(pred,gs)

Arguments

pred

A belief matrix, with rows coresponding to samples and columns to classes. The values are between 0 and 1 and sum on each row is 1. It needs to have row names. The belief values are the result of a prediction made by a model.

gs

A matrix, with rows coresponding to samples and columns to classes that give the true (gold standard) class membership of samples.

Details

See cited documents for more details.

Value

A named vector that includes the BCM, CCEM, AUPR_avg and Accuracy.

Author(s)

Adi Laurentiu Tarca <atarca@med.wayne.edu>

References

Adi L. Tarca, Mario Lauria, Michael Unger, Erhan Bilal, Stephanie Boue, Kushal Kumar Dey, Julia Hoeng, Heinz Koeppl, Florian Martin, Pablo Meyer, Preetam Nandy, Raquel Norel, Manuel Peitsch, Jeremy J Rice, Roberto Romero, Gustavo Stolovitzky, Marja Talikka, Yang Xiang, Christoph Zechner, and IMPROVER DSC Collaborators, Strengths and limitations of microarray-based phenotype prediction: Lessons learned from the IMPROVER Diagnostic Signature Challenge. Bioinformatics, submitted 2013.

See Also

predictDSC

Examples

#asume a 3 class classification problem; gs is the gold standard and pred are predictions
gs=cbind(A=c(1,1,1,1,0,0,0,0,0,0,0,0),B=c(0,0,0,0,1,1,1,1,0,0,0,0),C=c(0,0,0,0,0,0,0,0,1,1,1,1))
rownames(gs)<-paste("sample",1:12,sep="")
pred=cbind(A=c(0.6,0.9,1,0.3,0,0,0,0,0,0,0,0),B=c(0.4,0.1,0,0.7,1,1,0.7,1,0,0,0,0),C=c(0,0,0,0,0,0,0.3,0,1,1,1,1))
rownames(pred)<-paste("sample",1:12,sep="")
#male sure the sum per row is 1 is both gs and pred
apply(gs,1,sum)
apply(pred,1,sum)
#compute perfromance
perfDSC(pred,gs)




[Package maPredictDSC version 1.30.0 Index]