ADAM.tradeoffEO_TPR: Calculate ADaM threshold
In francescojm/ADAM: Adaptive Daisy Model

Description Usage Arguments Details Value Author(s) See Also Examples

This function finds the minimum number of cell lines in which a gene needs to be fitness in order to be called core-fitness. This is defined as the n providing the best trade-off between i) coverage of priori-known essential genes in the resulting set of predicted core-fitness genes, i.e. fitness in at least n cell lines, and ii) deviance from expectation of the number of fitness genes in n cell lines.

1	ADAM.tradeoffEO_TPR(EO,TPR,test_set_name)

`EO`	Profile of empirical odds values. Computed with the `ADAM.empiricalOdds` function.
`TPR`	Profile of True positive rates for across number of cell line. Computed with the `ADAM.truePositiveRate` function.
`test_set_name`	Name to give to the analysis, used for plotting titles.

Compare and plot the log10 odds ratios with the true positive rates to find the cross over point where the true positive rate falls below the odds ratio.

point

Number of cell lines for which a gene needs to be a fitness gene in order to be predicted as core-fitness gene.

C. Pacini & F. Iorio

ADAM.empiricalOdds, ADAM.truePositiveRate

data(exampleDepMat)

pprofile<-ADAM.panessprofile(depMat=exampleDepMat)

nullmodel<-ADAM.generateNullModel(depMat=exampleDepMat,ntrials = 1000)

data(curated_BAGEL_essential)

EO<-ADAM.empiricalOdds(observedCumSum = pprofile$CUMsums,
                       simulatedCumSum =nullmodel$nullCumSUM)

TPR<-ADAM.truePositiveRate(exampleDepMat,curated_BAGEL_essential)

crossoverpoint<-ADAM.tradeoffEO_TPR(EO,TPR$TPR,
                                    test_set_name = 'curated BAGEL essential')