estimateModelParameters: Estimate the SCAMPI model paramteres alpha, beta, mu and tau...

In protiq: Protein (identification and) quantification based on peptide evidence

Description

Estimate the parameter values with maximum likelihood (MLE) or a method of moments approach (LSE), or both (all)

Usage

estimateModelParameters(method = "all", ccList, peptides = NULL, 
                        numIter = 10, verbose = FALSE)

Arguments

method	method to be used for the parameter estimation; can be all (MLE and LSE, default), LSE or MLE
ccList	list of pre-processed connected components
peptides	data frame with pre-processed peptide info (only used for LSE)
numIter	number of successful munerical optimizations to perform (only used for MLE, see details)
verbose	If TRUE, detailed output is provided.

Details

To use method="MLE" the inverses of the covariance matrices (of the connected components) are needed. Depending on the chosen parameters, this can lead to stability issues. To avoid the function from crashing, a try(...) bolck is used: the parameter estimation is performed until it was successful numIter times. Among these numIter sets, the one with the lowest negative log-likelihood value is returned.

Value

Named list of vectors with at least four named elements: alphaH, betaH, muH and tauH (estimates for the four model parameters). The name of the list elements corresponds to the parameter estimation method, namely LSE or MLE.

Note

This function is called by one of the main functions of the package, runScampi. Calling this function directly is seldomly necessary.

Author(s)

Sarah Gerster [e-mail]

See Also

runScampi, iterateScampi

Examples

## Not run: 
data("leptoSRM")

dataChecked <- checkInputData(scampi(peptides=leptoSRMpeptides, 
                                     proteins=leptoSRMproteins, 
                                     edgespp=leptoSRMedgespp),
                              rescaling=FALSE)
tmpPrepro <- preprocessInputData(dataChecked)
dataPrepro <- tmpPrepro[["dataPrepro"]]
ppGraph <- tmpPrepro[["ppGraph"]]
myCCList <- tmpPrepro[["ccList"]]
rm("tmpPrepro")
scampiParam <- estimateModelParameters(method="all", ccList=myCCList,
                                       peptides=dataPrepro@peptides,
                                       numIter=10)

## End(Not run)

protiq documentation built on May 2, 2019, 9:06 a.m.