Selecting the number of mutational signatures using a perplexity-based measure and cross-validation

---

Documentation for package ‘selectKSigs’ version 1.14.0

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

calcPMSLikelihood	A function for calculating the log-likelihood from the data and parameters
Calculate_Likelihood_test	Output the maximum potential scale reduction statistic of all parameters estimated
convertFromTurbo_F	Restore the converted parameter F for turboEM
convertFromTurbo_Q	Restore the converted parameter Q for turboEM
convertToTurbo_F	Convert the parameter F so that turboEM can treat
convertToTurbo_Q	Convert the parameter Q so that turboEM can treat
cv_PMSignature	Output the maximum potential scale reduction statistic of all parameters estimated
getBG	Get the statsus of using the background signature
getCounts	Get the count data in a matrix
getExposures	Get a matrix of mutational exposures of signatures
getFeatures	Get a vector of possible features
getFeatureVec	Get a matrix of feature vector list
getK	Get the number of signatures
getLL	Get the values of loglikelihood
getLogLikelihoodC	Calculate the value of the log-likelihood for given parameters
getSamplelist	Get the sample list
getSamplelistG	Get the sample list
getSignatures	Get an array of signature feature distributions
getTranscription	Get the statsus of specifying the transcription bias
select_kth_fold	Output the training data or test data
splitG	Output the maximum potential scale reduction statistic of all parameters estimated