testDropout {LineagePulse} | R Documentation |
Performs one test for entire data set.
testDropout(objLP)
objLP |
(LineagePulseObject) LineagePulseObject with fitted null and alternative models. |
(data frame) Summary of hypothesis test
Gene: Gene ID.
p: P-value for existance of drop-out in data set. If high, the data can be explained with models based on NB noise and zero-inflation is not necessary: The null hypothesis of no zero-inflation cannot be rejected.
loglik_zinb: Loglikelihood of full model with ZINB noise (all genes).
loglik_nb: Loglikelihood of reduced model with NB noise (all genes).
df_full: Degrees of freedom of full model with ZINB noise (all genes).
df_red: Degrees of freedom of reduced model with NB noise (all genes).
David Sebastian Fischer
Called by user.
lsSimulatedData <- simulateContinuousDataSet( scaNCells = 100, scaNConst = 10, scaNLin = 10, scaNImp = 10, scaMumax = 100, scaSDMuAmplitude = 3, vecNormConstExternal=NULL, vecDispExternal=rep(20, 30), vecGeneWiseDropoutRates = rep(0.1, 30)) matDropoutPredictors <- as.matrix(data.frame( log_means = log(rowMeans(lsSimulatedData$counts)+1) )) objLP <- runLineagePulse( counts = lsSimulatedData$counts, dfAnnotation = lsSimulatedData$annot, strMuModel = "splines", scaDFSplinesMu = 6, strDropModel="logistic", matPiConstPredictors = matDropoutPredictors) testDropout(objLP)$p