evalLogLikMatrix {LineagePulse} | R Documentation |
This likelihood function is a wrapper computes loglikelihood of entire data set by parallelising loglikelihood computation over genes.
evalLogLikMatrix(matCounts, lsMuModel, lsDispModel, lsDropModel, matWeights = NULL, boolConstModelOnMMLL = FALSE)
matCounts |
(count matrix genes x cells) Observed read counts, not observed are NA. |
lsMuModel |
(list) Object containing description of gene-wise mean parameter models. |
lsDispModel |
(list) Object containing description of gene-wise dispersion parameter models. |
lsDropModel |
(list) Object containing description of cell-wise drop-out parameter models. Set to NULL to evaluate NB likelihood as oppose to ZINB. |
matWeights |
(numeric matrix cells x mixtures) [Default NULL] Assignments of cells to mixtures (for strMuModel="MM"). |
boolConstModelOnMMLL |
(bool) [Default FALSE] Whether to evaluate constant gene-wise mean model on mixture model likelihood. This is necessary for numeric stability when testing whether mixture model fits are worse than constant fits. |
vecLogLik (vector length number of genes) Loglikelihood of each gene under zero-inflated negative binomial model.
David Sebastian Fischer
Called directly by fitZINB
to track
convergence of estimation iteration on entire data set.