This function fits a negative binomial model to transcript counts of a group of cells thereby deconvoluting variability into sampling noise, global cell-to-cell variability of transcript counts, and residual variability, which corresponds to biological noise.
fitNBtb(z, gamma = 2, x0 = 0, lower = 0, upper = 100, grad = TRUE)
Data.frame with four columns:
Mean expression.
Biological noise.
Dispersion parameter capturing global cell-to-cell variability of transcript counts.
Dispersion parameter capturing global cell-to-cell variability of transcript counts and biological noise.
Transcript count matrix with cells as columns and genes as rows.
Positive real number. Scale paramter of the cauchy prior. Default is 2.
Real number greater or equal to zero. Location parameter of the cauchy prior.
Real number greater or equal to zero. Lower bound for the maximum a posterior inference of the biological noise. Default is 0.
Real number greater or equal to zero. Upper bound for the maximum a posterior inference of the biological noise. Default is 100.
Logical. If TRUE
then maximum a posterior value is inferred by determining the root of the gradient function. Otherwise, the maximum of the posterior probability is determined numerically. Default is TRUE
.