condLogLikDerSize {edgeR} | R Documentation |
Derivatives of the conditional negative-binomial log-likelihood (for each tag/transcript) with respect to the common dispersion parameter, for a single group of replicate libraries of the same size.
Parameterized in terms of size or precision (1/phi
).
condLogLikDerSize(y, r, der=1)
y |
matrix of (pseudo) count data |
r |
size parameter of negative binomial distribution |
der |
order of derivative required, either 0 (the function), 1 (first derivative) or 2 (second derivative) |
The library sizes must be equalized before running this function. This function carries out the actual mathematical computations for the conditional log-likelihood and its derivatives, calculating the conditional log-likelihood for each tag/transcript.
vector of function/derivative evaluations, one for each transcript
Mark Robinson, Davis McCarthy
y <- matrix(rnbinom(10,size=1,mu=10),nrow=5) condLogLikDerSize(y,r=1,der=1)