The GLM likelihood ratio test is based on the idea of fitting negative binomial GLMs with Cox-Reid dispersion estimates.
The R package glmmadmb [ 31] was used to fit negative binomial mixed effects models.
We assume that an individual receives n bites, where the size of each sporozoite inoculum, Si (such that the total initial inoculum S0 = Σ Si), is sampled from a negative binomial distribution obtained by fitting a negative binomial model to counts of parasites inoculated per-bite by anopheline mosquitoes.
All models will be corrected for possible overdispersion, either by using an adjustment parameter in the Poisson model or by fitting a negative binomial regression model.
We obtained sex-specific and age-group specific MN meage-group specificfitting a negative binoMNal randomeanfrequencies to the MN count of each subyect (Lindsey 1995).
Differential expression was evaluated by fitting a negative binomial generalized linear model for each feature and then adjusting the P-value for multiple testing using the Benjamini-Hochberg correction [ 78] with a false discovery rate (FDR) of 0.05.
Transition probabilities between states are determined by fitting a negative binomial distribution to the recombination site choice for VH, DH and JH and number of n nucleotides in the junctions as determined in a set of 293 unmutated sequences rearranged sequences (Fig. 2, Jackson et al., 2004).
Percent recovery was estimated by fitting a negative binomial regression model to the observed number of oocysts, with the number of spiked or expected oocysts functioning as the offset variable and fecal identification as a cluster variable due to using three replicates per concentration per fecal sample [ 25].
We compared our segmentation with that corresponding to the SGD annotation through the Hellinger distance by fitting a negative binomial distribution on each segment and repeated this comparison with the other two algorithms able to process long count datasets: PELT [ 16] and Binary Segmentation [ 6], both implemented in the R package changepoint for the Poisson distribution.
Three models were then compared, the Poisson, the best-fit negative binomial, and the geometric (negative binomial with dispersion parameter equal to 1).
