The heritability of Nematodirus egg counts using a mixture model

Gastrointestinal nematode infections are the
most devastating disease of Australian sheep and beef cattle. Following
infection, parasite eggs can be counted in the faeces and used to identify
relatively resistant and susceptible animals. Parasite egg counts are expected
to follow a mixture of two distributions because not all animals will have been
exposed to infection. For gastrointestinal nematodes like Nematodirus battus,
exposed animals are expected to follow a negative binomial distribution; this
distribution arises as a consequence of multiplicative variation among animals
and poisson variation in the counting process. Unexposed animals will conform
to a uniform distribution of zeros. In other words, egg counts will follow a
zero-inflated negative binomial distribution. This talk will describe the use
of Bayesian Monte Carlo Markov Chain procedures to estimate the heritability of
N. battus egg counts.