A pedigree-transmission likelihood for multiplex families

Family-based studies of a genetically inherited disease ascertain so-called multiplex pedigrees with several disease-affected members. A parameter of interest in these studies is the probability that a parent transmits a given genetic variant to their child. I describe a likelihood method for estimating the transmission probability of a rare genetic variant based on DNA sequencing data in affected relatives. These pedigree-transmission likelihoods are based on data for the presence or absence of the variant in the affected relatives. I describe how to implement the likelihoods using software for Bayesian Networks in R. The ideas are illustrated with several example pedigrees of various shapes and sizes.