Representing and solving complex forensic DNA identification cases using Bayesian networks

摘要

Object-oriented Bayesian networks (OOBNs) can be used to model and solve a wide variety of complex forensicudDNA identication problems, involving such complications as missing individuals, mutation, and null alleles. Weudprovide a brief overview of the approach and illustrate its use. In particular, we investigate the eect on paternityudratios of allowing for silent alleles, and show that this can be substantial even when the probability of silence isudvery small.