The effect of relatives on the likelihood ratio associated with DNA profile evidence in criminal cases

Abstract

In DNA profiling sometimes a match is declared between the DNA profile from a suspect and that from a scene-of-crime DNA sample. DNA evidence has frequently been presented in the form of a likelihood ratio, the ratio of the probabilities of the data set under the two hypotheses of a single and two sources for the matching DNA profiles. The calculation of the probability of a match is usually performed using a product rule with information from an appropriate database. This approach has been criticized for failing to allow for genetic relatedness, such that the suspect could be a close relative of the source of the scene-of-crime DNA profile. This paper suggests ways of incorporating the possibility of relatives into the likelihood ratio, and shows that unless there is strong evidence implicating a full sibling of the accused, allowing for possible relatedness has very little impact.