Paternity Testing Research Today is a free monthly online journal that collates and summarizes the latest research about Paternity Testing, including details on accuracy, methods, procedure, dna testing.

Discrimination of half-siblings when maternal genotypes are known.

Mayor LR, Balding DJ

Department of Epidemiology and Public Health, Imperial College, St. Mary's Campus, Norfolk Place, London W2 1PG, UK.

Given the DNA profiles of two individuals and one parent (say the mother) of each, we present likelihood ratios (LRs) comparing the hypothesis that they have the same father with the hypothesis of unrelated fathers. If the individuals have the same mother, the problem is to distinguish full- from half-siblings, otherwise we are comparing a half-sibling relationship with unrelated. We simulate STR profiles at up to 60 loci, based on allele proportions observed at 15 loci in three populations, and use them to approximate misclassification rates both for binary classification (e.g. "half-sib" versus "unrelated"), and when a third "cannot say" category is included. We find that reliable inferences in the absence of the mothers' profiles require many more STR loci than the 10-25 loci that are currently routinely available. However, profiling the two mothers conveys more discriminatory power than profiling the same number of additional loci in the individuals themselves. Our likelihood ratio formulas include a theta (or Fst) adjustment to allow for the individuals concerned to have recent shared ancestry (coancestry), relative to the population from which the allele frequency database is drawn. We illustrate that using an appropriate value of theta can reduce the average misclassification rate.

Published 19 May 2006 in Forensic Sci Int, 159(2): 141-7.
Full-text of this article is available online (may require subscription).

© 2004-2008 Paternity Testing Research Today. All Rights Reserved.