An evaluation of potential allelic association between the STRs vWA and D12S391: Implications in criminal casework and applications to short pedigrees Peter Gill a,b, *, Chris Phillips c , Catherine McGovern d , Jo-Anne Bright d , John Buckleton d a Department of Forensic Genetics, Norwegian Institute of Public Health, 0403 Oslo, Norway b Rikshospitalet, University of Oslo, Norway c Forensic Genetics Unit, Institute of Legal Medicine, University of Santiago de Compostela, Spain d ESR, PB 92021, Auckland, New Zealand 1. Introduction Forensic evidence associated with a match of DNA from a scene and a person of interest is preferably presented with an associated assessment of the weight of evidence. This weight of evidence typically involves using the probability that a match might be expected with a random unrelated person in a certain population. This is termed the match probability. The match probability is typically estimated using a population genetic model. The performance of these population genetic models is a matter of interest to the forensic scientist and the courts. Emerging STR multiplexes have loci physically close on the same chromosome. This leads to justifiable concerns about the possibility of allelic association. Tests for the hypothesis of no disequilibrium do exist but tend to have limited power to detect realistic levels of disequilibrium. Data requirements to identify realistic levels of disequilibrium are likely to be very large [1,2] and well beyond the studies already presented. Most frequency databases are comprised of convenience samples from multi-ethnic populations. Databases may be split into racial groups but substantial substructure is expected to remain. Therefore these databases may or may not show linkage disequilibrium caused by subpopulations effects, irrespective of any caused by allelic association due to physical linkage. Physical linkage is not a prerequisite for allelic association. Allelic association typically results from an evolutionary event such as genetic drift in finite populations [3], founder effects, natural selection, population admixture and the associated Wahlund effect. These affect both linked and unlinked loci. Linked and unlinked loci do however differ in the rate at which independence is re-established. The difference is therefore quantitative rather than qualitative. A very plausible candidate for a recent evolutionary event with the potential to cause linkage disequilibrium is admixture of two previously separated populations. Populations such as Hispanic, Forensic Science International: Genetics 6 (2012) 477–486 A R T I C L E I N F O Article history: Received 29 July 2011 Received in revised form 2 November 2011 Accepted 3 November 2011 Keywords: Allelic association vWA D12S391 Linkage Recombination Simulation A B S T R A C T An evaluation was carried out to determine the effect on routine forensic calculations when incorporating STRs D12S391 and vWA. These loci are co-located on the same arm of chromosome 12. It has been suggested that allelic association could result in over-estimates of strength-of-evidence calculations. In the first place, we argue that is very unlikely that genotypes collected from typical cosmopolitan forensic databases can provide meaningful information about effects attributable to physical linkage. Since admixture is the most likely cause of allelic association in modern populations we specifically evaluate this effect. We use computer simulation as the preferred approach to generate populations with disequilibrium and observe the effect on match probability. Although we have specifically evaluated the linkage between D12S391 and vWA, the methods described in this paper can be extended and generalized to evaluate linkage effects between any pair of loci where the recombination rate is known. Many jurisdictions apply a subpopulation correction following the standard method of Balding and Nichols. Such corrections would appear to be more than adequate to compensate for any increase in match probability that we were able to create by this admixture. Linkage is likely to have an appreciable effect on relatedness calculations in short pedigrees in some but not all instances. We examined those circumstances where an effect is likely and give formulae for some common situations. The complexity of these calculations is a cause for concern in some laboratories. We discuss possible strategies that might be employed and plausible effects. ß 2011 Elsevier Ireland Ltd. All rights reserved. * Corresponding author at: Department of Forensic Genetics, Norwegian Institute of Public Health, 0403 Oslo, Norway. E-mail address: peterd.gill@gmail.com (P. Gill). Contents lists available at SciVerse ScienceDirect Forensic Science International: Genetics jou r nal h o mep ag e: w ww .elsevier .co m /loc ate/fs ig 1872-4973/$ – see front matter ß 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.fsigen.2011.11.001