Interpreting low template DNA profiles David J. Balding a, *, John Buckleton b a Department of Epidemiology and Public Health, Imperial College, St Mary’s Campus, Norfolk Place, London W2 1PG, UK b ESR Private Bag 92021, Auckland, New Zealand 1. Introduction In 2003 James Garside and Richard Bates were convicted in London of the murder of Marilyn Garside, the estranged wife of James Garside. It was the prosecution’s case that Garside had hired Bates to murder his wife. The initial conviction was successfully appealed in 2004, leading to a retrial in 2005 which was followed by an unsuccessful appeal in 2006 [1]. Mixed DNA profiles from the crime scene formed a key part of the evidence. The major component of these crime scene profiles (CSP) corresponded to the victim and was assumed to be from her (her full profile was available). In addition, up to eight minor- component alleles were identified. Bates was subsequently arrested and profiled, revealing that the 17 distinct alleles in his 10-locus profile (Table 1) included all 8 CSP minor-component alleles, as well as a further six that were masked by the victim. However three of Bates’ alleles, one at locus D2 and two at D18, were not reported in any of the large number of electropherograms (epg) generated in the investigation, from different crime scene samples and different profiling runs under varying laboratory conditions. The prosecution allegation that Bates was the origin of the CSP minor component implies that allelic drop-out must have occurred at D2 and D18. Drop-out arises when an allele that is carried by an individual contributing DNA to a sample is not reported in a DNA profile obtained from that sample. A related phenomenon, drop-in, occurs when trace amounts of DNA, for example from a crime scene environment or laboratory plasticware, generate one or more spurious alleles in the profile. It is rare for drop-out or drop-in to occur with good-quality samples not subject to degradation or inhibition, but they become more likely as the amount of DNA template is reduced or environmental exposure increases. In Bates, alleles not attributable to the defendant were evident in some crime scene profiles, but not in the most incriminating CSP described in Table 1. Under conditions in which drop-out is plausible, it is difficult to entirely rule out the possibility of drop-in and we allow for this possibility in our analyses below. The prosecution in effect ignored locus D18 in computing its measure of evidential weight, which was equivalent to a ‘‘Random Man Not Excluded’’ (RMNE) probability, see [2] and Table 2. The widespread policy of ignoring loci showing no minor-component alleles is commonly thought to be neutral or conservative, a view sometimes supported by sayings such as ‘‘the absence of evidence is not evidence of absence.’’ Below, we show that in the presence of masking it can be very unfair to defendants. At D2, one allele corresponding to Bates was reported, but not his other allele. The RMNE calculation uses the ‘‘2p rule’’, under Forensic Science International: Genetics 4 (2009) 1–10 ARTICLE INFO Article history: Received 20 October 2008 Received in revised form 27 February 2009 Accepted 26 March 2009 Keywords: Low copy number LTDNA Drop-out Drop-in Masking Garside and Bates ABSTRACT We discuss the interpretation of DNA profiles obtained from low template DNA samples. The most important challenge to interpretation in this setting arises when either or both of ‘‘drop-out’’ and ‘‘drop- in’’ create discordances between the crime scene DNA profile and the DNA profile expected under the prosecution allegation. Stutter and unbalanced peak heights are also problematic, in addition to the effects of masking from the profile of a known contributor. We outline a framework for assessing such evidence, based on likelihood ratios that involve drop-out and drop-in probabilities, and apply it to two casework examples. Our framework extends previous work, including new approaches to modelling homozygote drop-out and uncertainty in allele calls for stutter, masking and near-threshold peaks. We show that some current approaches to interpretation, such as ignoring a discrepant locus or reporting a ‘‘Random Man Not Excluded’’ (RMNE) probability, can be systematically unfair to defendants, sometimes extremely so. We also show that the LR can depend strongly on the assumed value for the drop-out probability, and there is typically no approximation that is useful for all values. We illustrate that ignoring the possibility of drop-in is usually unfair to defendants, and argue that under circumstances in which the prosecution relies on drop-out, it may be unsatisfactory to ignore any possibility of drop-in. ß 2009 Elsevier Ireland Ltd. All rights reserved. * Corresponding author. Tel.: +44 20 7594 3309. E-mail address: d.balding@imperial.ac.uk (D.J. Balding). Contents lists available at ScienceDirect Forensic Science International: Genetics journal homepage: www.elsevier.com/locate/fsig 1872-4973/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.fsigen.2009.03.003