Forensic Science, Medicine, and Pathology

, 4:129

DNA reviews: low level DNA profiling

Authors

    • East Midlands Forensic Pathology UnitUniversity of Leicester
DNA Review

DOI: 10.1007/s12024-008-9044-x

Cite this article as:
Graham, E.A.M. Forensic Sci Med Pathol (2008) 4: 129. doi:10.1007/s12024-008-9044-x

Abstract

Low copy number (LCN) DNA profiling has recently been scrutinized in the United Kingdom following the comments of Mr Justice Weir made during the trial of suspected terrorist Sean Hoey. Mr Hoey was acquitted of all charges related to the Omagh bombing of 1998, following the inadmissibility of key DNA evidence during this trial. The Association of Chief Police Officers and Crown Prosecution Service, initially suspended the use of this technique, but quickly reinstated its use following an internal enquiry. This review describes the low copy number technique and the sample types that are now routinely collected from suspects, victims, and crime scenes for examination by this method.

Keywords

ForensicDNA transferLow copy number (LCN)Touch DNAStochastic effects

Introduction

Low copy number (LCN) DNA profiling has been offered as a specialist service by the Forensic Science Service (FSS) since 1999 and has been successfully applied to high profile cases in the United Kingdom and internationally including the trial of Australian, Bradley Murdoch, who was convicted of murdering British backpacker Peter Falconio in 2001 despite his body not being recovered. More recently, the LCN technique has been used in the UK to identify the “Suffolk Strangler” Steve Wright, the murderer of five prostitutes whose bodies were discovered over a 12-day period in December 2006. LCN has also been used in the investigation of the disappearance of Madeleine McCann from Praia da Luz, Portugal in May 2007, without success. Despite the admissibility of LCN DNA evidence in numerous cases to date, a recent judgment by Mr Justice Weir in the Omagh bombing case, R. v Hoey, delivered on December 20, 2007 questioned the reliability and scientific validation of this technique. During this trial it emerged that an LCN swab reportedly collected from part of the bomb timer, was matched to a 14-year-old boy from Nottingham, England who was completely unconnected with the terrorist attack carried out in Omagh, Northern Ireland in 1998. Questions and concerns raised during this trial led the Association of Chief Police Officers (ACPO) to suspend the use of LCN analysis in criminal investigations in England and Wales pending an internal review by the Crown Prosecution Service (CPS) on December 21, 2007. The use of the LCN technique was resumed on January 14, 2008 after the internal enquiry carried out by Professor Brian Caddy found that there was nothing to suggest that there were any current problems with LCN. The CPS concluded from this enquiry that LCN analysis provided by the FSS should remain available as potentially admissible evidence, while noting that the strength and weight of such evidence in any individual case remains a matter to be considered and should be presented and tested in the light of all other evidence in that case.

Low copy number DNA profiling

The term low copy number (LCN) DNA profiling was coined in a publication by Dr Peter Gill of the FSS in 2000 (1). This article describes the analysis of samples containing less than 100 pg, approximately 16 whole diploid cells, of DNA using the AmpFlSTR® SGM Plus® PCR Amplification Kit by Applied Biosystems. Commercially available kits such as this, are optimized and validated to produce DNA profiles when 0.5–2.5 ng is amplified for 28 cycles. Following this standard protocol, amplification of 100 pg DNA will typically result in allelic and locus dropout, or even complete failure. The sensitivity of this method can be radically improved by simply increasing the number of PCR cycles from 28 to 34 [1]. The increase in sensitivity opens up the potential of forensic DNA profiling to the analysis of many more sample types than had ever previously been imagined. The improved sensitivity gained by the use of six extra PCR cycles does, however, lead to additional issues in DNA profile interpretation. When undertaking any forensic DNA analysis procedure, the importance of correct handling and anti-contamination protocols cannot be underestimated. This is especially true when attempting to produce LCN profiles. The sensitivity of the AmpFlSTR® SGM Plus® PCR Amplification Kit when using 34 PCR cycles is such that laboratory derived contamination cannot actually be completely avoided [1]. The potential to acquire DNA profile information from minute template amounts will also lead to the production of partial DNA profiles, where both allele and locus dropout are observed. In order to combat these problems, along with increase stuttering and unbalanced amplification of heterozygous alleles, an entirely new set of DNA profile interpretation rules is required for LCN analysis.

The majority of the complications observed for LCN DNA profiles are due to stochastic variation. For example, a stutter band generated in an early PCR cycle, unassociated contaminating DNA fragments or one heterozygous allele can be preferentially amplified and therefore over-represented in the final DNA profile. The random nature of stochastic variation means that the same stutter product or contaminating DNA fragment will not be amplified identically in replicate analyses [2], leading to the first rule of LCN DNA analysis; an allele can only be reported if it is present in at least two replicate amplifications [3]. Once this rule has been applied, all remaining bands in the DNA profile can now be scrutinized. For standard 28 cycle DNA profiles produced with the AmpFlSTR® SGM Plus® PCR Amplification Kit, stutter peaks are not observed to exceed 15% of the parent allele’s peak height. In LCN analysis, stutter products are much larger, in the range of 20% when associated with peaks >10,000 RFU in area and 40% when associated peaks are <10,000 RFU [4]. As this cannot be avoided, it must always be considered that the stutter peak could be masking a true allele, potentially of a minor contributor to a mixed DNA profile. Similarly for heterozygosity balance, in a standard DNA profile the two associated alleles are always found to be within 60% of the peak area of each other, whereas in LCN DNA profiles the smaller allele of the heterozygous pair can be observed to measure just 20% of the larger one when peak areas are >10,000 RFU [4]. Despite the existence of rules and guidelines to LCN DNA profile interpretation, the lack of a steadfast understanding of the peculiarities of these profiles leaves a certain amount of subjectivity in their analysis, which can lead to reporting inconsistencies [5] which cannot be acceptable in forensic casework.

DNA transfer

By its nature, LCN DNA profiling is usually applied to minute samples such as swabbed fingerprints or other biological traces that cannot be visibly detected by the naked eye. The possibility of producing a DNA profile from “touch DNA” was first introduced by van Oorschot and Jones in their breakthrough 1997 Nature publication describing DNA fingerprints from fingerprints [6]. This coincided with the first attempts to identify the perpetrators of physical violence through DNA transferred to the victim during contact, in this case by DNA typing of epithelial cells on the neck surface of manual strangulation victims [7]. These papers describing the production of DNA profiles from trace evidence opened up an infinite number of potential sources of biological material in criminal investigations. A potential problem with analyses of this evidence type was also quickly acknowledged; secondary transfer. Although van Oorschot and Jones reported that substantial DNA transfer occurs during the initial contact with an object, they also noted that items handled by numerous individuals produced mixed DNA profiles, not always arising from the last individual to handle each particular object [6]. This persistence of DNA on handled objects could potentially hinder DNA profiling of trace evidence, such as fingerprints, by resulting in DNA mixtures, including components of innocent third parties [8]. A systematic analysis of secondary transfer was undertaken by Ladd and coworkers that identified two potential modes of secondary transfer that might be experienced in field situations or at crimes scene, namely skin-to-skin and skin-to-object transfer [8]. In order to demonstrate the potential problem a number of frequently handled objects, such as door handles, coffee cups [8], pens and telephones [6] were swabbed during DNA transfer investigations. The resultant DNA profiles showed, as anticipated, that DNA from multiple sources was detected on these items, with the majority of profiles displaying complex mixtures, with evidence of allele and locus dropout in some amplifications. In addition to secondary transfer, the possibility of tertiary transfer was commented upon by Rutty in 2002, who further demonstrated the transfer of “offender” DNA to the neck of “victims” during manual strangulation simulations [9]. Professor Rutty also demonstrated in 2003 that DNA arising from the mouth could be deposited on to objects without physical contact taking place [10], this finding was later confirmed by his group when it was demonstrated the oral DNA could be expelled 184 cm ahead of a speaking individual [11]. Aside from the DNA profile interpretation issue discussed above, another major consideration associated with the collection of trace DNA, is the origin of the biological sample being collected. Unlike larger visible stains, to which presumptive tests can be applied to determine the origin of the sample prior to DNA testing, during trace DNA investigations the source of any foreign biological material collected cannot be determined [12, 13]. Equally, it is not possible to determine exactly when or by what mechanism the foreign material was deposited onto the surface being sampled. These observations, along with the findings of multiple DNA contributors on handled objects were noted by Rutty, who also raised the question of what DNA is normally present on the human skin surface due to normal everyday interactions [9]. Despite the potential problems that may arise in LCN DNA profile interpretation if high levels of nonself DNA are normally found to be present on the human skin surface, it has not been thoroughly investigated to date, although this work is currently being carried out by Professor Rutty’s group [14].

Summary

LCN DNA profiling allows for the analysis of low level DNA samples that would not be possible using conventional protocols as recommended by the manufacturers of commercial DNA profiling kits. This technique, despite being accepted as admissible evidence in the United Kingdom is still subject to scientific query and criticism, largely due to the lack of consistency in collected samples and therefore DNA profiles produced by this in this way. This inconsistency is reflected in the scientific literature, with academic research papers currently asking more questions than providing answers for the reliability of LCN technique. The usefulness of this technique will ensure its continued use in forensic investigations but further research into DNA transfer and LCN DNA profile interpretation would increase its evidential value and decrease public misconceptions of DNA profiling methodology.

Educational message

  1. 1.

    The term low copy number DNA profiling refers to the amplification of less than 100 pg DNA using the AmpFlSTR® SGM Plus® PCR Amplification Kit at 34 PCR cycles.

     
  2. 2.

    Low copy number DNA profiles are admissible as evidence in the United Kingdom.

     
  3. 3.

    The interpretation of low copy number DNA profiles is hindered by increased artefact production, decreased heterozygosity balance and allele dropin/dropout.

     
  4. 4.

    It is currently not possible to determine when or by what mechanism biological material collected for low copy number DNA profiling was deposited at the sampled site.

     

Copyright information

© Humana Press 2008