Current and future trends in forensic molecular biology
Simon J. Walsh
Forensic science is part of a process beginning at a crime scene and concluding in a court room. This means that as one of the key forensic disciplines, the field of forensic molecular biology resides within the complex and adversarial context of the criminal justice system (CJS). The key areas of the CJS that are relevant to the use of forensic molecular biology are the domains of law enforcement and the justice system (Figure 1.1). Due to the intersection of these three domains, changes and developments in one can have a resultant impact on the other adjacent areas. Therefore, when considering the current and future trends in forensic molecular biology it is important to do so not only from the perspective of their effect within the forensic field itself, but also from the perspective of their interaction with neighbouring areas of the system. After all, it is in these neighbouring areas that forensic outcomes are eventually put to use.
Forensic molecular biology has developed rapidly into a comprehensive discipline in its own right and, perhaps more so than any scientific advance before it, has had a profound impact across the CJS. Within the forensic science discipline, as expected, development has been science and/or technology driven. It has followed a trend towards achieving greater sophistication, throughput and informativeness for the DNA-based outcomes of scientific analysis. Developments in forensic molecular biology that have influenced law enforcement could be thought of as operational developments as they predominantly apply to the manner or degree that forensic molecular biology is utilized. As such, they typically have both a technical and policy-oriented basis. Progress in forensic biology
Molecular Forensics. Edited by Ralph Rapley and David Whitehouse Copyright 2007 by John Wiley & Sons, Ltd.
Figure 1.1 A simplified representation of the areas of the CJS that are relevant to forensic molecular biology. A large number of cases flow directly between the police and legal domains (solid arrow) whereas a reduced number of cases flow through the forensic domain (dashed arrow). Each of the three areas can be thought of as intersecting and, as such, each has the capacity to exercise some effect on the others
Figure 1.1 A simplified representation of the areas of the CJS that are relevant to forensic molecular biology. A large number of cases flow directly between the police and legal domains (solid arrow) whereas a reduced number of cases flow through the forensic domain (dashed arrow). Each of the three areas can be thought of as intersecting and, as such, each has the capacity to exercise some effect on the others has also influenced the justice sector. This is characterized, for example, by the iterative response of both the legislature and the courts to changes in the volume and nature of forensic DNA tests. Throughout the history of the field there has also been associated debate and controversy accompanying these legal developments. This reflects the array of socio-legal and ethical issues associated with more widespread use of forensic molecular biology.
This chapter chiefly describes the process of development within the forensic molecular biology field. It also touches briefly on the way such developments intersect with the neighbouring fields of law enforcement and the justice system. By considering developmental trends in this way the overall impact of changes in forensic molecular biology can be appropriately placed in context, allowing reflection on their effect to date and foreshadowing their potential effect in the future.
1.2 Developments within the field of forensic molecular biology
From the time the field settled on a uniform technological platform (Gill, 2002), forensic molecular biologists have done a masterful job at extending the applicability of this testing regime as far as conceivably possible. The discriminating power of short tandem repeat (STR)-based tests has been increased by combining up to 16 (Collins et al., 2000; Krenke et al., 2002) STR loci into a single polymerase chain reaction (PCR; see Chapter 2). The sensitivity of the routine tests has also been driven downward so that successful analysis is now achieved from as little as 100 pg of starting template (Whitaker et al., 2001).
Advancing the capabilities of the DNA methodology has also expanded the range of criminal cases and sample types able to be successfully analysed. For many years forensic molecular biology was limited to testing templates such as blood, semen, hair and saliva. However, the increased efficiency of the STR-based methods now means that DNA can be successfully analysed from discarded clothing or personal effects (Webb et al., 2001), skin cell debris from touched or handled surfaces (Van Oorschot and Jones, 1997; Wiegand and Kleiber, 1997; Zamir et al., 2000; Bright and Petricevic, 2004), dandruff (Lorente et al., 1998), drinking containers (Abaz et al., 2002), food (Sweet and Hildebrand, 1999) and fingernail clippings and scrapings (Harbison et al., 2003). Recent approaches such as reduced-amplicon STR analysis (Butler et al., 1998, 2003; Wiegand and Kleiber, 2001; Coble and Butler, 2005) and low copy number (LCN) profiling (Gill, 2001a; Whitaker et al., 2001) have enhanced reaction sensitivity even further and improved the ability to analyse the most troublesome and highly degraded samples.
Many of the routine techniques have been adapted onto automated platforms so as to facilitate high-throughput analysis and reduce the amount of sample handling (Gill, 2002; Varlaro and Duceman, 2002; Fregeau et al., 2003) (see Chapter 3). Computer-assisted data analysis has also further streamlined the analytical process and reduced some areas of subjectivity, such as mixture interpretation (Perlin and Szabady, 2001; Bill et al., 2005; Gill et al., 2005, 2006b) (see Chapter 11). The next generation of laboratory instrumentation includes micro-scale electrophoresis devices (Woolley et al., 1997; Mitnik et al., 2002) that not only promise rapid analysis times but also allow for the possibility of remote or portable laboratory platforms (Hopwood et al., 2006).
The observable trend in the development areas mentioned above is that they are all directed towards improving the ability to undertake routine DNA-based identity testing. Whilst this refinement of routine typing technologies is of vital importance, it has meant that for the most part the field has sought only one dimension of information from biological evidence samples. Through recent research into the physical and genetic properties of human DNA this is now changing, allowing the forensic field to diversify its capabilities and begin to address questions beyond the identification of source.
There are already several examples of forensic molecular biology applications that either apply different forms of typing technologies or address a different line of genetic inquiry via new polymorphisms or loci. One such area is non-autosomal DNA profiling, particularly the analysis of mitochondrial DNA (mtDNA) and Y chromosome markers (see Chapters 7, 8 and 9). Whilst mtDNA analysis has been widely used in human evolutionary biology for a number of years (Cann et al., 1987), its routine application to forensic work has been consistently evolving. In forensic science, mtDNA is most often analysed in circumstances where nuclear DNA fails to give a result, such as in the analysis of telogenic hairs (Wilson et al., 1995), nail material (Anderson et al., 1999) and bone (Bender et al., 2000; Edson et al., 2004) or when distant relatives must be used as reference (Gill et al., 1994; Ivanov et al., 1996; Pfeiffer et al., 2003). Analysis typically involves direct sequencing of the hypervariable regions 1 and 2 (HV1 and HV2, respectively) (Tully et al., 2001) although SNP-based approaches offer the potential to complement or substitute the need for sequencing (Budowle et al., 2004; Coble et al., 2004; Quintans et al., 2004). Recent developmental progress in the forensic use of mtDNA has also been shaped by the context within which it has been required. In particular, the large-scale multi-national response to recent wars (Huffine et al., 2001; Andelinovic et al., 2005), refugee crises (Lorente et al., 2002) and mass fatalities (Roby, 2002; Vastag, 2002; Budjimila et al., 2003; Holland et al., 2003; Budowle et al., 2005) has seen a rapid evolution of these and other specialist identification sciences so as to respond to the unprecedented logistical and technical challenges presented by these circumstances.
The analysis of polymorphisms on the non-recombining portion of the human Y chromosome (NRY) (Jobling et al., 1997; Kayser et al., 1997) has also steadily developed into a valuable forensic technique (Gill et al., 2001; Gusmao and Carracedo, 2003; Gusmao et al., 2006). The male specificity of the Y chromosome makes it particularly suitable for the resolution of problematic situations such as complex mixtures. In a casework setting Y chromosome analysis is especially useful for typing mixed male-female stains that commonly occur as a result of sexual assaults (Dettlaff-Kakol and Pawlowski, 2002; Dziegelewski et al., 2002; Sibille et al., 2002). As with autosomal markers, microsatellites are favoured for forensic Y chromosome analysis and a number of suitable Y-STRs have been identified and validated for forensic use (Bosch et al., 2002; Butler et al., 2002; Redd et al., 2002; Hall and Ballantyne, 2003; Johnson et al., 2003; Hanson and Ballantyne, 2004; Schoske et al., 2004) and a selection of them included into commercially available multiplexes (Shewale et al., 2004; Krenke et al., 2005; Mulero et al., 2006).
Potentially the most valuable target markers for a diverse range of novel forensic molecular biology applications are single nucleotide polymorphisms (SNPs; see Chapter 6). These offer a range of forensic applications in traditional and novel areas and confer some particular advantages in comparison to STRs, including a low mutation rate (making SNPs highly suitable for kinship and/or pedigree analysis), amenability to high-throughput processing and automated data analysis, a shorter PCR amplicon size (assisting their ability to be multiplexed and making them good target loci for highly degraded samples), a vast abundance in the genome, and in some cases simplified interpretation (due to the absence of certain STR artifacts such as stutter). Single nucleotide polymorphisms are being investigated for use in forensics in both the identity testing and intelligence areas.
By virtue of the fact that there is greater allelic diversity at STR loci compared with SNPs, STRs have a profound advantage over SNPs in forensic identity testing. As a crude estimate, one would be required to type three to five SNP loci to discriminate between individuals at the same level as a single STR. This means that to approach the degree of certainty of the current STR kits up to 50 SNP loci would be needed, which presents a formidable technical challenge. In addition, changing routine target loci is undesirable, due largely to the significant investment in databases that has already occurred. In combination, these reasons make a universal change of DNA typing platform unlikely (Gill, 2001b; Gill et al., 2004). Nonetheless, the recent development of more advanced SNP genotyping technologies, and the desirable properties of SNP loci, has seen a continued focus on developing highly informative SNP-based multiplexes for forensic identity testing (Inagaki et al., 2004; Dixon et al., 2006; Kidd et al., 2006; Sanchez et al., 2006).
Single nucleotide polymorphism markers in coding regions linked to physical or behavioural (personality-related) traits are also being researched for forensic purposes. This research aims to provide investigators with an inferred description of an offender, based on biological evidence recovered from a particular crime and subsequent DNA analysis. In one example researchers have described approaches for screening genetic mutations associated with the red-hair pheno-type (Grimes et al., 2001; Branicki et al., 2006). A comprehensive candidate gene study for variable eye colour has also been conducted by an American company DNAPrint Genomics (Sarasota, FL, http://www.dnaprint.com) (Frudakis et al., 2003a). On the basis of this research (Sturm and Frudakis, 2004) DNAPrint Genomics have developed and validated RETINOME™, a high-throughput genetic test for predicting human iris colour from DNA. A blind validation test of RETINOME™ on 65 individuals of greater than 80% European ancestry revealed that the test was 97% accurate in its predictions.
Other SNP-based techniques potentially enable the inference of biogeographical ancestry from a DNA sample. As SNPs can be found in areas of the genome subject to evolutionary-selective pressures, such as coding and regulatory regions of DNA, they can exhibit far greater allele and genotype frequency differences between different populations than other forensic loci. In 2003, Frudakis et al. developed a classifier for the SNP-based inference of ancestry (Frudakis et al., 2003b). This research found that allele frequencies from 56 of the screened SNPs were notably different between groups of unrelated donors of Asian, African and European descent. Using this panel of 56 autosomal SNPs, Frudakis et al. report successful designation of the ancestral background of European, African and Asian donors with 99%, 98% and 100% accuracy, respectively. Applying a reduced panel of the 15 most informative SNPs the level of accuracy reduces to 98%, 91% and 97%, respectively (Frudakis et al., 2003b). This work represents the most significant step towards the development of a DNA-based test for the inference of ancestry in a forensic setting and has led to the generation of a commercially available tool known as DNA Witness™ (DNA-Print Genomics, Sarasota, FL).
A significant amount of research effort has also been invested in the study of non-autosomal SNPs. This approach is commonplace in human migration studies, with a large body of work examining SNP haplotype diversity on the
Y chromosome (Underhill et al., 2000, 2001; Jobling and Tyler-Smith, 2003) or mtDNA genome (Budowle et al., 2004; Jobling et al., 2004; Wilson and Allard, 2004). In the forensic context Y- or mtDNA-SNPs are also potential markers of biogeographical ancestry. They have often been preferred in this capacity as they can be locally customized and applied also to understand local population substructure, which in turn can support statistical interpretation models. Large-scale non-autosomal SNP multiplexes already exist (Sanchez et al., 2003; Brion et al., 2004, 2005; Coble et al., 2004; Quintans et al., 2004; Sanchez et al., 2005) and population data and supporting information are readily available (YCC, 2002).
Commensurate with the advances in the molecular tools available to forensic scientists, the interpretation of DNA evidence has also had to develop considerably over recent years. Early in the history of forensic molecular biology this was an area of heated dispute (Lander, 1989; Lewontin and Hartl, 1991) requiring concerted efforts to address concerns of the scientific and legal community (National Academy of Sciences, 1996). Now there is a far greater depth of understanding and an important sub-discipline of the field has developed (Robertson and Vignaux, 1995; Evett and Weir, 1998; Aitken and Taroni, 2004; Balding, 2005; Buckleton et al., 2005a). Nonetheless, each new molecular adaptation brings an associated requirement to reassess the weight or meaning of the outcomes statistically. Approaches are continually being refined to deal with routine complexities such as mixed profiles (Weir et al., 1997; Curran et al., 1999; Fukshansky and Bar, 2000; Bill et al., 2005; Wolf et al., 2005; Gill et al., 2006a), partial profiles (Buckleton and Triggs, 2006) and relatedness (Ayres, 2000; Buckleton and Triggs, 2005). In addition, novel theory has been needed to assess results obtained from LCN approaches (Gill et al., 2000), non-autosomal markers (Krawczak, 2001; Buckleton et al., 2005b; Fukshansky and Bar, 2005; Wolf et al., 2005), DNA database searches (Balding, 2002; Walsh and Buckleton, 2005), multi-trace cases (Aitken et al., 2003), mass disasters (Brenner and Weir, 2003; Brenner, 2006) and so on.
From this summary we can distil the following trends that appear set to characterize future years. The addition of more routine markers, and the wider use of known ones, appears likely to continue. Testing platforms will increase in their overall efficiency and move closer to the goal of rapid, portable microdevices. Taking the DNA science out of the laboratory is a move that could bring considerable advantage to many investigations but is also one with associated challenges. Progress will continue towards answering more diverse questions than 'who is the source of this DNA sample?'. There is almost limitless potential as to where this approach may lead as we unravel the full potential of information accessible via genetic testing. Of course we must observe that with this increased capability comes an associated increase in complexity. Scientists have the potential to step beyond the routinely applied testing regimes, but to do so they must understand the strengths and weaknesses of new approaches and, importantly, be equipped to deal with associated complexities such as the statistical assessment of outcomes. The forensic community must take ownership of this challenge and continue to ensure that proper validation, training and independent research occur. This will at times be awkward given the growing demands for all forms of DNA analysis and an increasingly commercialized operational environment. It will also be important to ensure appropriate management of expectations regarding emerging capabilities on the part of police, legal professionals and the general public.
1.3 Developments influencing law enforcement -operational impacts
The current environment where forensic molecular biology operates as a tool of the law enforcement community is starkly different to the mid-1980s, when its role in this context first began. This is unsurprising given the rapid evolution of the techniques, as described above. The most notable operational difference is the frequency of use of DNA evidence in criminal casework. Across the world the overall number of cases submitted annually for DNA analysis has increased by many fold. In the UK the average annual inclusion of crime samples onto the national DNA database (NDNADTM) increased from 14 644 for the period 1995-2000 to 59 323 for the period 2000-2005. In Canada, 7052 crime samples were added to the national DNA databank in 2005 compared with 816 in 2000. In NSW (the most populous State of Australia) the annual DNA case submissions have risen from 1107 in 1998 to 10 146 in 2005.
The major driver of this change in case volume has been the global implementation of forensic DNA databases. Forensic DNA databases have altered the landscape of the criminal justice system and irrevocably re-shaped the field of forensic science. Their growth has been rapid with millions of STR profiles now held from convicted offenders, suspects and unsolved crimes (Table 1.1). Links provided through DNA database searches have contributed valuable
Table 1.1 Size and effectiveness of major national DNA databases
Database and date
Table 1.1 Size and effectiveness of major national DNA databases
Database and date
Was this article helpful?
Are You Suffering From Social Withdrawal? Do People Shun Or Ostracize You Because You Have A Hard Time Getting Some Of Your Words Out? Or Does Your Child Get Teased At School Because They Stutter And Cant Speak Like Everyone Else? If you have answered yes to any of the above, then you are in the tiny percentage of people that stutter.