A Review of the Science of Low Template DNA Analysis
Executive Summary
From our detailed review we find that the science supporting the delivery of Low Template DNA (LTDNA) analysis
is sound and that the three companies (the Forensic Science Service Ltd, LGC Forensics and Orchid Cellmark Ltd) providing
this service to the Criminal Justice System have validated their processes in accord with accepted scientific principles using
both 28 and 34 PCR cycles for extracts containing less than 200 picograms (pg) of DNA. At these levels, stochastic and inhibition
effects have an impact upon the DNA profiles produced and all those involved in this process have established guidelines for
profile interpretation. Work on interpretation is continuing and it is for the Forensic Science Regulator to monitor this
and to bring about some standardisation in interpretation amongst all providers.
Now that suitable commercial analytical systems are available for the quantification of low levels of DNA, it
becomes important that all DNA samples submitted for analysis under the Criminal Justice System must, as a matter of best
practice, be quantified before attempts are made to produce DNA profiles.
Training both laboratory personnel and those involved in the recovery of DNA samples from crime scenes requires
to be standardised. It is for the Forensic Science Regulator through a dialogue with all providers and ACPO to establish what
those standards should be, to implement them and to monitor their application.
The use of DNA-free consumables both in the laboratory and for those working at the crime scene is essential.
Standards for these need to be set and quality control mechanisms put in place to monitor the status of these materials/chemicals.
Although used for a number of years we do not yet have any reliable measure of the success rate of LTDNA analysis
and this need to be corrected.
We have been made aware that there is concern by the forensic science providers of the role of Police Laboratories
established by the larger police forces. This concern relates to the quality of work and their future role in the provision
of forensic science services. Additionally they have expressed concern over the funding of forensic science. It is for the
Forensic Science Regulator to enter into a dialogue with ACPO to allay these concerns and /or to develop mechanisms that can
overcome any envisaged problems.
We have become aware that there is a desperate need for independent research funding in order to advance the
discipline of forensic science. Such funding should be open to all on a competitive basis in the same way as the government
funded research councils, who are at present unable or unwilling to finance any developments in this arena.
Finally, we believe that it is important for the Forensic Science Regulator to monitor all documentation associated
with accreditation and validation.
APRIL 2008
B. Caddy
G. R. Taylor
A.M.T. Linacre
A Review of the Science of Low Template DNA Analysis
by
Professor Brian Caddy, BSc., PhD., CChem., MRSC.*
Dr Graham R. Taylor, BSc., PhD., FRCPath., SRCS**
Dr Adrian M.T.Linacre, BSc., DPhil., RFP*
* University of Strathclyde ** Cancer Research UK
1. Introduction
1.1 Doubts have been expressed about the use of Low Template DNA analyses (LTDNA) for legal purposes (as opposed
to the investigating process) and it has been reported that few countries use this process routinely. Since such concerns
have been expressed, the Forensic Science Regulator instigated this review and appointed Prof. Brian Caddy of Strathclyde
University to chair this process with the support of two DNA experts, Dr Graham Taylor of Cancer Research UK and Dr Adrian
Linacre of Strathclyde University.
The terms of reference given to this review panel are:
i). To examine low template DNA profiling techniques, including the Low Copy Number (LCN) technique employed
by the Forensic Science Service Ltd (FSS), and analogous processes used by other providers of DNA profiling services to the
UK Criminal Justice System (CJS), to generate DNA profiles from samples which may not yield useable results from the standard
DNA profiling (termed SGM Plus® process discussed in section 1.4). This is to include processes which seek to obtain profiles
from DNA samples below 200 picogram (pg) and the application of supra-28 cycle amplification;
ii). To advise upon the scientific validity of those techniques, having regard to any novel issues raised (in
comparison with accepted SGM Plus® techniques) and the variations in approach adopted by different providers, recommending
best practice in the light of current scientific knowledge and opinion;
iii). To comment upon the interpretation of the results and how they should be presented to the customer and
to the court in any criminal proceedings;
iv). To advise upon the creation of a national minimum technical standard for low template DNA analysis, to
include extraction, quantification/dilution and interpretation criteria; and
v). To make other relevant recommendations.
1.2 The approach adopted for this review has been to visit the three laboratories that constitute the main providers
of forensic science services to the CJS and to speak with personnel involved with LTDNA analysis at both the bench and research
levels as well as the management of the laboratories. Additionally, we have visited Harperley Hall (National Policing Improvement
Agency Training Centre) to speak with the trainers of crime scene officers involved in the recovery of DNA samples from crime
scenes. Furthermore, we have interviewed: a representative of United Kingdom Accreditation Service (UKAS) who are involved
in accrediting forensic science laboratory processes, the Custodian of the National DNA Database (NDNAD) and the person responsible
for forensic science support at the Metropolitan Police. We have also participated in a forum with senior officers of the
Metropolitan police and have corresponded with a representative number of police forces. Our view from the commencement of
this review was that for completion it was not sensible to divorce the laboratory processing surrounding LTDNA from those
of police activities associated with the recovery of DNA samples. Much of our focus therefore has been on the validity of
the scientific processes and the need for instituting strict regimes to eliminate contamination.
Forensic science has long been thought of as a process starting at the crime scene and ending
in the court. However the LTDNA process starts beyond the scene and includes the manufacture of items for use at the scene
and in the laboratory. There has been insufficient time to make a full comparison of all aspects of LTDNA analysis between
all three forensic science providers. For example no direct comparison has been made of their individual extraction processes
but since these processes have been well documented and validated for many years when employing standard SGM Plus® systems
we do not feel that this is a serious omission. It is assumed that all information provided was accurate, but should further
information be made available we are prepared to reconsider our recommendations.
1.3 Dr Peter Gill who was one of the members of staff of the FSS with whom we spoke in relation to this review,
has from 1st April 2008, become
an employee of the University of Strathclyde.
1.4 With the introduction of the NDNAD, forensic science, as opposed to police science (fingerprints) is, for
the first time, able to provide direction to the investigator. The current process of DNA profiling used in the UK examines
10 hypervariable regions of DNA (short tandem repeats or STRs loci) plus a marker to determine gender. This process is termed
SGM Plus®. The use of short tandem repeats was introduced in 1994 when 4 loci were analysed, this extended to 6 loci plus
the gender test (SGM) and in 1999 the 10 locus test (SGM Plus®) was introduced. These tests were all standardised for 28 cycle
amplification from originally 2 nanogram (ng) of DNA before the increase in sensitivity in 1999 permitted 1ng of DNA as the
standard starting template.
1.5 The introduction of the polymerase chain reaction (PCR) permits the analysis of tiny bloodstains and other
body fluids. The analysis of these PCR products originally took place on gel systems but this latter process has now been
superseded by the use of capillary electrophoresis, which produces a series of peaks referred to as a "DNA profile". This
process is applied routinely to traces of body fluids that have been identified and recovered from clothing and surfaces and
the amplification process in such cases is undertaken 28 times using a pre-prepared SGM Plus® kit. This process has been validated,
is well documented and used routinely both in the United Kingdom and in other countries [1-5].
1.6 The use of 28 cycles was introduced so that STR profiles will be obtained optimally at 1ng of DNA. The commercial
kit (SGM Plus®) is optimally designed to amplify a full DNA profile from 1ng of DNA, although DNA profiles of varying quality
can be generated from less template DNA. This mass (1ng) of DNA equates to approximately 160 human somatic cells. This number
of cells can be visualised as, for example, by a tiny blood spot. The ideal DNA test will detect DNA transferred during an
incident but it should be noted that any item examined may have had DNA transferred prior to the alleged incident (considered
in this review as incidental DNA). Additionally, DNA may be transferred after the alleged incident (termed contamination DNA).
If anticontamination procedures are used it is aimed that 28 cycles of amplification should not generate full DNA profiles
associated with low level contamination.
1.7 Standard DNA profiling which uses 28 cycles works effectively with identifiable traces of body fluids but
there are times when no identifiable body fluid is present. The amount of DNA in these samples may be present at very low
levels perhaps corresponding to one or more human cells. Some of these samples are sometimes referred to as 'touch DNA' and
may be present at levels similar to incidental DNA or that of low level contamination that would not normally be detected
using standard DNA profiling. Modifications to obtain an STR profile from less than 200 picogram (pg) include: optimisation
of the electrophoresis system, increasing the number of amplification steps from 28 to 34 cycles [6] and/or purification of
the PCR product from a 28 cycles process. Any of these modifications results in an increase in the sensitivity of the test
but may also increase stochastic effects and the opportunity for detecting DNA not related to the alleged incident (either
incidental or due to contamination). The stochastic effects include allelic "drop out", random allelic "drop-in" and an increase
in stutter products. These processes confuse the outcome of such DNA profiling and are usually dealt with by repeating the
process a small number of times, usually twice is sufficient. The stochastic effects are not limited to increased cycle number
but occur even with 28 cycles when using low template DNA. With the introduction of more sensitive systems, e.g. capillary
electrophoresis it is possible to detect very low levels of DNA using 28 cycles. Sometimes identifiable cellular material
may be present but either the DNA is degraded such that a full DNA profile cannot be obtained, or the presence of inhibitors
prevents further analysis. Since the aim is to generate as complete a DNA profile as possible, modifications as described
above will be used.
1.8 Additionally, in order to take into account those processes used by some forensic science providers that
do not employ additional cycles of amplification to detect very low levels of DNA we have adopted the following definition
of LTDNA analysis viz.
"An ultra sensitive technique that has the potential to yield a DNA profile from sub-optimal biological samples
e.g. Low Copy Number DNA analysis."
2. Recovery of Samples for Low Template DNA Analysis
2.1 There are two aspects to this process, the first is the recovery of samples, often by Scene of Crime Officers
or Crime Scene Investigators (SOCO/CSI) from the crime scene and the second is the recovery of biological samples by the scientists
in the laboratory from items submitted.
Recovery of Samples from Crime Scenes
2.2 The first of these rests on the decision of the SOCOs/CSIs often in conjunction with the police Senior Investigating
Officer of what samples should be considered for subsequent LTDNA analysis. It is considered important that this type of decision
is taken early in an investigation to reduce the possibilities of contamination of such samples. One view expressed to the
review is that at this stage of an investigation it is not possible to identify such samples and that in anyway DNA samples
would be recovered by a standard procedure irrespective of its importance to the investigation. However, trainers from the
FSS have reported confusion amongst some police force scene personnel as to when and how samples for LTDNA are to be recovered.
This same problem does not appear to have been encountered by some other forensic science providers but these have probably
undertaken less training and are only recently involved directly with crime scene work. As a consequence we do not entirely
support this view but believe that given appropriate training, more focused decisions in respect of recovery of LTDNA samples
can be made and any appropriate rigorous decontamination procedures implemented. Where such decisions become difficult, then
direct contact with laboratory staff needs to be encouraged. Because of the commercialisation of forensic science provision
we are told that some police forces believe that the advice given may be commercially driven and consequently an independent
advice service would be looked upon with favour. If this is accepted then the only potential source for such advice may have
to come through the National Policing Improvement Agency (NPIA) who would be required to provide trained personnel for this
purpose. This suggests that some conformity in approach needs to be established. It is for the Forensic Science Regulator
to institute such training programs and mechanisms for the resolution of these issues.
2.3 A most important aspect of this recovery process is associated with the use of DNA free materials particularly
in respect of swabs and containers. The FSS do provide these under their SceneSafe Kits especially the K555 kit used to recover
‘touch DNA’ such as would occur with finger prints. They have entered into agreement with manufacturers to provide
such DNA-free materials using a specific ethylene oxide decontamination procedure they have developed. We were informed that
batches of these materials are quality controlled (we believe that other forensic science providers should introduce similar
controls over their consumables). Additionally they have generated a database of personnel working for the supply companies
that can be used for elimination purposes. Such precautions, which the review panel consider valuable, do have a cost implication
which may lead some police forces to sanction the use of cheaper substandard kits for such purposes. Other forensic science
providers have taken a different view and have put the onus on the manufacturer to provide DNA clean materials. They have
encountered some minor problems using this approach leading to “drop in “ alleles present in the profiles but
these have not been considered a real problem for interpretation. One of these organisations has looked at the use of ethylene
oxide decontamination and believed it to be a promising process. Neither of these organisations batch quality control their
consumables nor do they provide police forces with DNA recovery kits. A national standard for "DNA clean" recovery materials
needs to be set.
2.4 Some police forces are making use of kits provided by the FSS while others, basically on cost, may choose
to go to alternative suppliers. It is not clear how such forces choosing the latter route determine the suitability of these
consumables. In all cases surveyed it would seem that wet and dry swabbing are the standard procedure employed. It is for
the Forensic Science Regulator in consultation with all parties to establish the standard for DNAfree recovery kits and to
make sure that all police forces employ these. Additionally, such kits must be quality controlled.
2.5 Some forces issue detailed instructions on how to recover LTDNA samples while others only give general guidance.
There seems to be a feeling in some police forces that LTDNA is a panacea for the solution of all crime because of little
understanding of the concepts and making submissions without consideration of contamination issues. Clearly there is a requirement
to communicate the limitations of these procedures.
2.6 It should be noted that failure rates for LTDNA analysis are high. One police force has estimated success
in achieving a full profile at about 6% although this might be a reflection of the substrates sampled. We have sought detailed
information on the success rate of LTDNA analyses but no such information seems to be available. The main reason for this
lies in a lack of agreement on what constitutes a success. One consideration may be those situations where standard SGM Plus®
has failed to produce any useable profile but that LTDNA analysis has produced a full or useable partial profile. It is for
the Forensic Science Regulator to come to an agreement with all parties on what constitutes LTDNA success and to then to institute
an appropriate survey.
2.7 Some police forces are aware that because LTDNA analysis cannot often be related to a particular act then
this limits its usefulness to an investigation. Other forces are not so aware of the difficulties that arise from LTDNA analyses
and an education programme is required to inform and correct any misconceptions. There is some agreement amongst police forces
that national guidance on the use of this technique would be useful since even when the scientific advice is that such samples
may be better analysed by the standard SGM Plus® system or that the item is unlikely to provide any useful data, some police
forces choose to go ahead with such LTDNA analyses. It is for the Forensic Science Regulator to institute appropriate training
programs and to set standards that will enable police forces and their crime scene personnel to have a full grasp of what
constitutes LTDNA analysis, how such samples are to be collected and stored and the likelihood of success.
Isolation of DNA samples in the Laboratory
2.8 The isolation of DNA from exhibits differs little whether standard or low template DNA samples are being
recovered. The main difference is that for the former, automation of the DNA extraction process can be employed while for
the latter manual extraction would be more common. Methods for the recovery of DNA from exhibits are well documented using
a commercial kit from Qiagen or the phenol/chloroform method and have been validated over many years [7-11].
2.9 In relation to LTDNA analyses it is desirable that a decision is made at an early stage as to whether such
a sample submitted to the laboratory would be best analysed by this method. This requires that staff to be suitably trained
for this role and we have been advised that in the FSS such specialised training does take place and that the staff are mentored
after training and their casework peer reviewed. In addition such staff must undergo a specific written test. We are advised
by UKAS that these training programs are of a high standard and that detailed records of conformity to these standards of
performance are maintained. For other providers, for this specialised work we were informed that they have recruited personnel
who already have the necessary skills although we are aware that at least some of these personnel transferred to these organisations
prior to the introduction of LCN DNA analysis by the FSS. For these organisations training programs comparable to those of
the FSS are in place or being put in place for new recruits who do not yet possess the necessary skills. The levels of such
competencies will need to be evaluated by UKAS as part of their accreditation process. It is for the Forensic Science Regulator
to oversee compliance with these standards of competence for LTDNA specialists and when and where appropriate to suggest modifications
to such training programs and record keeping.
2.10 In all laboratories we have visited special precautions have been taken in respect of LTDNA analysis to
minimise the risk of contamination. These include pressurised laboratories, ultraviolet irradiation of benches, a strict bench
cleaning regime and deep cleaning processes. Contamination levels are continuously monitored. We have observed details of
contamination levels and the corrective procedures. All those who work within this environment are suitably dressed with clean
overwear, masks, footwear, and double rubber gloves etc. Maintaining such laboratories is expensive and this will be reflected
in the cost of providing this service. Some more recent providers of this type of service are fortunate in having newly constructed
laboratories of a high standard while the FSS and LGC Forensics has had to adopt much older laboratories and while this does
not mean an inferior service it will probably be more costly to maintain.
3. The
Analytical Process
3.1 Having recovered the low levels of DNA from exhibits it becomes necessary to submit the sample for analysis.
This is a two stage process involving the amplification of DNA from the recovered template DNA using the PCR and finally analysing
this product using capillary electrophoresis. In earlier days and at the commencement of the implementation of LTDNA analyses
the FSS employed gels for the separation of the DNA products.
Quantification of DNA
3.2 The first question that must be resolved is whether or not the isolated DNA (assumed to be <200pg) should
be quantified. Because the amount of DNA is so small the FSS take the view that it is unnecessary to quantify because it uses
up too much DNA in the process thereby reducing the chances of producing a successful profile. Other forensic science providers
routinely quantify the DNA extract believing that in so doing it reduces the chances of generating over amplification of the
PCR products, something the review has observed with some FSS analyses. Additionally these providers believe that this enables
a better estimate of potential inhibition. Unless there is a recognised method for addressing problems relating to over amplification,
the reviewers would favour this second approach. The availability of real time PCR quantification makes this not only feasible
but uses acceptable levels of DNA product leaving sufficient to carry out a full analysis in duplicate with sufficient material
being left over for a third analysis should it become necessary. Further research into the best ways of quantifying very small
DNA samples (for example using a repetitive DNA target) may be indicated. The Forensic Science Regulator should monitor the
use of DNA quantification procedures.
3.3 Two different approaches have been adopted by different forensic science providers to the amplification
process but the aim in each case is the same, namely to boost the signal obtained from the PCR product in relation to the
background noise. The method adopted by the FSS has been to increase the number of PCR cycles from the standard 28 to 34 thereby
producing an increase in the amount of PCR product. From such a process an increase in "drop-out" of the second peak for heterozygote
profiles, or difficulty in allocating the second peak, can occur. Additionally stochastic problems become more acute and problems
can be encountered with the effects of inhibition but the review has been told by the FSS that these effects can be reduced
by dilution. [Rules about inclusiveness have been developed. (See below under interpretation)].
3.4 The method adopted by other forensic providers is to carry out the amplification to 28 cycles and then to
submit the product to a "clean-up" procedure prior to analysis by optimised capillary electrophoresis. This has produced profiles
that are directly comparable with those of the FSS without requiring additional amplification. In an earlier case, an extract
from a tape taken at a crime scene failed to produce a profile of a suspect when analysed by the FSS but that same extract
produced a profile when analysed by one of the alternative suppliers in 2004. The reason for this difference in result seems
to lie in the effect of excessive DNA inhibition on LCN analyses conducted by the FSS which they did not fully recognise until
2003 and did not correct for by dilution of the sample until 2005. Inhibition of the PCR when using the standard 28 cycles
and the SGM Plus® kit was well known in 2001. This is only one example of improved results being obtained from the same sample
by two different forensic science providers.
3.5 The way in which scientific methods often develop is that a process will be introduced that seems to meet
all validation criteria but that after a period of time using this same process it becomes evident that modifications require
to be instituted to optimise the system. This appears to have been the case with the FSS LCN DNA system when it was discovered
in 2003 that inhibition processes were preventing DNA profiles being obtained from some samples. Their solution to this problem
was to dilute the samples thereby diluting the inhibiting factor, including the amount of DNA, and enabling a profile to be
obtained. This discovery meant that the FSS were required to re-test some 5,000 samples. The Reviewers note that there was
a failure on behalf of the FSS to inform the police forces of this finding and the length of time, until September 2005, the
dilution process took to implement. Moreover, since inhibition was at the heart of the problem and one aspect of this was
the presence of excessive amounts of DNA, this should have triggered the thought processes into implementing some quantification
of the extracted DNA prior to amplification and analysis. However, it was after July 2005 that a reliable quantification method
became available. Prior to this only a rough guide as to the amounts of DNA in an extract was obtained from a measurement
of the intensity of a fluorescent dye binding to the DNA fragments. In some cases this may have been sufficient to evaluate
the presence of excessive amounts of DNA and enabled appropriate dilutions to have been made. Unlike all other providers the
FSS do not, even now, routinely quantify their DNA extracts. The Forensic Science Regulator should insist that as a matter
of best practice a DNA quantification step is implemented for all DNA analyses submitted to the CJS and should monitor its
implementation.
3.6 A valuable test would be for a direct comparison study of the two different approaches to LTDNA analyses
to be undertaken. The Review has been informed that a scientific paper by one of the forensic science providers describing
this is expected to be published soon. Whether or not this scientific paper is published the Forensic Science Regulator should
provide a mechanism that, while retaining appropriate confidentiality for the research, he is able to evaluate the information
for all forensic science providers and thereafter take appropriate decisions as to best practice. All providers agree that
regardless of which signal enhancement method is selected, the problems of allelic drop out due to stochastic effects in the
presence of low quantities of template and that of increased noise will occur in sub-optimal DNA samples.
Interpretation of LTDNA Profiles
3.7 An ideal profile, such as could be obtained from fresh blood or a buccal sample, would contain alleles for
each of the loci in the test kit. LCN-like approaches deal with minute samples of DNA. These may also be contaminated with
inhibitors of the PCR, extraneous DNA and be below the threshold of detection by PCR. All approaches to increasing the sensitivity
of detection face the same problems: loss of alleles by chance because they are not amplified from minute starting quantities
(allele drop out); high cycle numbers may increase the risk of polymerase induced “stutter-bands” leading to additional
alleles and detection of low levels of extraneous contamination.
3.8 In summary therefore the main problems arising from LTDNA profiles reside in inhibition, the “drop
out and in” of different alleles, an increase in the number of irrelevant alleles detected and mixture interpretation.
Inhibition is well known to occur with some dyes e.g. from blue denim garments, but there may be other inhibitors including
chelating agents, organic contaminants like haem, and heparin that are difficult to remove [12]. Excessive DNA may also affect
the interpretation of the resulting DNA profile and this is not restricted to LTDNA.
3.9 In respect of "drop-in" alleles these are considered a random and infrequent process associated with contamination
especially of the consumables but also arising from crime scene and laboratory personnel. While laboratory personnel can usually
be eliminated from a DNA profile fairly quickly, the incomplete nature of the Police Elimination DNA database is a problem
and the Forensic Science Regulator needs to pursue this problem with ACPO. "Drop-in" alleles can sometimes be explained by
reference to a manufacturer's and laboratory personnel database, but may also be helped by employing assured DNA free consumables
and reagents. "Drop-out" alleles are also a problem and rules have been developed by the FSS following a study of many profiles
under standard but not LTDNA profiling conditions [13, 14]. Other providers have developed similar guidelines for the interpretation
of potential allelic "drop-out" and "drop-in".
3.10 Efforts have been made to develop a consensus for the interpretation of STR profiles from low template
DNA [15-17]. Peak areas are normally measured in relative fluorescence units (rfu’s) and it has been reported that provided
that the weaker of two peaks for a heterozygote is 60% or greater of the larger peak then it should be included in the profile.
Using 34 cycles, where the relative fluorescent units are greater than 10,000 and there is no other detectable allele, then
this should be considered a homozygote by the FSS, although this view is not accepted by all the providers. These data are
based upon 34 cycle amplification by the FSS and not using the methods adopted by the other two providers.
3.11 All providers agree that regardless of which signal enhancement method is selected, the problems of allele
drop out due to stochastic effects in the presence of low quantities of template and that of increases noise will occur in
sub-optimal DNA samples. The recent publication [18] also contributes to development of quality guidelines. The peer reviewed
scientific paper discussed in the Omagh Bombing case (The Queen v Sean Hoey (Belfast Crown Court 2006) [19] provides a mathematical
model designed to predict the detection of alleles under “normal” and low concentration conditions. Whilst the
paper is insightful it cannot be regarded as definitive and the work needs replication by an independent group before it can
be accepted. It seems, therefore, that the current state of the art with respect to the analysis of low concentration DNA
STR profiles is not yet represented by a legal and scientific consensus, regardless of the quality of the data presented in
a particular case. The lack of clear, explicit consensus reflects the extremely challenging nature of the analysis. At the
same time, it is clear that the need to articulate such a consensus at national and ideally at international level is pressing.
The recent publication [18] is a step in the right direction. As it stands, this represents an agreement from the providers
themselves. It does not include a "user" perspective and needs to actively take account of legal as well as scientific arguments.
Furthermore, it does not go into sufficient detail regarding criteria for data acceptability or how these are achieved.
3.12 The Forensic Science Regulator should develop a consensus from all the forensic science providers in consultation
with all stakeholders on how profiles and mixed profiles are to be interpreted. Once these criteria have been agreed then
the Regulator should monitor their implementation.
Validation
3.13 A key question is whether or not the process(es) involved in LTDNA analyses have been adequately validated
and is such a validation accepted by the international forensic science community. The question that arises from this is "What
is to be expected of a validation process?" Mr Justice Weir, the judge in the Omagh Bombing case, made some useful comments
concerning validation. He said:
“Validation is the process whereby the scientific community acquires
the necessary information to:
• Assess the ability of a procedure to obtain reliable results
• Determine the conditions under which such results can be obtained
• Define the limitations of the procedure
The validation process identifies aspects of a procedure that are critical and must be carefully controlled”
Mr Justice Weir went on to say that:
“The absence of an agreed protocol for the validation of scientific techniques prior to their being admitted
in court is entirely unsatisfactory”
3.14 The Reviewers are entirely in agreement with these statements and seek to assess how far the providers
of LTDNA analyses comply with these. Because science is fundamentally an exoteric process, it is the norm in empirical science
that findings and data are independently replicated prior to widespread acceptance. Lack of refutation is not sufficient of
itself, regardless of the source of the original work. The lack of a funding mechanism to enable this type of scientific enquiry
is a barrier to the process of validation of new approaches. The Forensic Science Regulator should seek funding for independent
research and validation that is open to national competition.
3.15 To provide validation it is normal practice to begin with samples of known provenance and to submit them
to the process and then to see how they comply with the expected outcome. This latter may require a statistical evaluation.
For the LTDNA analyses the parallel would be the taking of a large number of DNA samples of known profiles reflecting the
different alleles at the 10 different loci. These samples would then be serially diluted to provide masses of say 10, 20,
40, 80, 100 and 200pg. Each of these would then be submitted to the extraction procedure, appropriate PCR regime followed
by capillary electrophoresis. The profiles so produced would then be compared with the known profiles of these samples and
an evaluation made in respect of all the parameters discussed above. This process should then be repeated enough times to
obtain a statistically robust measure of the reproducibility of the system. This would address the well-founded criticism
of Mr Justice Weir who commented on how many times the analysis should be performed and who was not convinced by the response
of duplicate analyses when it was demonstrated that a third analysis had provided a 'different' profile. In this context "the
same" does not mean identical since not all alleles may be amplified in the replication process. Casework restrictions in
terms of sample availability will always have to be taken into consideration. Similarly, mixtures of two profiles (or more)
at these same mass levels could be constituted, extracted, analysed and evaluated.
3.16 The review is in receipt of comprehensive documentation from Orchid Cellmark Ltd and LGC Forensics that
relates to their validation processes for LTDNA analyses systems Enhance and DNA senCE respectively. These systems employ
28 cycle PCR followed by a clean-up procedure prior to capillary electrophoresis. This documentation sets out a logical progression
of experiments to determine the limitations of their processes, these include, optimisation of the capillary electrophoresis
system, limits of detection, the reproducibility and evaluation of stochastic effects, drop in and out and inhibition in a
way very similar to the pattern of experiments the review advocates above. These experiments represent a true internal validation
which UKAS have accepted as complying with ISO 17025. Additionally, these organisations have effectively validated, in a similar
way, LCN using 34 cycles as used by the FSS. These last experiments serve as independent confirmation of the system employed
by the FSS, although it must be recognised that the FSS did not quantify the DNA prior to analysis.
3.17 The FSS initially provided the review with documentation in support of a validation process from the early
years of LTDNA analyses but this documentation was mainly centred on published scientific papers especially the papers of
Gill et al [16] and Whittaker et al [20]. This documentation did not follow what we would consider the normal approach to
a validation process such as that submitted by the other forensic science providers but we were assured by the representative
of UKAS that in order to comply with ISO 17025, the FSS must have provided them in 2000 with material of the kind expected
and as the use of the technique developed, additional criteria would have had to have been met. In order to assess the validity
of these publications we have examined in detail a sample of the raw experimental data associated with this work. While always
responding to our requests for information in a timely manner the FSS found difficulties, even after providing the validation
model illustrated above, in providing what was required. We eventually received from the FSS experimental data setting out
a validation process comparable to the suggested model and including reproducibility information, part of which is from a
previous publication [16] and some of which appears as an in-house study. In our opinion, collectively the documentation and
data provided represent a validation of LCN DNA analyses. We find this inability to respond quickly to our original requests
difficult to rationalise because we were assured by UKAS that copies of the validation procedure accepted by them (UKAS) as
conforming to ISO 17025 should always be kept by the accredited laboratory and should be readily available for inspection.
The fact that all three organisations have independently demonstrated the viability of LTDNA analyses employing 34 cycles
demonstrates that this technique is fit for purpose although the other two providers use this process as a last resort preferring
their own modified 28 cycle systems. The FSS has undertaken to develop and align its validation procedures with the recommendations
of ENFSI and SWGDAM. The Forensic Science Regulator should institute a regular program of inspections of documentation associated
with all validations.
3.18 We note that an additional validation process was undertaken in 2007 by Orchid Cellmark following developed
validation studies over the period 2000 to 2005. LGC Forensics validation on the 28 cycle (DNA SenCE) was from 2001 onwards
and the FSS validation studies were conducted from 1999 onwards.
3.19 There is not currently one standard method employed by these three organisations to achieve DNA typing
of these sub-optimal DNA templates. The term LCN was one adopted by the FSS for their own methodology but any differentiation
from this process, such as that conducted by the other two forensic providers, can also be termed low template DNA typing
(LTDNA).
3.20 External validation can only be achieved if the process is accepted by the wider scientific community.
The International Society for Forensic Genetics has published guidelines for sub-optimal DNA testing [21]. The use of LTDNA
in the international forensic science community has been limited until recent times with only the main laboratory in the Netherlands
(NFI) having implemented the LCN 34 cycle DNA typing method in Europe. Other forensic science providers who use 34 cycles
routinely are: Environmental Science and Research Ltd (ESR) in New Zealand and the Office’s of the Chief Medical Officer
in Switzerland and New York State. Similar systems of amplification of sub-optimal DNA operate in Bosnia-Herzegovina, Spain,
Italy and Germany The review is in possession of details provided by the Dutch laboratory which demonstrate an independent
validation of the LCN DNA process similar to that implemented by the FSS. This process is accepted by the Dutch Criminal Justice
System. Such a validation shows international acceptance of the principal of using 34 cycle processes. Additionally, while
some countries do not operate such a system of analysis, LTDNA evidence is accepted in their courts.
3.21 International acceptance requires the implementation of the technique by countries other than the United
Kingdom. Some of those countries who have presently implemented such a scheme as detailed above, and while not large in number,
supports limited international recognition. This does not in itself negate the technique as being scientifically justified
and as being used in the United Kingdom CJS. Scientific papers [17, 21] provided by the FSS indicate that a consensus accepting
the general process leading to profiles generated by a validated LTDNA regime has been accepted by the international community
such as ENFSI and the European DNA Profiling Group (EDNAP) but these agreements do not provide experimental details of the
initial validation process. Nor is it clear if or how they address alternative proposals for dealing with incomplete data.
4. Transfer of Cellular Material
4.1 Research has discovered that individuals vary in their propensity to transfer their cellular material to
an object. Further factors that affect the transfer of cellular material include time and pressure. Transfer of cellular material
from a person to an object is termed primary transfer. Once transferred the cellular material may be transferred to another
surface if direct contact is made between these two surfaces. This secondary transfer is dependent upon the same factors of
time and pressure. Additionally the nature of the two surfaces will affect whether cells transfer readily or remain predominantly
on the primary surface. Transfer from a primary surface to a secondary surface will not be 100% and therefore there will be
less cellular material transferred to the secondary surface than that on the primary.
4.2 There was a caveat in the LCN DNA profiling reports by the FSS which stated that any cellular material generating
a result using LCN may have transferred by means unknown. This caveat no longer seems to be stated in recent witness statements
relating to touch DNA. LTDNA is used on bone samples, teeth and material removed from microscope slides, in which case the
cell type is either known or inferred. There may be good reason for this omission arising from additional research, but any
such work has not been made available to the Reviewers. The Forensic Science Regulator should encourage openness in the availability
of information that may have an impact on the way DNA profiles are interpreted in the context of a case.
5. Persistence of Cellular Material
5.1 Once cellular material has been transferred from an individual to an inert surface the cellular material
will start to be degraded by cellular enzymatic actions. The speed of this degradation is affected by temperature and the
presence of moisture. Due to the large number of factors that affect the degradation of cellular material, and in turn DNA,
it is unusual to be able to comment upon the time that the cellular material was deposited unless the cells are associated
with a particular stain type. The nature and reason for using LCN precludes such association unless the DNA can be associated
with discrete items such as bones, teeth and degraded bloodstains.
6. The Omagh Bombing Trial
6.1 Following the initiation of this review by the Forensic Science Regulator based upon a high profile case,
it became known that there was likely to be difficulties associated with LTDNA analyses in respect of the Omagh Bombing case.
The reviewers felt it desirable to study the evidence given by forensic scientists during the trial in order to evaluate whether
or not additional problems associated with the use of LTDNA could be identified.
6.2 Much of the focus of the trial was on the validity of LCN DNA analyses as a technique and as to whether
or not the process had been properly validated both internally and externally. The approach that seems to have been adopted
by the FSS as described in court was to see the LCN technique as an extension of the established and internationally validated
standard procedure using 28 cycle PCR methodology. This validated procedure was then extended by two main publications in
Forensic Science International (15,16). It was declared that LCN was a technique that was "reliable, robust, reproducible
and fit for purpose".
6.3 The validation was severely criticised by the defence in terms of reproducibility, lack of quantification,
interpretation and the establishment of guidelines. These areas of criticism have been dealt with in discussions detailed
above. Contamination was also an issue especially any arising from imperfect packaging and storage. While in the laboratory
contamination was under control, the FSS accepted that insecure packaging could lead to contamination problems in relation
to traces of DNA.
6.4 Since the judicial decision in this case the FSS have presented additional experimental data to the review
and we have accepted these as supporting a proper validation of this methodology. The Forensic Regulator will be best placed
to review any future validation studies.
7. LCN in the Criminal Justice System
7.1 The increase in the cycle number from 28 to 34 as employed by the FSS, or the 28 cycle plus clean-up as
used by other providers has led to the opportunity to obtain DNA profiles that may not otherwise be generated. With the increase
in the sensitivity of the methods come additional interpretation problems and stochastic effects.
7.2 Since the introduction of LCN DNA profiling by the FSS in 1999 to the CJS there had been limited challenge
to the process in the UK until the Omagh Bombing case. Questions were raised by Mr Justice Weir as to the validity and reliability
of the LCN technique. It is our opinion that LCN and LTD are extensions of the internationally accepted process of standard
DNA profiling. Since 1999 there have been a number of advances that have increased the sensitivity of DNA testing such that
full DNA profiles using SGM Plus® are possible from less starting material. One consequence has been the increase in the number
of mixed DNA profiles and low level minor profiles from cellular material that may not have been detected previously. The
forensic science community in the UK has developed interpretation guidelines to assist with these types of cases [18, 21].
7.3 Initially it was our opinion that there remained some ambiguity as to the extent of validation conducted
by the FSS with regard to the LCN technique. These reservations have been allayed from a study of the raw data produced by
the FSS that has been provided to the review, recent experimental work conducted by the FSS and also from detailed information
submitted by the other forensic science providers which clearly demonstrate the soundness of LTDNA analysis (including LCN)
providing all the appropriate conditions are met. It appears that it is the position of the FSS and UKAS that the technique
of LCN is fit for purpose and we would wish to support this view. The LTDNA/LCN DNA technique conversely has yet to be implemented
widely by the international forensic science community.
7.4 It is our opinion that any LTDNA profile should always be reported to the jury with the caveats: that the
nature of the original starting material is unknown; that the time at which the DNA was transferred cannot be inferred; and
that the opportunity for secondary transfer is increased in comparison to standard DNA profiling. There may perhaps be some
exceptions (see section 4.2 above).
7.5 It is our opinion that when DNA profiles match as a result of LCN DNA profiling, the significance of the
match should be reported on the probability that the two DNA profiles match only [6]. As the results were obtained from LCN
it is inappropriate to comment upon the cellular material from which the DNA arose or the activity by which the DNA was transferred.
8. General Discussion
8.1 Forensic DNA analysis is a huge asset to national and international law enforcement, yet at the extremes
of its applications there are limitations and the boundaries of these limits are now under critical review. This is entirely
appropriate and timely. It should not be regarded as a threat to service delivery and does not undermine the basic principles
of forensic DNA analysis. Nevertheless the challenges in terms of statistical interpretation of the data and in communicating
them to a largely innumerate criminal justice system should not be under-estimated, nor should the importance of earning and
maintaining public confidence in the system. This will be best achieved by encouraging open review, debate and investigation
of the process.
8.2 It is not our purpose to be too prescriptive in our recommendations but rather advisory since it is important
that this review does not restrict innovation but encourages research and further developments. To this end it is important
that a pool of research funding, open to all by direct competition, is made available since at present such funding is not
accessible through normal government funded science research bodies.
8.3 All forensic science providers have expressed to us their concern over the quality of police forensic science
laboratories established by the larger police forces as a financial saving and used for screening of items to establish what
to submit to the established forensic science providers. Such laboratories have implications for contamination issues relating
to all types of trace evidence, including DNA. They also present problems for interpretation when only selections from specific
exhibits are submitted to forensic science providers. Under these conditions no-one has a complete overview of a case and
this can lead to future problems. Finally, such laboratories do not operate a properly constituted Quality Assurance system
and are not accredited to ISO 17025 standards but should be. This position needs to be remedied. The Forensic Science Regulator
needs to enter into a dialogue with ACPO as to the way such laboratories are to be integrated into the scheme of forensic
science provision.
8.4 One forensic provider commented that, in their opinion, some police forces do not allocate sufficient resources
to ensure the delivery of high quality forensic services. It was suggested that, if this is a problem, spending on forensic
services should be ring-fenced. The Forensic Science Regulator may wish to explore the validity of this comment, although
we understand that the overall spend on forensic services does already account for a sizeable proportion of a force's budget
and that, as is the nature of such things, a force's spend on forensic services does have to compete with spending on other
policing priorities. Furthermore, increased competition in the procurement of forensic services should help to ensure that
police forces obtain additional value for money.
9. Conclusions
9.1 DNA profiling using STR variants is a well established and robust method. Low template DNA profiling techniques
pose additional challenges for data analysis in terms of allele dropout, band stutter and allele gains. Many services have
moved into the low template domain by dint of improved technology such as capillary electrophoresis without necessarily being
explicit that this is the case.
9.2 The scientific validation of the approach is complete although developing aspects of evidence interpretation
are still on-going. This needs to be continued with the explicit goal of delivering clear, easily understandable guidelines
for 9.3) and 9.4) below. It is however, our belief that the organisations providing such a DNA service in terms of LT DNA
analysis have complied with a normal in house validation process based upon sound scientific principles. This has been seen
by UKAS as complying with the ISO 17025 standard and we would endorse this. For completion, we would normally wish to see
such validation independently ratified by another such laboratory to conform to normal scientific practice of repeatability
by an independent laboratory. The commercialisation of forensic science makes this very difficult and it is only through this
review that we have seen that this has been effectively achieved by the FSS’s commercial rivals. This should perhaps
be a driver for the establishment of a professional organisation of all forensic science providers where cross validation
can be discussed. The Forensic Science Regulator should explore the means of establishing a professional forensic science
provider’s organisation in order to develop mutually agreed standards.
9.3 Interpretation of the results is complex for two reasons: the statistics are challenging and probably hard
to comprehend by a non-specialist and the decision how and when to apply certain statistical methods has not yet reached a
clear consensus, such that the methods applied by laboratory "a" can be directly compared to the methods applied by laboratory
"b", e.g. New York and UK laboratories. This is bound to create confusion in a court setting and a clear protocol to avoid
this confusion is required.
9.4 National minimum technical standards for extraction, quantification/dilution and interpretation criteria
need to be agreed by not only the providers, NDNADB and UKAS but also by users including the police and the criminal justice
system. As the providers have made significant progress in this direction already, it may be possible to achieve this in a
few months. This would be a highly desirable outcome. The Forensic Science Regulator needs to coordinate all the information
already available that is associated with these techniques and by agreement with all stakeholders establish appropriate standards.
9.5 DNA technology continues to advance rapidly and there needs to be a mechanism for horizon scanning and for
the funding of ambitious, servicedriven scientific and technical development to enable the UK to maintain its pre-eminent
position in forensic DNA analysis.
10. Recommendations
1. For SOCOs/CSIs and SIOs, there needs to be a national education programme setting out the advantages and
limitations of LTDNA in order to establish a conformity of approach to crime scene work. From this should be developed national
guideline documentation. It is for the Forensic Science Regulator to institute such training programs and mechanisms for the
resolution of these issues.
2. It is for the Forensic Science Regulator to come to an agreement with all parties on what constitutes LTDNA
success and to then to institute an appropriate survey.
3. It is for the Forensic Science Regulator to institute appropriate training programs and to set standards
that will enable police forces and their crime scene personnel to have a full grasp of what constitutes LTDNA analysis, how
such samples are to be collected and stored especially in relation to issues of contamination and the likelihood of success.
4. The Forensic Science Regulator should monitor the use of DNA quantification procedures.
5. We have been told that there is an urgent need for the DNA profiles of all serving operational police officers
and crime scene personnel to be included on the Police Elimination Database and for forensic science providers to have direct
access to it as a means of eliminating irrelevant DNA profiles. While laboratory personnel can usually be eliminated from
a DNA profile fairly quickly, the incomplete nature of the Police Forces DNA database is a hindrance and the Forensic Science
Regulator needs to pursue this problem with ACPO. As an alternative, financial support needs to be provided to enable the
DNA profiles of Police Officers and crime scene personnel involved in a specific investigation to be obtained at the same
time as the suspect samples.
6. A national standard needs to be established for "DNA clean" consumables, especially in relation to crime
scene recovery kits. The Forensic Science Regulator should ensure that only kits which meet such a standard should be used
by police forces.
7. The Forensic Science Regulator should ensure the batch testing of all DNA reagents to ensure that they are
DNA free prior to their use.
8. It is for the Forensic Science Regulator to oversee compliance with standards of competence for LTDNA laboratory
specialists and when and where appropriate to suggest modifications to such training programs and record keeping.
9. Those police forces that have made the decision to carry out preliminary forensic testing by the establishment
of a police forensic science laboratory must have such laboratories accredited to a standard comparable to those of forensic
science providers and should comply with ISO 17025 through UKAS. The Forensic Science Regulator needs to enter into a dialogue
with ACPO as to the way Police laboratories are to be integrated into the scheme of forensic science provision.
10. For all LTDNA samples and taking into account the limitations of the amount of DNA extracted from crime
samples, quantification of the material extracted for analysis must be undertaken. Satisfactory commercial kits are now available
for this purpose. Further research is required into the best ways of quantifying very small samples of DNA such as using repetitive
DNA target. The Forensic Science Regulator must insist that as a matter of best practice a DNA quantification step is implemented
for all DNA analyses submitted to the CJS and should monitor its implementation.
11. There needs to be a national agreement on how LTDNA profiles are to be interpreted especially in relation
to "allele drop in and out", stochastic effects, inhibition, and mixtures. This should be aided by regular circulation of
appropriate test profiles and interpretation by ALL providers of this service and any results should be coordinated through
the forensic science regulator. The Forensic Science Regulator should develop a consensus from all the forensic science providers
in consultation with all stakeholders on how profiles and mixed profiles are to be interpreted. Once these criteria/standards
have been agreed then the regulator should monitor their implementation. The Forensic Science Regulator should encourage openness
in the availability of information that may have an impact on the way DNA profiles are interpreted in the context of a case.
12. The Forensic Science Regulator should institute a regular program of inspections of documentation associated
with all validations.
13. Appropriate caveats should be stated in witness statements/court reports, in most instances, when LTDNA
analyses have been undertaken.
14. Any new methods of analysis used by a forensic science provider that will result in the presentation of
evidence to the courts must be validated using appropriate and sound internationally recognised scientific principles. The
details of such validation, including copies of raw data, should be lodged with the forensic science regulator before it is
introduced into service. At least once a year or when the regulator decides it appropriate, such validations will be reviewed
by an independent internationally recognised expert panel the composition of which will be determined by the regulator.
15. An independent study should be undertaken to assess the advantages and disadvantages of the two different
approaches to LTDNA analysis. The Review has been informed that a scientific paper by one of the forensic science provides
describing this is expected to be published soon. Whether or not this scientific paper is published the Forensic Science Regulator
should provide a mechanism that, while retaining appropriate confidentiality for the researchers, enables other providers
to evaluate the research.
16. Improve existing guidelines and standards. Active development of a consensus approach to the analysis of
partial or contaminated DNA profiles is already underway and needs further work and an inclusive structure that takes account
of all of the stakeholders. National and international providers have led the way, but this now needs to include close consultation
with users. Within the UK a steering or advisory group comprising providers, users and independent legal advice, with perhaps
lay representation should endeavour to develop documentation that would guide the courts in the interpretation of evidence.
Educationalists and users should evaluate its comprehensibility and review it in a timely manner in the light of legal precedent
and scientific advances.
17. The use of STR's clearly has limitations and is essentially 15-year-old technology. The world of genetics
and genome analysis has moved on a great deal since then and there may be benefits to be had from alternative technologies.
For example, the world of forensic archaeology has been transformed by the use of next generation sequencers [22, 23] and
it seems likely that these could have a huge impact on forensic DNA analysis. Furthermore PCR and other amplification technology
has improved, with the development of emulsion PCR [24, 25] that may have substantial advantages and enable backward compatibility
with the existing STR database. At present any service developments would have to be recovered as a service cost overhead.
This would preclude radical advances requiring substantial funding. These developments may require high-level academic input
and a competitive funding mechanism similar to those used by the Research Councils. Opportunities to tap into the international
expertise of the Wellcome Trust Sanger Institute, a world leader in DNA sequencing technology, could be investigated. A UK
working group focussed on such developments should develop an option appraisal. The risk of not doing this is the stagnation
and decline of standards on forensic DNA analysis, whereas a successful programme would secure a world lead for the UK.
18. An open funding mechanism needs to be put in place that will support an independent validation process of
new developments. The Forensic Science Regulator should seek funding for independent research and validation that is open
to national competition.
19. National minimum technical standards for extraction, quantification/dilution and interpretation criteria
need to be agreed by all forensic science providers. These standards should also be agreed by the Forensic Regulator's Forensic
Science Advisory Council. The Forensic Science Regulator needs to coordinate all the information already available that is
associated with extraction etc. techniques and by agreement with all stakeholders establish appropriate standards.
20. Since this matter of financing forensic science has been brought to our attention by the forensic science
providers we believe it is for the Forensic Science Regulator to quantify this problem and to explore mechanisms to correct
any problems his inquiry may reveal.
21. The Forensic Science Regulator should explore the means of establishing a professional forensic science
provider’s organisation in order to develop mutually agreed standards.
11. References
1. Cotton, E.A., et al., Validation of the AMPFlSTR(R) SGM Plus(TM) system
for use in forensic casework. Forensic Science International, 2000. 112(2-3): p. 151-161.
2. Gill, P., et al., Considerations from the European DNA profiling group
(EDNAP) concerning STR nomenclature. Forensic Science International, 1997. 87(3): p. 185-192.
3. Tully, G., et al., Considerations by the European DNA profiling (EDNAP)
group on the working practices, nomenclature and interpretation of mitochondrial DNA profiles. Forensic
Science International, 2001. 124(1): p. 83-91.
4. Gill, P., et al., The evolution of DNA databases--Recommendations for
new European STR loci. Forensic Science International, 2006. 156(2-3): p. 242-244.
5. Dixon, L.A., et al., Analysis of artificially degraded DNA using STRs
and SNPs--results of a collaborative European (EDNAP) exercise. Forensic Science International,
2006. 164(1): p. 33-44.
6. Gill, P., Application of Low Copy Number DNA Profiling. Croatian Medical Journal, 2001. 42(3): p. 229-232.
7. Maniatis, T., E. Fritsch, and J. Sambrook, Molecular Cloning: A Laboratory
Manual. 1982: Cold Spring Harbor Press.
8. Montpetit, S., I. Fitch, and P. O'Donnell, A simple automated instrument
for DNA extraction in forensic casework. Journal of Forensic Sciences, 2005. 3: p. 555-563.
9. Castella, V., et al., Forensic evaluation of the QIAshredder/QIAamp DNA
extraction procedure. Forensic Science International, 2006. 156(1): p. 70-73.
10. Greenspoon, S., et al., QIAamp spin columns as a method of DNA isolation
for forensic casework. Journal of Forensic Sciences, 1998. 5: p. 1024-1230.
11. Weiner, M., S. Gabriel, and J. Stephens, Genetic Variation: A Laboratory
Manual. Cold Spring Harbour Laboratory Press, 2007.
12. Moreira, D., Efficient removal of PCR inhibitors using agaroseembedded
DNA preparations. Nucleic Acids Research, 1998. 26(13): p. 3309-3310.
13. Collins, P.J., et al., Developmental validation of a single-tube amplification
of the 13 CODIS loci, D2S1338, D19S433, and amelogenin: the AmpFISTR Identifier PCR Amplification Kit. Journal
of Forensic Sciences, 2004. 49: p. 1265-1275.
14. SWGDAM, Short tandem repeat (STR) interpretation guidelines. Forensic Science Communications, 2000.
15. Whitaker, J., E. Cotton, and P. Gill, A comparison of the characteristics
of profiles produced with the AMPFlSTR(R) SGM Plus(TM) multiplex system for both standard and low copy number (LCN) STR DNA
analysis. Forensic Science International, 2001. 123(2-3): p. 215-223.
16. Gill, P., et al., An investigation of the rigor of interpretation rules
for STRs derived from less than 100 pg of DNA. Forensic Science International, 2000. 112(1): p. 17-40.
17. Dixon, L.A., et al., Validation of a 21-locus autosomal SNP multiplex
for forensic identification purposes. Forensic Science International, 2005. 154(1): p. 62-77.
18. Gill, P., et al., National recommendations of the Technical UK DNA working
group on mixture interpretation for the NDNAD and for court going purposes. Forensic Science
International: Genetics, 2008. 2(1): p. 76-82.
19. Gill, P., J. Curran, and K. Elliot, A graphical simulation model of
the entire DNA process associated with the analysis of short tandem repeat loci. Nucleic Acids
Research, 2005. 33(2): p. 632-643.
20. Whitaker, J.P., E.A. Cotton, and P. Gill, A comparison of the characteristics
of profiles produced with the AMPFlSTR(R) SGM Plus(TM) multiplex system for both standard and low copy number (LCN) STR DNA
analysis. Forensic Science International, 2001. 123(2-3): p. 215-223.
21. Gill, P., et al., DNA commission of the International Society of Forensic
Genetics: Recommendations on the interpretation of mixtures. Forensic Science International,
2006. 160(2-3): p. 90-101.
22. Green, R.E., et al., Analysis of one million base pairs of Neanderthal
DNA. Nature, 2006. 444(7117):
p. 330-336.
23. Margulies, M., et al., Genome sequencing in microfabricated highdensity
picolitre reactors Nature, 2005. 437(7057):
p. 376-380.
24. Wetmur, J.G., et al., Molecular haplotyping by linking emulsion PCR:
analysis of paraoxonase 1 haplotypes and phenotypes. Nucelic Acid Research, 2007. 33(8): p. 2615-2619.
25. Hori, M., H. Fukano, and Y. Suzuk, Uniform amplification of multiple
DNAs by emulsion PCR. Biochem Biophys Res Commun, 2007. 352(2): p. 323-328.
12. Acknowledgements
The authors of the review would wish to acknowledge the help and full cooperation of the following organisations
and personnel without which this review could not have been written:
The Forensic Science Service Ltd
Orchid Cellmark Ltd
LGC Forensics
Harperley Hall Training Centre, NPIA, Mr Keith Fryer and Mr Duncan Brown
Gary Pugh of the Metropolitan Police
The Metropolitan Police Focus Group
United Kingdom Accreditation Service, Dr Jane Beaumont
Custodian of the NDNAD, Dr Michael Prior
Greater Manchester, Hampshire, Lincolnshire and West Yorkshire Police Forces
The Dutch National Forensic Science Laboratory.
13. Glossary
