USE OF LOW COPY NUMBER (LCN) DNA IN FORENSIC INFERENCE

C. Murray, et al.
The Forensic Science Service, London, UK
 

Published Online: 2001

ABSTRACT

Development of a low copy number DNA profiling method has allowed the detection of very low levels of DNA. However, issues concerning the transfer and persistence of LCN DNA have required further research to be carried out. This has included studies of individuals' tendency to deposit and transfer DNA onto surfaces they have touched. Additionally, the effect on DNA of some of the enhancement treatments used on latent finger marks has been tested.

INTRODUCTION

Low copy number (LCN) DNA profiling using the SGM Plus multiplex (1) has been used in casework since January 1999. The sensitivity of the profiling process has been increased by raising the number of PCR cycles from 28 to 34. This allows DNA to be detected at much lower quantities, even down to the level of a single cell (2). Hence, strict guidelines are in place regarding the collection and processing of samples, and also the interpretation of results (3,4).

The LCN method can be used to analyse samples which have simply been touched; this follows findings reported by workers (5), who used 28 PCR cycles to obtain genetic profiles from objects such as a telephone handset, pens and briefcase handles. The authors also reported that the amount of DNA recovered varied depending on the individual.

However, profiles obtained using LCN DNA analysis often originate from samples that show no visible source of biological material, consequently no presumptive test is available for contact traces. The question arises whether the profile is actually relevant to a case. Specific caveats are written into the court statements, pointing out that it is not possible to make conclusions about where the DNA originated, when it was deposited, or about the transfer and persistence of DNA. Further research into LCN DNA is driven by the need to answer these issues.

Questions regarding transfer have been previously investigated (again using 28 PCR cycles) by Ladd et al. (6) who tested whether secondary transfer of DNA could occur from individual A to individual B and then onto an object, or from individual A to an object and then onto individual B. They found some minor peaks attributable to secondary transfer but concluded that this was not likely to be an issue when presenting analysis results to a court. Van Oorschot and Jones reported that they had observed secondary transfer, as well as the persistence of DNA on experimental items for up to 84 days and on a casework item (a glove) for two years (5,7).

Finger marks found at a crime scene potentially offer two highly discriminating forms of evidence. Where marks are not smeared there will be the fingerprint ridge detail and the DNA of the blood or skin cells that have been transferred to a surface on deposition of the mark. Marks are usually enhanced using light sources and chemicals to allow greater visualization, but a dilemma may be faced by investigators who also wish to obtain a DNA profile. It becomes necessary to be aware of any detrimental effects that enhancement methods may have on DNA profiling.

Previous research has been carried out in this field, predominantly into the effect of chemical treatments on marks made in blood. Several authors have reported that cyanoacrylate (CNA), when used to enhance bloody finger marks, has no detrimental effect on subsequent DNA analysis (8, 9, 10, 11). Zamir et al, 2000 have reported that CNA does not effect STR profiling of latent marks. The effect of dactyloscopic powders on DNA analysis has also been investigated (10, 11, 13). While powders such as White and Black powder were not found to inhibit DNA processing, metallic powders were found to limit the amount of DNA that could be recovered and profiled from latent marks (14). Two chemical treatments often applied to marks on paper are ninhydrin and 1, 8 diazafluoren-9-one (DFO). These enhancers have been observed to have little adverse effect on DNA profiling from marks in blood (11, 15), for ninhydrin this applies even when the mark has been left in its enhanced state for up to 56 days (16). Physical developer and iodine are also often applied to enhance marks on paper, both have been reported to degrade DNA in latent and bloody marks respectively (17, 8).

This document examines further studies on shedder type (the tendency of an individual to deposit his/her DNA profile on a touched surface), transfer and persistence of LCN DNA, and discusses research that has been carried out to determine the effects of various enhancement methods on the DNA integrity of latent finger marks.

SHEDDER INDEX

A group of 29 people were tested for their ability to deposit their DNA profile onto touched objects. It was found that a typical good shedder leaves a complete profile on the surface of a plastic tube after contact of only 10 seconds, whereas at the other end of the scale a poor shedder will leave only a few alleles, possibly with several loci dropping out completely. The results from this group suggest that with the collection of data from more individuals, a continuous distribution would result.

PRIMARY TRANSFER

Work was carried out to determine whether DNA profiles could be obtained from clothing; specifically, plain white t-shirts. After 8 hours wear, more of the wearer's DNA was recovered from the front of the t-shirt than the back. Targeting the neck area maximized the chance of obtaining a useful result. In a series of simulated assaults, where one person grabbed the shoulder of another for a period of 30 seconds, mixed profiles were obtained from the grabbed area of the t-shirts. The "assailant" always contributed the major component to this mixture, regardless of his/her shedder type.

SECONDARY TRANSFER

Experiments were carried out to determine whether it was possible for individual A to transfer his DNA to individual B through contact, who could in turn transfer individual A's DNA onto an object. We began with a scenario which was most likely to yield a result: a good DNA shedder (A) shook hands with a poor shedder (B), who then gripped a plastic tube for 10 seconds. The results from swabs of the tubes showed that on five separate occasions all of the good shedder's profile was recovered, with none of the poor shedder's alleles appearing.

The experiment was then repeated, but with the introduction of a delay of 30 minutes between the time of the handshake and the tube-holding. The results indicated that although the poor shedder deposited some alleles, secondary transfer of the good shedder's DNA still occurred.


PERSISTENCE

Many factors may affect the persistence of low level DNA; time, temperature, humidity, etc. While it is unreasonable to test every combination of variables, some generic experiments have been undertaken and certain scenarios addressed.

A time-study of the persistence of DNA is currently underway, where touched items have been stored at room temperature and tested to find out how much DNA can be recovered after certain periods of time. Full profiles were still recovered from surfaces touched by a good shedder -- even after 4 months. Whereas a marked decrease in the recovery of the poor shedder's DNA was observed.

An exchange of identical wrist-watches between certain shedder types was carried out to ascertain the period of time needed for the original wearer's DNA profile to be replaced by that of the new wearer. Generally we found that a good shedder completely replaced the original wearer's profile in 2-3 weeks, and after only a few days had become the major component of a mixture. An example of this is shown in Figure 4. In contrast, a poor shedder typically took around 2 weeks just to comprise the major component.

ENHANCED LATENT FINGER MARKS

The effect of various treatments used to enhance latent marks on either porous or non-porous surfaces was investigated. The finger marks used for these analyses were all deposited by the same individual on either acetate or paper for non-porous and porous surfaces respectively. The chemical treatments tested were CNA (in and out of a vacuum), aluminium powder, metal deposition, DFO, ninhydrin and physical developer. The effect of these chemicals on STR profiling was observed on freshly enhanced marks and marks that had been left for 100 days before DNA analysis was carried out.

It was observed that overall, better recovery of DNA was possible from marks deposited on the non-porous surface. Marks enhanced with CNA, aluminium powder and metal deposition yielded full DNA profiles when DNA processing was carried within a week of treatment. However, recovery of DNA decreased when marks had been left in the enhanced state for 100 days. While some of the drop in DNA recovery can perhaps be accounted for by general degradation, the inability to recover any alleles from the marks treated with metal deposition suggests that some chemicals do radically effect DNA over time.

A similar result was obtained from ninhydrin treated marks. Further observation of the results appears to indicate that the recovery of DNA from vacuum CNA treated marks increases over time post enhancement. However, this is unlikely to be the case. The reason for improved recovery may be due to two different pieces of vacuum CNA equipment being used for each sample set. Slight modifications to enhancement methods have been shown previously to improve recovery of DNA from marks in blood.

An additional experiment using aluminium powder was also carried out. The relative levels of profile recovery from powdered marks in situ and from tape lifts of the marks were investigated. It was found that equal amounts of DNA could be recovered from the mark in situ and the lifted mark, both yielding approximately 70% of the fingerprint donors profile. This finding was important as it suggested that a mark could be lifted and preserved on tape for ridge detail analysis while the original deposit could be swabbed for DNA.

~~~~~~~~~~~~~

REFERENCES

1. Cotton E.A., et al.
Validation of the AMPFlSTR SGM Plus system for use in forensic casework
Forensic Sci. Int., Vol. 112, p.151 (2000)

2. Findlay, I., et al.
DNA fingerprinting from single cells
Nature, Vol. 389, p. 555 (1997)

3. Gill, P., et al.
An investigation of the rigor of interpretation rules for STRs derived from less than 100pg of DNA.
Forensic Sci. Int., Vol. 112, p.17 (2000)

4. Whitaker, J.P., Cotton, E.A., Gill, P.
A comparison of the characteristics of DNA profiles for standard STR and low copy number (LCN) STR analysis
Forensic Sci. Int., Vol. 52, p. 345 (2001)

5. van Oorschot, R.A.H., Jones, M.K.
DNA fingerprints from fingerprints
Nature, Vol. 387, p.767 (1997)

6. Ladd, C., et al.
A systematic analysis of secondary DNA transfer
J. Forensic Sci., Vol. 44, p.1270 (1999)

7. van Oorschot, R.A.H., et al.
Retrieval of DNA from touched objects
In: Proc. 14th Int. Symp. Forensic Sci. (1998)
Australian / New Zealand Forensic Science Society

8. Lee, H.C., et al.
The effect of presumptive test, latent fingerprint and some other reagents and materials on subsequent serological identification, genetic marker and DNA testing in bloodstains
J. Forensic Ident., Vol. 39, p.339 (1989)

9. Shipp, E., et al.
Effects of argon laser light, alternate source light, and cyanoacrylate fuming on DNA typing of human bloodstains
J. Forensic Sci., Vol. 38, p.184 (1993)

10. Newall, P.J., et al.
Successful amplification and STR typing of bloodstains subjected to fingerprint treatment by cyanoacrylate fuming  Can. Soc. Forens. Sci. J.,  Vol. 29, p.1 (1996)

11. Roux, C., Gill, K., Sutton, J., Lennard C.
Effect of fingerprint enhancement techniques on the DNA analysis of bloodstains
J. Forensic Ident., Vol. 49, p. 357 (1999)

12. Zamir, A., Springer, E., Glattstein, B..
STR typing of DNA extracted from adhesive tape after processing for fingerprints
J. Forensic Sci., Vol. 45, p.687 (2000)

13. Van Rentergeum P., Leonard, D., De Greef C.
Use of latent fingerprints as a source of DNA for genetic identification.
In: Progress in Forensic Genetics 8. Elsevier Science, p.501 (2000)

14. Van Hoofstat, D.E.O., et al.
DNA typing of fingerprints and skin debris:
Sensitivity of capillary electrophoresis in forensic applications using multiplex PCR.
In: Proceedings from the 2nd  European Symposium of Human Identification
Innsbruck, Austria, Promega Corporation, p. 131 (1998)

15. Fregeau, C.J., Germain, O., Fourney, R.M.
Fingerprint enhancement revisited and the effects of blood enhancement chemicals on subsequent Profiler Plus TM fluorescent short tandem repeat DNA analysis of fresh and aged bloody fingerprints.
J. Forensic Sci., Vol. 45, p.354 (1999)

16. Stein, C, Kyeck S.H., Henssge, C.
DNA typing of fingerprint reagent treated biological stains
J. Forensic Sci., Vol. 41, p.1012 (1996)

17. Presley, L.A., Baumstark, A.L., Dixon, A.
The effects of specific latent fingerprint and questioned document examinations on the amplification and typing of the HLA DQ alpha gene region in forensic casework.
J. Forensic Sci., Vol. 38, p.1028 (1993)
~~~~~~~~~~~~~~~~~~~~~~

* Note: Low copy number (LCN)  testing is used when only very small samples of DNA are available. The techniques and results are, theoretically, just as valid as any other type of DNA testing. Problems can arise, however, with contamination of samples from other sources. In general, this is more likely to result in a guilty person going free rather than an innocent person being convicted of a crime. This is because extraneous DNA which contaminates a sample may make it hard, or impossible, to show that the 'scene of crime' sample being tested comes from a guilty suspect.  On the other hand, an innocent suspect's DNA should not occur in a tested sample unless the laboratory has been grossly negligent in its procedures.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Published Online: 2001

Croation Medical J., Vol. 42, p. 229 (2001)

Application of Low Copy Number DNA Profiling

Peter Gill
Forensic Science Service, Trident Court, Birmingham, UK


Generally, the lower limits of sensitivity recommended by manufacturers of short tandem repeat (STR) multiplex systems are in the region of 250 pg. Multiplexes usually work at their optimum efficiency when 1 ng of DNA is analyzed (1,2) and not more than 28-30 cycles of amplification are carried out. Interpretation of DNA profiles is assisted by the use of systems that are not too sensitive. This is important because the scientist often needs to associate the presence of a bloodstain (or other evidence) with the DNA profile itself. A highly sensitive system that may reveal DNA from sources other than the body fluid analyzed would require careful consideration when the evidence was interpreted. For this reason, validation exercises often include studies on the effect of rough handling, coughing, or sneezing onto garments to determine if it is possible to transfer casually DNA to evidential material.

Nevertheless, forensic scientists always seek to increase the sensitivity of their methods and the easiest way to do this is simply to raise the number of polymerase chain reaction (PCR) amplification cycles. Findlay et al (3) demonstrated that single (buccal) cells could be analyzed when 34 cycles were used with second generation multiplex (SGM) system. Interpretation was not straightforward. Additional alleles were observed, the sizes of stutters were enhanced, and allele drop out was common. However, such profiles may be interpreted if robust guidelines are used. Subsequently, increasing the sensitivity of PCR by raising the number of cycles has been used to increase the range of evidence types analyzed. For example, Wiegand and Kleiber (4) and Wiegand, et al. (5) analyzed epithelial cells transferred from an assailant after strangulation using 30-31 cycles of PCR. Van Hoofstat, et al. (6) analyzed fingerprints from grips of tools with 28-40 cycles. Barbaro, et al. (7) reported analysis of STRs from hair shafts in the absence of the root using 35-43 cycles.

Increased PCR cycles are routinely used by anthropologists and forensic scientists to identify ancient DNA from bones. We (8) used 38-43 cycles to analyze STRs from 70-year-old bone from the Romanov family. Schmerer, et al. (9) and Burger, et al, (10) analyzed STRs from bone thousands of years old (60 and 50 PCR cycles, respectively). Some other authors used modified PCR methods, such as a nested primer PCR strategy (11). The nested primer PCR strategy used a first round amplification with 40 cycles, with subsequent analysis of a portion with further 20-30 cycles. This method was used to analyze DNA from charred human remains and minute amounts of blood.

Comparison of both different methods available to analyze DNA in amounts less than 100 pg and varying cycling conditions between 28-60 cycles showed that the optimum for both SGM and AMPflSTR-SGM Plus systems (Applied Biosystems, Foster City, CA, USA) was 34 cycles (12). There was little to be gained by increasing the cycle number further, since it did not result in increased sensitivity, but encouraged artifact production. The extreme sensitivity of the method suggested that analysis should only be attempted in a sterile environment to reduce the possibility of contamination from personnel within the laboratory itself.

Nevertheless, all methods used to analyze low copy number (LCN) DNA suffer from several disadvantages that are primarily derived from stochastic variation. When present in low copy number, a molecule that is amplified by chance during the early rounds of the PCR is likely to be preferentially amplified. There are, therefore, several consequences that cannot be avoided:

a) Allele drop out may occur because one allele of a heterozygote locus can be preferentially amplified;

b) Stutters may be preferentially analyzed -- these are sometimes known as false alleles;

c) The method is prone to sporadic contamination -- amplifying alleles that are unassociated with the crime stain, or sample.

This means that different DNA profiles may be observed after replicate PCR analyses. Tarbelet et al (13) suggested a method of replicated analyses that comprised a rule that an allele could only be scored if observed at least twice in replicate samples. This theory was expanded by us (12), who adopted Tarbelet's duplication rule and demonstrated that it was conservative in relation to a new likelihood ratio (LR) method that assessed DNA profiles in relation to sporadic allelic contaminants, stutters, and allelic drop-out. Provided that the level of sporadic contamination was not high (<30% per locus), the duplication method was demonstrated to be conservative relative to the likelihood ratio method.

Interpretation of LCN DNA Profiles

In conjunction with the increased use of DNA profiling, there has been a parallel development in interpretation methodology. Cooke, et al. (14,15) and Evett (16) introduced the notion of the "hierarchy of propositions". This has led to a much deeper understanding of the interpretative process. However, there is currently considerable lack of understanding about issues of transfer and persistence, and further work is being undertaken in this area.

Hierarchy of Propositions

The hierarchy of propositions takes as its premise that scientific evidence may only be interpreted if at least two competing propositions are considered. The top level, level III, of the hierarchy represents the offense level. These are the propositions that are most usually seen to be the province of the jury. For example:

a) The suspect is the offender.

b) The suspect is unconnected with the incident.

These embody the assumption that an offense has indeed been committed and, in general, this would seem to be a level that scientists would prefer to avoid.

The second level, level II, represents the activity level. These would be pairs of propositions that the scientist may feel qualified to address, given adequate information about the case circumstances. For example:

a) The suspect broke the window at the scene.

b) The suspect is unconnected with the incident.

In general, such propositions invoke the classic forensic considerations of transference and persistence. The third level, level I, is the source level, which comprises propositions that relate to the origin of recovered material. For example:

a) The bloodstain came from the suspect.

b) The bloodstain came from some unknown person.

Such propositions would be appropriate when the circumstances are such that the scientist is unable to express an opinion with regard to particular actions or activities. Inevitably, the lower the level of the propositions, the more of the interpretation is left to the jury.

When samples comprise <100 pg of DNA, even level I propositions may be inappropriate. We may have a DNA profile from the crime sample but because of several uncertainties we may not be justified in inferring that it came from the stain that was sampled.

Because of this, we have introduced a new idea: "sub-level I propositions". For example:

a) The DNA profile came from the suspect.

b) The DNA profile came from some unknown person.

Addressing propositions at this level means that the scientist is unable to express a substantive opinion of how the DNA arrived at the site from which it was recovered, or even whether it came from the stain that may have been the reason that sampling was carried out. All considerations between the sub-level I propositions and the level III propositions that the jury must ultimately address must be left to the court, though the scientist has a clear duty to advise the court on the issues that are relevant.

Association of the DNA Profile with the Evidential Material Analyzed

There are two broad categories of evidence types: discrete (e.g. bone, hair) and non-discrete (e.g. blood stains). When using LCN, it is generally easier to associate a DNA profile with a discrete evidence type.

This is because the analysis of bone samples is not attempted without removing the outermost layer by physical methods (e.g. sandpaper) to minimize the possible contamination from modern DNA. Similarly, hair shafts can be washed in a detergent solution to remove adhering DNA. This cannot be done with evidence types that are not discrete, eg, blood stained cloth, hence the chance is increased that a DNA profile may not be directly associated with the evidential body fluid that is "apparently" analyzed.

Because there is a serious possibility of transferring LCN DNA from a modern source, to either minimize the chance of contamination or to identify an occurrence, we use the following guidelines:

1. DNA extractions and setting up PCR reactions are carried out in a dedicated laboratory.

2. Personnel wear disposable laboratory coats, gloves, and face masks.

3. Benches and equipment are frequently treated with bleach (or equivalent) and irradiated with UV light.

4. PCR amplification is carried out in a separate laboratory or laboratory area.

5. Negative controls are used with every test to demonstrate absence of contamination.

6. PCR tests are duplicated wherever possible. All results are compared against a staff database. A database to eliminate investigators of crime scenes as potential contributors is also under preparation.

Defining when DNA Transfer Can Occur

Before and after a crime event, there is the potential for adventitious transfer of cells. Note that the term contamination is reserved for transfer of DNA after the crime event. Adventitious transfer and laboratory contamination usually involves low levels of DNA.

The association of body fluid and the DNA profile is not implicit. If the body fluid giving a positive presumptive test is small or degraded then the DNA profile may have originated from an alternative source. For example, a fresh saliva stain, the latter solely contributing to the observed result, may mask a small-degraded blood-spot. The scientist cannot infer either the type of cell donating the DNA or the time when the cells were deposited.

An estimate of the quantity of DNA is useful to assist in the interpretation of the relevance of a DNA profile. For example, if a visible fresh bloodstain yields several micrograms of DNA, it is not unreasonable to associate the DNA profile with the bloodstain according to level I proposition. However, the association is uncertain if the bloodstain is minute, old, and yields just a few picograms of DNA. It may be appropriate to use a sub-level I proposition. Inevitably, there is a direct relationship between the quantity of DNA present and the relevance of the evidence. The interpretation of the case can only follow after an assessment of all the available evidence, taking into consideration the scenarios offered by prosecution and defense lawyers.

Assessment of Contamination Risks

DNA can be transferred at any time before, during, and after the crime. The foregoing discussion has covered the possibility of adventitious transfer at a period before the crime and it is implicit that the DNA profile matches a suspect. If the DNA profile does not match the suspect, then post-crime transfer must be considered. Contamination is transfer of DNA after the crime event. Potential sources of contamination are:

a) Investigative officers, pathologists, etc, at the crime scene;

b) Laboratory staff;

c) Cross contamination from samples processed in the laboratory (e.g. by aerosol);

d) Plastic-ware contamination (may be contaminated at the manufacturing source).

Whereas a) and b) can be covered by reference to staff databases, and databases of investigating officers, c) and d) are more difficult to detect but are minimized by good laboratory design, staff wearing anti-contamination clothing and face masks, and UV sterilization of plastic-ware.

Transfer of DNA by individuals unassociated with the crime before the crime event itself is defined as adventitious transfer. When a DNA profile does not match the suspect, the following possibilities apply:

a) The suspect is innocent and the perpetrator profile has been visualized.

b) Cells have been transferred by an innocent individual before the crime (perpetrator has not shed cells) "adventitious transfer".

c) Cells have been transferred by an investigator after the crime event (perpetrator has not shed cells)  "contamination".

Note that mixtures may show DNA profiles arising from a combination of the three different events listed.

The circumstances of the victim leading up to the crime event is unknown to the scientist, hence the possibilities of adventitious transfer cannot be directly ascertained. Once the crime has been discovered, the scene and the associated evidence enter a controlled environment, where the risk of contamination is minimized by the adoption of good laboratory and investigative practice.

The primary risk of contamination is wrongful exclusion ? particularly if the contaminant masks the perpetrator?s profile. For the converse to apply ? wrongful inclusion, either tube mix-up or gross contamination (e.g., use of pipette tips contaminated in the laboratory ? e.g., used twice) would be required. Good laboratory practice renders this a virtual impossibility and is not considered further here.

Current Reporting of Sub-level I Propositions in Statements

Because of uncertainties that surround persistence and transfer, the statements are written to reflect this and interpretation usually proceeds according to sub-level I principles. Examples of the wording used in statements are given below. Interpretation is dependent upon a full analysis of the circumstances of the crime and based on a careful consideration of all of the non-DNA evidence.

Observation of Mixtures

With LCN, mixtures are commonly encountered. It cannot be determined whether recovered DNA profiles are associated with a crime event. An example statement follows:

The observation of mixed DNA STR profiles (ie, from more than one individual) can be anticipated. E.G. From past experience it is not unusual to detect DNA profiles on items that match the profile of an individual who has habitually worn that item. However, currently we have no information to assist with questions of transfer and persistence of low levels of DNA on items such as XXXXXX. Thus consideration should be given as to how the DNA detected has been transferred to that item, and consequently to the relevance of finding profiles matching the individuals in the case.?

In the following, two alternatives are considered. No reference is made about the origin of the body fluid type ? it is simply stated that DNA was recovered from the item. ?Either the majority of the DNA originated from Mr. X; or the majority of the DNA originated from someone other than and unrelated to Mr. X. If this DNA had, in fact, originated from Mr. X, then I would expect to obtain matching profiles.?

In the summary section the following paragraphs are included ? this statement was specifically written for a case where DNA from a watchband matched a suspect.

When very small amounts of DNA are analyzed, special considerations arise as follows:

a) Although a DNA profile has been obtained, it is not possible to identify the type of cells from which the DNA originated, neither is it possible to state when the cells were deposited.

b) It is not possible to make any conclusion about transfer and persistence of DNA in this case. It is not possible to estimate when the suspect last wore the watch, if it is his DNA.

c) Because the DNA test is very sensitive, it is not unexpected to find mixtures. If the potential origins of DNA profiles cannot be identified, it does not necessarily follow that they are relevant to this case, since transfer of cells can occur as a result of casual contact.

Effectively, the strength of the LCN DNA evidence is decreased compared to conventional DNA analysis. This inevitably arises from uncertainties relating to the method of transfer of DNA to a surface and uncertainties relating to when the DNA was transferred. It is emphasized that the relevance of the DNA evidence in a case can only be assessed by a concurrent consideration of all the non-DNA evidence. Research is currently being undertaken to devise a probabilistic Bayesian method that encapsulates the DNA and non-DNA evidence.
 

~~~~~~~~
References

~~~~~~~~

1. Sparkes, R., et al.
The validation of a 7-locus multiplex STR test for use in forensic casework 
I. Mixtures, ageing, degradation and species studies
Int. J. Legal Med., Vol.109, p. 186 (1996)

2. Sparkes, R., et al.
The validation of a 7-locus multiplex STR test for use in forensic casework
II. Mixtures, ageing, degradation and species studies.
Int. J. Legal Med., Vol. 109, p.186 (1996)

3. Findlay, I., Frazier, R., Taylor, A., Urquhart, A.,
Single cell DNA fingerprinting for forensic applications
Nature, Vol. 389, p. 555 (1997)

4. Wiegand P, Kleiber M.
DNA typing of epithelial cells after strangulation
Int. J. Legal Med., Vol.110, p.181 (1997)

5. Wiegand, P., Trubner, K., Kleiber, M.
STR typing of biological stains on strangulation tools
Progress in Forensic Genetics, Vol. 8, p.508 (2000)

6. Van Hoofstat,  D., et al.
DNA typing of fingerprints and skin debris: sensitivity of capillary electrophoresis in forensic applications using multiplex PCR
In: Proc. 2nd European Symp. Human Ident., p. 131
Promega Corp., Innsbruck, Austria, (1998)

7. Barbaro A, Falcone G, Barbaro A.
DNA typing from hair shaft
Progress in Forensic Genetics,  Vol. 8, p.523 (2000)

8. Gill, P., et al.
Identification of the remains of the Romanov family by DNA analysis
Nat. Genet., Vol. 6, p.130 (1994)

9. Schmerer ,W.M., et al.
Optimized DNA extraction to improve reproducibility of STR genotyping with LCN DNA 
Electrophoresis Vol. 20, p. 1712 (1999)

10. Burger, J., et al.
DNA preservation: A microsatellite-DNA study on ancient skeletal remains
Electrophoresis Vol. 20, p.1722 (1999)

11. Strom, C.M., Rechitsky, S.
Use of nested PCR to identify charred human remains and minute amounts of blood
J. Forensic Sci., Vol.43, p.696 (1998)

12. Gill P, et al.
An investigation of the rigor of interpretation rules for STRs derived from less than 100 pg of DNA
Forensic Sci. Int., Vol.112, p.17 (2000)

13. Taberlet, P., et al.
Reliable genotyping of samples with very low DNA quantities using PCR
Nucleic Acids Res., Vol. 24, p.3189 (1996)

14. Cook, R, Evett, I.W., Jackson, G, Jones P.J., Lambert, J.A.
Hierarchy of propositions: deciding which level to address in casework
Sci. Justice,  Vol. 38, p. 231 (1998)

15..Cook, R, Evett, I.W., Jackson, G., Jones, P.J., Lambert, J.A.
Case pre-assessment and review in a two-way transfer case
Sci. Justice,  Vol. 39, p.103 (1999)

16. Evett , I.W., Jackson, G., Lambert, J.A.
Hierarchy of propositions: The distinction between explanations and propositions
Sci. Justice, Vol. 40, p.3 (2000)