How DNA Evidence Is Used in Sexual Assault Cases - A Real Case Explained

The sampling strategy adopted during forensic examination is a critical component of DNA evidence generation and interpretation. In this matter, the high and low vaginal swabs were combined prior to testing. This approach affects how the biological material is represented in subsequent laboratory results and has implications for the interpretation and evidential value of any DNA findings.

Determining an appropriate sampling strategy requires consideration of multiple factors, including the allegation, the anatomical location of interest, the expected distribution of biological material, the time since the alleged activity, and the potential for transfer or degradation. In practice, however, laboratory workflows are often oriented toward maximising the likelihood of obtaining a DNA profile. While this is an important operational objective, it is distinct from the question of evidential value – that is, what the resulting DNA findings can meaningfully say about the alleged events.

Once swabs from different anatomical locations are combined, the ability to attribute any resulting DNA profile to a specific site is lost. If a DNA profile is subsequently obtained, it is no longer possible to determine whether it originated from the high vaginal region, the low vaginal region, or a combination of both. In certain case contexts, this distinction may be scientifically and contextually important, as the distribution of biological material within the vaginal canal can vary depending on time since intercourse, movement, drainage, and other post-event factors.

In this matter, no DNA profile was ultimately obtained from the combined swabs, and the impact of the sampling strategy is therefore limited. However, had a profile been generated, its anatomical origin would have remained uncertain. This may have influenced the interpretation of the findings, particularly if questions arose regarding persistence, timing, or the mechanism of deposition.

Laboratory sampling strategies are not described in detail within forensic reports. Nonetheless, the case file, examination notes, and laboratory documentation can provide insight into how samples were collected, processed, and prioritised. In reviewing this material, I considered whether the approach taken was appropriate in the context of the case, and whether alternative sampling or testing strategies may have been available. In some circumstances, targeted examination of individual swabs, or the use of complementary recovery techniques, may assist in identifying the presence – or absence – of biological material in specific locations and may provide information of potential evidential value, including exculpatory findings.

A range of established recovery methods are available in forensic biology, including site-specific swabbing, staged sampling, and adhesive tapelift techniques for recovering cellular material from skin or substrates. The selection of method influences both the likelihood of recovering DNA and the interpretive value of the results. For this reason, sampling strategy is not merely a technical step in the laboratory process; it is integral to how forensic findings are generated, understood, and ultimately evaluated in the context of the case.

In forming opinions on sampling and testing strategy, I rely on a substantial body of peer-reviewed research addressing evidence recovery methods, anatomical sampling, and the performance of swabbing and tapelift techniques in forensic casework. These studies demonstrate that recovery outcomes and interpretive value are influenced by sampling location, method selection, and handling practices, and that different strategies may produce materially different evidential outcomes.

1.     Sewell, J., & Quinones, I. (2014). Collection of biological evidence in sexual assault investigations. Journal of Forensic Nursing.

2.     Verdon, T.J., Mitchell, R.J., & van Oorschot, R.A.H. (2014). The influence of sample type and DNA extraction method on the detection of DNA in mixed biological samples. Forensic Science International: Genetics.

3.     Foster, E.A., et al. (2013). Evaluation of swabbing techniques for the recovery of DNA from skin surfaces. Forensic Science International: Genetics.

4.     Daly, D.J., Murphy, C., & McDermott, S.D. (2012). The transfer of touch DNA and the effectiveness of recovery techniques. Forensic Science International: Genetics.

5.     Fieldhouse, S., & Gwinnett, C. (2012). The use of tapelifts in the recovery of DNA evidence from surfaces. Journal of Forensic Identification.

6.     Rutty, G.N. (2001). An investigation into the effectiveness of swabbing and tapelift techniques in forensic sampling. Forensic Science International.

No spermatozoa were detected on the high vaginal swab in this matter. In those circumstances, semen cannot be confirmed at that site on the basis of the laboratory findings. The absence of sperm cells is a scientifically relevant observation, but it does not, in isolation, establish that ejaculation did not occur.

The persistence of spermatozoa within the vaginal environment is known to be time-dependent and highly variable. Following ejaculation, sperm cells may be progressively lost, degraded, or redistributed through a combination of biological and mechanical processes. These include natural drainage, vaginal physiology, microbial activity, enzymatic degradation, and the effects of post-event activities such as washing, urination, menstruation, or further consensual sexual activity. The timing of sampling relative to the alleged event is therefore critical when interpreting the absence of spermatozoa.

Time-since-intercourse research demonstrates that sperm recovery rates decline over time and that detection is influenced by multiple variables, including the volume of ejaculate, condom use, anatomical differences, movement and posture, and the collection method used during examination. Even within relatively short timeframes, spermatozoa may not be recovered from all vaginal sampling sites, and distribution within the vaginal canal is not uniform. The absence of sperm on a high vaginal swab must therefore be considered holistically alongside findings from other sampling locations, laboratory results, and the broader case circumstances.

Importantly, the absence of spermatozoa does not preclude the presence of seminal fluid or male DNA. Sperm cells may degrade or be absent while other components of semen — including proteins, enzymes, or non-sperm cellular material — persist and may still be detectable through biochemical or DNA testing. For this reason, interpretation must consider the full range of laboratory findings rather than relying on microscopy alone.

In forming opinions on the absence of spermatozoa and its significance, reliance is placed on a substantial body of peer-reviewed research examining sperm persistence, time-since-intercourse dynamics, vaginal biology, and the recovery of semen and DNA following sexual activity. These studies consistently demonstrate that sperm detection is influenced by time, environment, and post-event factors, and that non-detection cannot be interpreted in isolation. This literature provides the scientific framework for evaluating the evidential significance of the findings in this matter.

1      Willott, G.M. (1989). An examination of the persistence of spermatozoa in the vagina. Forensic Science International.

2      Davies, A., & Wilson, E.M. (1974). The persistence of seminal constituents in the human vagina. Forensic Science.

3      Hanson, E.K., & Ballantyne, J. (2010). Time since intercourse: the challenge of sperm persistence and recovery in sexual assault casework. Forensic Science International: Genetics.

4      Mayntz-Press, K.A., & Morling, N. (2015). Forensic investigation of sexual assault – time since intercourse and biological evidence. Forensic Science International: Genetics.

5      Girardet, R., et al. (2009). Collection and persistence of biological evidence following sexual assault. Journal of Forensic Sciences.

6      Anderson, S., Howard, B., Hobbs, G.R., & Bishop, C.P. (2006). A method for determining the time since intercourse from biological samples. Forensic Science International.

Spermatozoa are microscopic reproductive cells produced in the testes. Only a small portion of semen is made up of these cells; the majority is a complex biological fluid containing proteins, enzymes, and other biochemical components contributed primarily by the prostate gland, seminal vesicles, and other accessory glands.

Importantly, semen does not always contain sperm cells. Conditions such as vasectomy, certain medical treatments, age, or naturally low sperm counts (including azoospermia) can result in seminal fluid being present in the absence of detectable spermatozoa.  A study by Soares-Vieira et al. (2007) analysed the DNA content of semen from 105 males, determining that the concentration of DNA present varied substantially.  Therefore the visualisation of sperm cells within the sample is crucial to the interpretation of the forensic DNA results.

Laboratory testing often targets specific biochemical markers that are characteristic of seminal fluid. Key components include enzymes and proteins such as acid phosphatase, prostate-specific antigen (PSA), and semenogelin.

Acid phosphatase is an enzyme produced in high concentrations by the prostate gland. Presumptive acid phosphatase testing is commonly used as an initial screening method because elevated enzyme activity may indicate the presence of seminal fluid. Early work Early work by Kind et al (1954) demonstrated that the acid phosphatase test has practical detection thresholds and may fail to identify very dilute or degraded seminal material.  The limitations of this test are also extensively reported in the literature; and an intimate understanding of the methodology applied within the specific laboratory assists in identifying the potential for false positive reactions in the specific matter.

Prostate-specific antigen (PSA), also known as p30, is a protein produced by prostatic epithelial cells and is sometimes used as a more specific marker for seminal fluid. PSA testing can detect semen even where spermatozoa are absent, but it too is not entirely unique to semen and may occasionally be identified in other bodily fluids at low concentrations.

Semenogelin, the target protein of the RSID Semen test is a major structural protein derived from the seminal vesicles and forms the coagulated matrix of freshly ejaculated semen. Its presence can assist in confirming seminal material and in understanding the biochemical processes involved in semen liquefaction and degradation over time.

Understanding the biological function and origin of these enzymes and proteins is central to forensic interpretation. Laboratory tests do not directly answer whether ejaculation occurred, when it occurred, or the circumstances in which biological material was deposited. Rather, they detect markers consistent with seminal fluid, each with its own specificity, sensitivity, and potential for cross-reactivity. The interpretation of results therefore requires careful consideration of the underlying biochemistry, the limitations of each test, and the broader evidential context in which the findings are being applied

.

The presence of the terms sperm fraction and non-sperm fraction in a forensic DNA report indicates that a process known as differential extraction was used during laboratory testing. This is important, as it provides insight into how the sample was handled and how the resulting DNA profiles were generated.

Differential extraction is a laboratory technique designed to separate sperm cells from other cellular material within a biological sample, most commonly in the examination of sexual assault evidence. The aim is to preferentially recover male DNA from spermatozoa while isolating DNA from epithelial and other non-sperm cells into a separate fraction.

The process is carried out in stages. Initially, non-sperm cells — such as skin or vaginal epithelial cells — are lysed using detergents and enzymes, releasing their DNA into solution. This material is collected as the non-sperm fraction. The remaining cellular material is then treated with additional reagents, including the reducing agent dithiothreitol (DTT), which disrupts the structural integrity of sperm heads and allows DNA from spermatozoa to be released and collected separately as the sperm fraction. DTT is required because spermatozoa are resistant to standard lysis; the DNA within the sperm nucleus is tightly compacted by protamines and stabilised by extensive disulfide cross-linking within sperm head proteins.

However, this method is not perfectly efficient. Carryover of DNA from non-sperm cells into the sperm fraction is well recognised. In this matter, DNA attributable to a female contributor is detected within the sperm fraction, despite not originating from spermatozoa themselves. This observation is consistent with the known limitations of differential extraction and must be considered in interpretation.

For this reason, the presence of DNA within the sperm fraction must be approached cautiously. It does not automatically mean that the DNA originated from sperm, nor does it, in isolation, establish direct sexual contact. DNA detected in this fraction may arise from other biological material, including epithelial cells introduced through prior or indirect contact, or through transfer during handling and processing. Scientific interpretation is therefore required to understand what the findings do — and do not — indicate.

Interpretation in this matter is further complicated by the case file recording the identification of only a single spermatozoon on the vulval swab. In such circumstances, careful consideration must be given to whether the DNA profile observed in the electropherogram represents the expected outcome if the DNA were derived from sperm, as opposed to another cellular source. This type of evaluation requires detailed knowledge of the extraction process, the behaviour of mixed cellular material during laboratory processing, and the way DNA from different sources is represented in resulting profiles.

In reviewing the evidence and forming my opinions, I rely on an extensive body of peer-reviewed scientific literature addressing differential extraction, sperm cell lysis, DNA recovery from mixed biological samples, and the interpretation of sperm and non-sperm fractions in forensic casework. Foundational studies describing the development and performance of differential extraction methods, research examining the effects of carryover between fractions, and contemporary work on DNA interpretation in complex mixtures collectively inform the evaluation of laboratory findings. These sources provide the scientific framework within which the results in this matter are assessed and interpreted.

Differential extraction techniques, their efficiency, and their limitations in mixed biological samples have been extensively described in the forensic literature (Gill et al.; Yoshida et al.; Horsman et al.; Verdon et al.; Norris et al.; Garvin et al.; Castro et al.).

1      Gill, P., Jeffreys, A.J., & Werrett, D.J. (1985). Forensic application of DNA ‘fingerprints’. Nature, 318, 577–579.

2      Yoshida, K., Sekiguchi, K., & Mizuno, N. (1995). A simple method for differential extraction of sperm DNA from mixed biological stains. Journal of Forensic Sciences.

3      Horsman, K.M., Barker, S.L., Ferrance, J.P., Forrest, K.A., Koen, K.A., & Landers, J.P. (2006). Separation of sperm and epithelial cells in a microfabricated device: potential application to forensic analysis of sexual assault evidence. Analytical Chemistry, 78, 3122–3128.

4      Verdon, T.J., Mitchell, R.J., van Oorschot, R.A.H. (2014). The influence of sample type and DNA extraction method on the detection of male DNA in mixed samples. Forensic Science International: Genetics.

5      Norris, J.V., Manning, K., & Linke, S.J. (2007). Evaluation of differential extraction methods for forensic DNA analysis of sexual assault samples. Journal of Forensic Sciences.

6      Garvin, A.M., Bottinelli, M., Gola, M., Conti, A., & Soldati, G. (2009). DNA preparation from sexual assault cases by selective degradation of contaminating DNA from the victim. Forensic Science International: Genetics

7      Castro, A., Alonso, A., & Martín, P. (2014). Challenges in differential extraction and interpretation of sperm and epithelial fractions in forensic casework. Forensic Science International: Genetics.

8      Green, R.L., & Sambrook, J. (2012). Molecular cloning: laboratory approaches to DNA extraction and protein denaturation relevant to sperm lysis.

This number is what is known as a likelihood ratio. In forensic DNA testing, a likelihood ratio is a scientific expression of how much more probable the observed DNA results are under one proposition compared with an alternative. Most commonly, the comparison is between the proposition that the DNA originated from a named individual and the proposition that it originated from an unknown, unrelated person. It is a measure of the strength of the scientific findings; it is not a statement about guilt, and it does not explain how or when the DNA was deposited.

In this matter, the likelihood ratio evaluates two competing propositions: that the DNA originated from the accused (or one of his paternal relatives), or that it originated from an unknown, unrelated individual. It does not address the mechanism of transfer, the timing of deposition, or the activity that led to the DNA being present. It speaks only to the relative probability of the DNA findings under the competing propositions about source.

A likelihood ratio of 5,600 means that the DNA results are 5,600 times more probable if the DNA came from the named individual (or a paternal relative) than if they came from someone unrelated. In scientific terms, this is typically described as providing strong support for the proposition that the DNA originated from that paternal lineage, within the limits of the assumptions and propositions being tested.

When assessing the robustness of a likelihood ratio, several factors must be considered. These include the quality and completeness of the DNA profile, the number of contributors present in the sample, the allele frequency databases used, and the population groups selected for statistical evaluation. Population genetics is central to likelihood ratio calculations: the rarer a genetic profile is within a relevant population, the stronger the statistical support that may be attributed to a match. The choice of database and the representativeness of the reference populations therefore directly influence the resulting likelihood ratio.

The type of DNA testing used is also critical. In autosomal DNA analysis, likelihood ratios are often generated using probabilistic genotyping software, such as STRmix, which applies complex statistical models to interpret mixed DNA profiles. These systems incorporate allele frequency databases, account for stochastic effects such as allele drop-out and drop-in, and evaluate multiple competing contributor combinations. While such methods are scientifically validated and widely used, they rely on modelling assumptions and population data. The possibility of false inclusions – where an individual may appear to be supported by the DNA evidence due to statistical modelling, mixture complexity, or shared alleles within a population – must therefore always be considered alongside the reported likelihood ratio.

In contrast, Y-STR haplotype evidence is interpreted differently from autosomal DNA. Because Y-STR profiles are inherited along the paternal line and shared among male relatives, statistical evaluation is based on haplotype frequency rather than independent allele combinations. Likelihood ratios derived from Y-STR testing therefore reflect the frequency of that paternal lineage within relevant populations, rather than the probability of a unique individual profile. As a result, Y-STR results cannot distinguish between fathers, sons, brothers, or other paternal relatives, all of whom may be potential sources of the DNA.

Accordingly, a likelihood ratio based on Y-STR findings must be interpreted within this framework. It can indicate that the DNA is consistent with a particular paternal lineage and that such a lineage may be relatively uncommon in the reference population, but it cannot identify a specific individual to the exclusion of all other males within that lineage, nor can it determine the circumstances in which the DNA was deposited.

In reviewing the likelihood ratio evidence and forming my opinions, I rely on a substantial body of peer-reviewed scientific research and validation data relating to population genetics, allele frequency databases, probabilistic genotyping, and the statistical interpretation of DNA mixtures. This includes studies examining the performance and limitations of likelihood ratio approaches, the operation and validation of software such as STRmix, the risk of false inclusions in complex mixtures, and the interpretation of Y-STR haplotype frequencies using large reference datasets. These sources provide the scientific and statistical foundation for evaluating the strength, reliability and limitations of likelihood ratio evidence in the context of the specific circumstances of the case.

1.     Evett, I.W., Gill, P.D., Jackson, G., Whitaker, J., & Champod, C. (2002).  Interpreting DNA evidence: a review. International Statistical Review, 70(3), 473–495.

2.     Buckleton, J., Triggs, C.M., & Walsh, S.J. (2005).  Forensic DNA Evidence Interpretation. CRC Press.

3.     Gill, P., Brenner, C., Buckleton, J., et al. (2006). DNA commission of the International Society of Forensic Genetics: recommendations on the interpretation of mixtures. Forensic Science International, 160, 90–101.

4.     Bright, J.A., Taylor, D., Curran, J.M., & Buckleton, J. (2016). Developing allelic and stutter peak height models for a continuous method of DNA interpretation. Forensic Science International: Genetics, 21, 13–22.

5.     Taylor, D., Bright, J.A., & Buckleton, J. (2013). The interpretation of single source and mixed DNA profiles. Forensic Science International: Genetics, 7, 516–528.

6.     Bleka, Ø., Storvik, G., Gill, P., & Egeland, T. (2016).  EuroForMix: an open source software based on a continuous model to evaluate STR DNA profiles from a mixture. Forensic Science International: Genetics, 21, 35–44.

7.     Taylor, D., Bright, J.A., Buckleton, J., & Curran, J.M. (2013).  An illustration of the use of STRmix™ for DNA mixture interpretation.  Forensic Science International: Genetics, 7(2), 226–233. 

8.     Willuweit, S., & Roewer, L. (2015). The new Y Chromosome Haplotype Reference Database (YHRD): an extended online resource for Y-STR haplotypes. Forensic Science International: Genetics, 15, 43–48.

Y-STR DNA testing targets genetic markers located on the Y chromosome and is used to detect and profile male DNA within biological samples. Because the Y chromosome is present only in males and is passed from father to son largely unchanged, Y-STR analysis allows male DNA to be identified even in the presence of large amounts of female DNA. For this reason, it is commonly used in the examination of sexual assault evidence, particularly where spermatozoa are absent, where differential extraction has been unsuccessful, or where standard autosomal DNA testing yields a profile dominated by female DNA.

Unlike autosomal DNA profiling, Y-STR testing does not produce an individualising profile. All males within the same paternal lineage share the same Y-STR haplotype, meaning the result cannot distinguish between fathers, sons, brothers, or other paternal relatives. Instead, Y-STR analysis is best understood as identifying a paternal line rather than a specific individual. It can assist in determining whether male DNA is present and whether it is consistent with a particular male or male lineage, but it cannot establish the identity of a single contributor to the exclusion of all others.

Y-STR testing is particularly useful in complex mixtures. It can recover male DNA signals that would otherwise be obscured by high levels of female DNA and can provide investigative or evaluative information where conventional autosomal testing is uninformative. However, this sensitivity also brings limitations. The technique may detect very small amounts of male DNA, including DNA introduced through indirect transfer, background contact, or handling. As with other forms of DNA testing, detection does not determine when the DNA was deposited, how it was transferred, or whether it relates to the alleged activity under investigation.

Interpretation of Y-STR findings relies heavily on haplotype frequency data. Because Y-STR profiles are shared within paternal lineages, their evidential value is assessed by reference to how commonly a particular haplotype occurs within relevant populations. This is typically undertaken using large population databases such as the Y Chromosome Haplotype Reference Database (YHRD), which collates Y-STR profiles from populations worldwide. The number of observed haplotypes and their distribution within the database inform statistical assessment of how rare or common a given Y lineage may be.

Careful consideration must be given to the selection of appropriate population data when evaluating Y-STR evidence. Subpopulation structure, geographic ancestry, and sampling composition can influence haplotype frequency estimates. The choice of database and reference population is therefore not merely a technical step; it is central to the proper interpretation of the findings and the weight that may be attributed to them.

In forming opinions on Y-STR findings, reliance is placed on a substantial body of peer-reviewed research addressing the development, validation and interpretation of Y-chromosome markers in forensic science. Foundational studies describing Y-STR population genetics and forensic application (Kayser; Jobling & Tyler-Smith), the development of haplotype databases and statistical approaches to their evaluation (Roewer; Willuweit & Roewer), and validation and casework performance (Butler; Ballantyne et al.) collectively inform the evaluation of results. These works, together with research examining transfer, persistence and interpretive limitations in mixed samples, provide the scientific framework within which Y-STR findings are assessed and explained in the context of the specific circumstances of a case.

1      Jobling, M.A., & Tyler-Smith, C. (2003).  The human Y chromosome: an evolutionary marker comes of age. Nature Reviews Genetics, 4, 598–612.

2      Kayser, M. (2017). Forensic use of Y-chromosome DNA: a general overview. Human Genetics, 136, 621–635.

3      Butler, J.M. (2003). Recent developments in Y-short tandem repeat and Y-single nucleotide polymorphism analysis. Forensic Science Review, 15, 91–111.

4      Ballantyne, K.N., Goedbloed, M., Fang, R., et al. (2012).  Mutability of Y-chromosomal microsatellites: rates, characteristics, molecular bases, and forensic implications. American Journal of Human Genetics, 87, 341–353.

5      Prinz, M., Boll, K., Baum, H., & Shaler, R. (1997). Multiplexing of Y chromosome STRs and performance for mixed samples in forensic casework. Forensic Science International.

6      Hanson, E.K., Ballantyne, J. (2007). An ultrahigh discrimination Y-chromosome short tandem repeat multiplex DNA typing system. Forensic Science International: Genetics.

7      Sibille, I., Duverneuil, C., Lorin de la Grandmaison, G., et al. (2002). Y-STR DNA amplification as biological evidence in sexually assaulted female victims with no cytological detection of spermatozoa. Forensic Science International, 125, 212–216.

8      Roewer, L. (2009). Y chromosome STR typing in crime casework. Forensic Science, Medicine, and Pathology, 5, 77–84.