Bridging Law and Science
THE INTERPRETATION PROCESS OF COMPLEX MIXED DNA PROFILES FROM SOURCE TO ACTIVITY LEVEL BY MEMBERS OF THE COURT
Joyce Cuijpers (12825077) Literature Thesis 5 EC
University of Amsterdam
Institute for Interdisciplinary studies
Master Forensic Science
District Court of The Hague
26 October 2020 – 21 December 2020
Supervisor
Wanda Remijn MSc.
Examiner
Dr. Maarten J. Blom
Coordinator Prof. Dr. Arian C. van Asten
ABSTRACT
Developments in forensic biology research have advanced DNA profiling techniques immensely in both sensitivity and capacity. As a result, the finding of complex DNA mixture samples has become increasingly common in forensic casework, including rape, robbery, and homicide. However, research has shown that mixed DNA profiles are more difficult to interpret than single DNA profiles for both experts and members of the court. Nowadays, it seems that in most cases with complex DNA mixture evidence, the activity as to how the DNA ended up at the crime scene is disputed instead of the source of the DNA mixture sample. Generally, the forensic expert performs evidence interpretations at the source level, whereas members of the court evaluate the evidence at the offense level. However, the evidence interpretation at the activity level is currently being performed by both forensic experts and members of the court, indicating that there is no clear task separation regarding the evidence evaluations at the activity level. Interestingly, it seems that members of the court have taken the task of activity level evaluations upon themselves when being confronted with alternative scenarios by the defense regarding indirect DNA transfer. Therefore, this literature thesis assessed whether it is feasible for members of the court to perform the translation process from source to activity level evaluations of complex DNA mixture evidence themselves. After reviewing Dutch case law, members of the court should be made aware of the possibilities in the forensic field with regard to indirect DNA transfer examination as some may not fully comprehend the relevance and rather presume that their own knowledge is sufficient. Therefore, forensic advisors could play a more prominent role in advising whether activity level examinations should be performed or not since they possess knowledge of both law and science. However, we have to keep in mind that the forensic expert should not be seated in the chair of the judge, and the judge always has the final say.
CONTENT
Abstract ... i
Introduction ... 1
The translation process from source to activity level interpretations ... 3
Forensic activity level evaluation of indirect transfer is disregarded by the court ... 3
Controversy regarding the responsibility of indirect transfer evaluations ... 6
The judicial approach versus the forensic approach ... 8
Discussion and Conclusion ... 10
Acknowledgement ... 13 References ... 13 Scientific publications... 13 Case law ... 15 Appendix ... 16 Search strategy ... 16
INTRODUCTION
Recent developments in forensic biology research have advanced DNA profiling techniques immensely in both sensitivity and capacity. Nowadays, forensic DNA examiners are able to analyze low quantities of trace-material for DNA on a large-scale (Buckleton et al. 2016). As a result of these improvements in DNA typing technologies, the finding of DNA mixture samples has become increasingly common in forensic casework (Tasselli et al. 2015). DNA mixture samples contain DNA profiles derived from two or more contributors. The information that can be acquired from these DNA mixture samples can be highly relevant for criminal investigations and court proceedings involving multiple suspects and/or victims (Gill et al. 2018; Twisk et al. 2019).
However, due to the environmental factors present at the crime scene and the rate of DNA degradation, it is challenging to obtain useful information from DNA mixture evidence as small samples are often acquired. Therefore, the examination of DNA mixture samples is performed through STR analysis, which can aid criminal investigation as it augments small DNA samples and yields low detection rates (Hu et al. 2014). A difference in quantity between DNA profiles in a mixture can indicate the amount of DNA each contributor donated to the mixture. The DNA profile of the highest quantity is referred to as the major component/contributor, while the other DNA profile(s) is/are referred to as the minor component(s)/contributor(s). However, it should be noted that there are cases where no clear major contributor can be distinguished in the DNA mixture sample, leaving a mixture containing only minor contributors (Bille et al. 2019; Budowle et al. 2020). Thus, to the author’s belief, a relatively new concept arises, namely complex DNA mixture samples that are defined by consisting of three or more contributors with no major contributor present. These samples can also include the presence of incomplete profiles. Because of these complex DNA mixture samples, it can become challenging to determine the number of contributors and the quantity that each person contributed to the mixture sample (Hu et al. 2014). Further complicating the DNA mixture analysis are artifacts within the obtained DNA mixture profile – such as the drop-out, drop-in, or stutter – that can interfere with the interpretation process. In reaction to these interpretation issues, the field of forensic biology has taken great measures to establish mathematical models and bio-statistical software to increase the efficiency and accuracy of DNA mixture interpretation (Hu et al. 2014; Fenton et al. 2020).
Despite these statistical improvements, the interpretation of DNA mixtures still mainly relies on the experience and knowledge of the forensic DNA expert, which can be subjective or biased. The opinion of the expert also plays a major role in determining the strength of the evidence (de Keijser et al. 2016). This value is obtained by using the Likelihood Ratio (LR) from Bayes’ Theorem. The LR is determined by establishing mutually exclusive propositions and assigning probabilities to these propositions. Generally, one of the established propositions is the scenario as expressed by the prosecution (Hp), while the alternative proposition is the scenario as expressed by the defense (Hd). When evaluating the strength of the evidence (LR), the probability of the observed results (E) is compared under both propositions if either one would be true. Thus, the LR is equal to the probability of the observed results if the prosecution proposition would be true, divided by the probability of the observed results if the defense proposition would be true (Robertson et al. 2016; Prakken & Meester, 2017; Twisk, 2019).
𝐿𝑅 = 𝑃(𝐸|𝐻𝑝) 𝑃(𝐸|𝐻𝑑)
In the courtroom, DNA mixture evidence is found in various cases – such as rape, robbery, and homicide – and can be perceived as entirely unambiguous by members of the court. However, research has shown that mixed DNA profiles are more difficult to interpret than single DNA profiles for both experts and members of the court (de Keijser et al. 2016; Twisk et al. 2019; Malsch et al. 2016). The interpretation of complex DNA mixture evidence can be performed on multiple levels that constitute a hierarchical structure, namely source, activity, and offense level (Figure 1). Interpretations at the source level are regarding the origin of the trace-material, whereas interpretations at the activity level are concerned with the question of how the trace-material got there. Interpretations at the offense level concern legal issues, in which guilt and innocence are weighed – e.g. did the suspect indeed murder the victim or is he innocent? It should be noted that the forensic expert is only allowed to evaluate the strength of the evidence at source and activity level. Although in practice, activity level evaluations are increasingly performed by the judge, suggesting that there is no apparent task separation regarding the evidence evaluations at the activity level between forensic experts and members of the court. Nowadays, it also seems that in most cases with DNA evidence, the source of the evidence is not disputed by the defense as high evidential values are often obtained. However, the activity as to how the DNA evidence ended up at the crime scene is (Gill et al. 2020). Evidence interpretations at the offense level are solely the task of the judge since prior odds are needed for this type of interpretation. Prior odds represent the view of the prosecution and defense before the evaluation of scientific evidence. These views are shaped by tactical information or knowledge from non-scientific evidence, such as CCTV footage or witnesses. To avoid contextual bias in the forensic expert, the evidence examination is performed while only having access to the relevant and necessary contextual information (Robertson et al. 2016). Additionally, the communication between the forensic scientists and members of the court involved in the
decision-making process is often strained due to the difference in background knowledge and training, allowing risks for misinterpretations in (high profile) criminal cases. However, it is important to note that the forensic expert can only aid or advise the judge during their decision-making process since the judge always has the final say in court (Twisk, 2019; Gill, 2019).
Currently, there is a lack of data on how members of the court deal with the interpretation of complex DNA mixtures, particularly the translation from source to activity level in DNA mixture evidence. While the objective of the court is to make statements regarding the complex DNA evidence at the offense level, the evidence first needs to be interpreted at the activity level. As mentioned before, forensic experts use the Bayesian approach to evaluate the strength of the evidence first at source level and later at activity level. It is important to note that interpretations made at the source level can influence interpretations made at the activity level. Moreover, no conclusion can be made at the activity level from source level evaluations since forensic experts need additional information for this type of evaluation, including transfer, persistence, and recovery probabilities (Gill et al. 2020). However, from case law research by Twisk (2019), it appears that in cases with DNA mixture evidence, members of the court themselves perform the translation from source to activity level interpretations without additional information and guidance from a forensic DNA expert. Consequently, the question arises whether members of the court – the judge in particular – have sufficient knowledge/skills to address the alternative statement of the defense at the activity level without further consulting a forensic DNA expert when assessing complex DNA mixture evidence.
Therefore, this literature thesis analyzes the interpretation process of complex mixed DNA profiles from source to activity level by members of the court. The overall structure of the paper takes the form of three sections. The first section will examine what evidential value members of the court attach to complex DNA mixture profiles and how these members deal with this type of evidence in criminal court cases. The second section will discuss the limitations and subsequent recommendations regarding the translation of DNA mixture evidence from source to activity level as described by the forensic science community. Subsequently, the third section will examine how these limitations and recommendations relate to the use and evaluation of complex DNA mixture evidence in criminal cases by members of the court. Finally, this literature thesis will assess whether it is feasible for members of the court to perform the translation process from source to activity level evaluations of complex DNA mixture evidence themselves, given the complexity of DNA mixture profiles.
THE TRANSLATION PROCESS FROM SOURCE TO ACTIVITY LEVEL INTERPRETATIONS
FORENSIC ACTIVITY LEVEL EVALUATION OF INDIRECT TRANSFER IS DISREGARDED BY THE COURT
Forensic biological research makes a valuable contribution to criminal court proceedings by determining the evidential strength of acquired cell material and DNA samples when being correctly interpreted by members of the court. Nevertheless, misinterpretations can occur, and a well-known fallacy that can be made by members of the court is the prosecutor’s fallacy, also known as transposing the conditional. In this fallacy, the conditional probabilities of the observed results are mistakenly taken for conditional probabilities of the established hypotheses (Robertson et al. 2016; Twisk, 2019). Another kind of fallacy that can be made by members of the court is the defense attorney’s fallacy. For example, when looking at the interpretation of DNA evidence, members of the court can incorrectly assume that the frequency in which DNA characteristics are present in a population is equal to the frequency in which the obtained DNA profile occurs in that same population. This type of assumption or interpretation would lead to a lower evidential value (Buckleton et al. 2016). Consequently, misinterpretations and fallacies made by members of the court can lead to a change in the strength of the evidence, which could have adverse effects on the case as a whole.
In this section, we will examine what evidential value members of the court attach to complex DNA mixture profiles and how they deal with this type of evidence in criminal court cases by looking at Dutch case law published from 2019 to 2020. A case law study approach was adopted to gain a detailed understanding of how judges handle the translation of complex DNA mixture evidence interpretation from source to activity level. The Dutch case law is obtained from the website rechtspraak.nl, of which relevant case law from the district courts and the courts of appeal are filtered and selected by using the following search terms: ‘mixed DNA profiles’, ‘alternative scenario’, ‘activity level’, ‘explanation’, ‘not excluded’, ‘denies’, ‘at least’, ‘secondary transfer’, and ‘plausible’. The search on rechtspraak.nl resulted in a total of fifteen case law articles1, of which the latest is published on October 13th, 2020. Case law of the supreme court has not been included in the examination since the supreme court deals with cassation complaints concerning the incorrect application of the law by the district court and the court of appeal. Furthermore, the inclusion criteria were case law containing DNA mixture evidence with at least three or more contributors and defenses that present alternative scenarios at the activity level regarding the DNA mixture evidence. From the fifteen selected case law articles, three articles contain rulings from the courts of appeal, and twelve articles contain rulings from the district courts throughout the Netherlands. The selected case law contains judgements on various charges, which include six charges of (attempted) murder or manslaughter, four charges of robbery, one charge of sexual assault, two charges of illegal possession of firearms, and two charges concerning the illegal possession and detonation of explosives.
According to the Bayesian framework, forensic evidence can be evaluated and interpreted at the source and activity level by the forensic expert. The expert presents the obtained findings in a report, which will be sent to the prosecutor, defense, and the court. However, the evidence interpretation at the source level and the translation toward the activity level may pose difficulty for members of the court as this could be outside their field of expertise (Robertson et al. 2016). To determine the manner in which court members approach these interpretations, it is necessary to first analyze how DNA mixture evidence is examined and interpreted at the source level.
DNA analysis of biological traces on evidence items is performed by forensic DNA experts from different institutions. In the Netherlands, most cases are handled by the Netherlands Forensic Institute in The Hague, which functions as a central forensic body for both national and international criminal cases. Out of the fifteen case law articles, it appears that in eight articles the DNA analysis and source level examination were performed by experts from the Netherlands Forensic Institute (NFI). In three other case laws, it appears that DNA analysis and source level examination were performed by experts from a different institute than the NFI, including the Independent Forensic services (IFS), the Maastricht Forensic Institute (TMFI), and the Forensic Laboratory for DNA-research (FLDO). Moreover, DNA analysis and source level examination may be performed by multiple experts from different institutions. For example, when the defense questions the impartiality of the forensic expert, the reliability of the used methods, or the reliability of the results from the forensic DNA analysis report, they can request a counter-investigation or -assessment (Depauw, 2020). In one case law article from the Court of Appeal of ‘s-Hertogenbosch2, both the NFI and TMFI performed the DNA-analysis and source level examination of biological trace material acquired from the victim’s jeans. However, in this case law, it is was not explained why both the
1The reference list includes all selected case law under the heading ‘Case law’. When specific case law is mentioned in this paper, it will include a
footnote with the corresponding ECLI number, which stands for the European Case Law Identifier. The ECLI number is a unique code for court rulings within the European Union.
NFI and TMFI were involved in the DNA evidence examinations. In the remaining three case laws, it was not indicated who performed the forensic DNA-analysis and source level examination.
Interestingly, the location of DNA recovery was provided by all case law, whereas the type of biological trace that DNA was extracted from was provided by none – only the words ‘cell material’ were used to describe this. Moreover, some case laws were more elaborate on their descriptions compared to others. During DNA analysis, the forensic expert can examine how many individuals have contributed to the DNA sample obtained from a piece of evidence. It is relevant to know the number of contributors to a DNA sample since this type of information influences the evidential value. To clarify, in case of a DNA mixture profile, the individual DNA profiles of multiple contributors need to be untangled by the expert, which lowers the evidential value of mixtures compared to single profiles at the source level since mixture DNA profiles may be explained in multiple ways in contrast to single DNA profiles (Buckleton et al. 2016). Additionally, it should be noted, when establishing propositions for the evidence evaluation, the forensic expert conditions on known individuals, such as the victim or suspect (Gill et al. 2018). Out of the fifteen case laws, ten cases observed a DNA mixture profile with at least three contributors, and five other cases observed a DNA mixture profile of at least four or more contributors. Interestingly, in one case law3, the results from the DNA analysis state that a DNA mixture profile with at least two contributors is observed, yet in the established hypotheses the assumption of three contributors is used instead of two. However, from the case law it is unclear why this shift from two to three contributors to the DNA mixture profile has occurred. Even though propositions are usually established before an evaluation, a possible explanation could be that extra allele peaks were present in the DNA mixture profile that could not be accounted for by the suspect and one extra unknown contributor. Therefore, they might have used propositions with three contributors to the mixture profile. Generally, when the number of contributors cannot reliably be assigned by the forensic DNA scientist, the DNA mixture profile may be considered unsuitable for DNA evidence interpretation (Bille et al. 2019). Furthermore, in case law from the District Court of Rotterdam4, a partial DNA mixture profile with at least three contributors, including the victim, was obtained from a tie used to bind the victim’s wrists. In partial or degraded DNA profiles, genetic information is missing, so the obtained DNA profile or DNA mixture profiles may resemble the DNA profile of multiple individuals in the population. Consequently, this can affect the strength of the evidence by providing a lower LR (Gill et al. 2018).
In all selected case law, source level examinations were performed by the forensic expert, who expresses the evidential value at the source level by stating the obtained LR. Although the LR is a numerical value, it can also be expressed by using a verbal scale. However, most forensic scientists prefer the numerical value over the verbal equivalent since this allows less room for subjective interpretations (Buckelton et al. 2016; Robertson et al. 2016). In thirteen case law articles, the interpretations at source level were expressed by only stating the numerical value of the LR. Conversely, in one other case law from the District Court of Zeeland-West-Brabant5, the evidential value was expressed by solely stating the verbal equivalent of the LR. Additionally, in one case law from the District Court of Rotterdam6, the numerical value of the LR is mentioned before expressing the strength of the evidence officially with its verbal equivalent. In most case law, the obtained LR values were around one million and one billion, or its verbal equivalent of ‘extremely probable’. In two case law, from the District Court of Amsterdam7 and the District Court of Rotterdam8, the obtained LR values were around one and thirty-nine thousand, with a verbal equivalent of ‘more probable’.
When expressing the LR, several kinds of fallacies can be made by members of the court.During the case law research, it was discovered that there were fallacies present within some of the evidence interpretations made at the source level. Out of the fifteen case laws, two9 contained a prosecutor’s fallacy in the source level interpretations of the DNA mixture evidence. In case law from the Court of Appeal of Arnhem-Leeuwarden, the court has transposed the conditional by stating how likely the hypotheses are given the obtained results instead of stating how likely the obtained results are given the established hypotheses. Similarly, case law from the District Court of Amsterdam stated how likely the established hypotheses were instead of the obtained results. The prosecutor’s fallacy can have adverse effects by overstating the evidential value of the DNA mixture evidence. To clarify, when evaluating the strength of DNA evidence, typically high LRs are obtained that can be perceived as very incriminating even if the DNA evidence is viewed alongside exculpatory evidence. As Robeiro et al. (2020) argue, misinterpreting and overstating the evidential value could lead to viewing DNA evidence as a ‘heuristic of guilt’.
3The District Court of Rotterdam, published 15 July 2019, ECLI:NL:RBROT:2019:5623 4The District Court of Rotterdam, published 19 November 2019, ECLI:NL:RBROT:2019:8958 5The District Court of Zeeland-West-Brabant, published 13 October 2020, ECLI:NL:RBZWB:2020:4932 6The District Court of Rotterdam, published 19 November 2019, ECLI:NL:RBROT:2019:8958 7The District Court of Amsterdam, published 5 August 2019, ECLI:NL:RBAMS:2019:5690 8The District Court of Rotterdam, published 19 November 2019, ECLI:NL:RBROT:2019:8958
9The Court of Appeal of Arnhem-Leeuwarden, published 26 March 2019, ECLI:NL:GHARL:2019:2604; and The District Court of Amsterdam,
The ultimate goal of the court is to decide whether someone is guilty, or innocent of the crime committed. Although DNA evidence interpretations at source level can indicate who was present at the scene, it does not necessarily mean that DNA was transferred through criminal activity (Gill et al. 2020). During the trial, the defense can present alternative scenarios for the obtained evidence in which the activity that has led to the DNA transfer is not crime related (Gill et al. 2020). In thirteen case laws, the defense has indicated that secondary or tertiary DNA transfer cannot be excluded. For instance, in case law from the District Court of Amsterdam10, the defense argues that the suspects’ contribution to the DNA mixture sample was due to secondary or tertiary transfer. As an alternative scenario, the defense states that the acquired cell material was previously present on the clothing of the victim due to consented sexual contact with the suspect. From the clothing of the victim, the cell material of the suspect was indirectly transferred onto one of the garbage bags – in which parts of the victim’s body were found – through the actions of the perpetrator. Alternatively, in case law from the District Court of Rotterdam11, the defense argued that the presence of the suspect's DNA was due to contamination of the evidence. Specifically, a tie used to bind the victim's wrists were brought into direct contact with other pieces of evidence while they had been contained within the same paper bag, thereby transferring the DNA of the suspect onto the aforementioned tie. Apart from this, in case law from the Court of Appeal from ‘s-Hertogenbosch12, the accused provided the court with an alternative scenario after evaluating all the evidence against him. In the alternative scenario, he did not dispute the source of the DNA evidence or the activity of moving the victim’s body into the bushes. However, he stated that the victim was shot by another individual who had first moved the victim’s body onto the side of the road.
To verify the alternative statements made by the defense, the court can request an additional examination at the activity level by the forensic expert. From the fifteen selected case laws, only one13 contains an activity level examination that is performed by forensic DNA experts from the NFI and TMFI. The investigative judge had requested the NFI to investigate whether the obtained results are more likely when the suspect is the only person who has moved the victim’s body compared to when the suspect has moved the victim’s body after another individual has moved the body. When analyzing the established propositions, the forensic expert has taken into account context information provided by the court. However, in this case law, no information about the used probabilities is specified and only the verbal equivalent of the LR is stated. The results from the NFI and TMFI indicated that the observed DNA mixture evidence is more probable when the suspect is the only person who moved the victim’s body than when an unknown individual moved the victim’s body before the suspect has moved the body. Conversely, in all other case law, the court themselves has performed the translation from source to activity level by stating whether the alternative scenario is plausible or not while using tactical evidence, witnesses, and logical inference. To illustrate, case law from the District Court of Gelderland14 states that in theory, the suspect’s DNA profile can be present on the victim’s coat via secondary transfer. However, after reviewing all other evidence, such as the suspect’s motive, the suspect’s car placed at the scene, and the suspect’s flight behavior to Italy have ensured that the court does not consider the explanation of secondary transfer as plausible. Additionally, in other case law from the District Court of Gelderland15, the defense argued that secondary transfer cannot be ruled out since the DNA mixture evidence is obtained from movable objects. Be that as it may, the court rejected the defense since the suspect has not given any explanation for the presence of his DNA on the evidence items. Similarly, in several other case law16, the alternative explanation of secondary transfer was rejected because the suspect or accused did not provide sufficient justification for the presence of his DNA via secondary transfer. In particular, the case law from the District Court of Amsterdam and the Court of Appeal from Arnhem-Leeuwarden17 – which contained the prosecutor’s fallacy at the source level interpretations – did not indicate that these fallacies negatively influenced the suspect’s case, as the DNA mixture evidence was viewed in conjunction with other evidence to evaluate the alternative explanation of the defense concerning secondary transfer. Interestingly, in contrast to the previously mentioned case law, case law from the District Court of Midden-Nederland18, argues that the defense of secondary transfer cannot be excluded since there is not enough evidence to place the suspect at the scene or to demonstrate complicity. Even though the suspect’s profile is present in the DNA mixture evidence, the court states that it can be explained by the fact that the suspect has been several times inside the car where the firearm was found. Therefore, the court decided to acquit the defendant, considering that
10The District Court of Amsterdam, published 5 August 2019, ECLI:NL:RBAMS:2019:5690 11The District Court of Rotterdam, published 19 November 2019, ECLI:NL:RBROT:2019:8958 12The Court of Appeal s’-Hertogenbosch, published 30 January 2020, ECLI:NL:GHSHE:2020:242 13The Court of Appeal s’-Hertogenbosch, published 30 January 2020, ECLI:NL:GHSHE:2020:242 14The District Court of Gelderland, published 10 October 2019, ECLI:NL:RBGEL:2019:4497 15The District Court of Gelderland, published 8 October 2019, ECLI:NL:RBGEL:2019:4472
16The District Court of Amsterdam, published 27 June 2019, ECLI:NL:RBAMS:2019:4443; The District Court of Amsterdam, published 29 July
2020, ECLI:NL:RBAMS:2020:3567; The Court of Appeal of Arnhem-Leeuwarden, published 26 March 2019, ECLI:NL:GHARL:2019:2604; and The District Court of Rotterdam, published 15 July 2019, ECLI:NL:RBROT:2019:5623
17The Court of Appeal of Arnhem-Leeuwarden, published 26 March 2019, ECLI:NL:GHARL:2019:2604 and The District Court of Amsterdam,
published 29 July 2020, ECLI:NL:RBAMS:2020:3567
his DNA could have transferred onto the firearm via contact with other surfaces that contained the defendant’s DNA.
Overall, the court evaluates the alternative scenario regarding indirect transfer and contamination themselves by using witnesses, tactical evidence, and logical inference. When the defense presents an alternative explanation concerning direct transfer through a legitimate activity, then the court requests a forensic expert to evaluate this alternative explanation, comparing it to the crime-related activity. Conversely, only forensic DNA experts examine and assess the source level interpretations of the obtained DNA mixture evidence. These source level interpretations are used by the court to place the suspect at the scene of the committed crime.
CONTROVERSY REGARDING THE RESPONSIBILITY OF INDIRECT TRANSFER EVALUATIONS
In response to the increased frequency of encountering DNA mixture profiles and recent controversy regarding the evaluation of DNA mixture profiles, it is necessary for the forensic biology community to elucidate limitations in DNA mixture evidence interpretations. Subsequently, recommendations and guidelines must be provided to structure and objectify the assessment of DNA mixture evidence, particularly for the translation from source to activity level interpretations since members of the court are more interested in knowing how or when a certain DNA trace is deposited at the crime scene (Kokshoorn et al. 2017). This section will discuss the limitations and subsequent recommendations regarding the translation of complex DNA mixture evidence from source to activity level as described by the forensic science community.
Although the evaluation of DNA mixture profiles at the source level has become more standardized over the last decade, DNA mixture evidence needs to be examined with caution (Bleka et al. 2016; Aben, 2018; Kruijver et al. 2019; Bright et al. 2020). As forensic DNA examiners are able to recover and analyze minute quantities of DNA traces, it is possible that the recovered DNA profiles have no connection to the crime committed. Therefore, it is relevant to point out that finding a DNA ‘match’ between the reference DNA profile of the suspect and the trace DNA profile obtained from the crime scene does not necessarily mean that the suspect committed the crime. There are multiple ways for DNA to be present at the scene. For instance, DNA traces can be deposited via indirect transfer, or it can be recovered as a result of evidence contamination by crime scene investigators or forensic experts (Kokshoorn et al. 2017; Gill et al. 2020). A case-example of indirect transfer is David Butler’s case, in which Butler’s DNA was recovered from underneath a murdered sex worker (Gill, 2019). However, the DNA appeared to be indirectly transferred to the victim due to his dry-skin condition that increased the probability of legitimate-transfer. Moreover, DNA can already be present at the scene before the crime was committed, also known as background DNA (Buckelton et al. 2016; Gill, 2019). As a result, high LRs can lose their bearing in court when the defense disputes the manner of DNA deposition rather than its origin. Therefore, newly formulated questions are posed regarding the activity in which DNA could be deposited (Kokshoorn et al. 2017; Bright et al. 2020; Gill et al. 2020).
One limitation in the evaluation of complex DNA mixture evidence – and all other type of evidence in general – is the confusion regarding the hierarchy of propositions. It is essential to first carefully formulate propositions that do not cross the boundary of offense level evaluations, as this is the domain of the judiciary. However, this may be difficult since forensic experts receive established propositions at the activity level by the prosecution and the defense based on context information of the case. Hence, one proposition will be regarding innocent DNA transfer, whereas the other will be regarding crime-related DNA transfer (Robertson et al. 2016; Kokshoorn et al. 2017; Taylor et al. 2018). As a consequence of the language used in these propositions, some members of the court may falsely assume that the propositions contest the innocence or guilt of the suspect. To illustrate, in the case of sexual assault, an example of a proposition at the offense level for the prosecution would be “the victim was sexually assaulted by the suspect” while an example of a proposition at the activity level for the prosecution would be “the victim was penetrated by the suspect”. When not properly informed, the court might believe that the activity level proposition is equivalent to the offense level proposition, even though penetration could have been consented by the victim (Taylor et al. 2018; Fenton et al. 2020). Thus, it is necessary that the court views the weight of the complex DNA mixture evidence in conjunction with other obtained evidence and prior odds (Robertson et al. 2016; Twisk, 2019). Furthermore, forensic experts are often not informed or get insufficient information about the alleged activity to evaluate exclusive hypotheses, which may lead to incorrect or irrelevant propositions. However, Biedermann et al. (2016) argue that it is the task of the forensic scientist to request additional information to perform an adequate assessment of the acquired evidence.
After activity level propositions have been established, the forensic expert has to assess whether DNA has been transferred to a specific surface by criminal or legitimate activity, in which extra information about the transfer, persistence, and recovery of DNA traces is required (Gill et al. 2020). Currently, forensic scientists from various
fields are investigating these three phenomena and their mechanisms (Ruan, et al. 2018; Samie et al. 2020; Boyko et al. 2020). Despite the increasing body of literature on transfer, persistence, and recovery probabilities, there are still forensic experts who feel uncomfortable evaluating evidence at the activity level as this would be ‘overly speculative’ (Biedermann et al. 2016). Several limitations and issues were specified by these forensic experts.
First of all, even though the transfer, persistence, and recovery probabilities are increasingly studied in laboratories under specific circumstances, some forensic and judicial experts claim that this data should not be applied in real cases (Biedermann et al. 2016; Taylor et al. 2018). Experimental data is acquired from a closed and controlled environment, taking into account every aspect that might be of influence. Conversely, information obtained from the crime scene is subjected to many uncertainties since this environment is open and unregulated. Therefore, using probabilities calculated in settings with no ambiguity may result in overstating the evidential value at the activity level. However, when the forensic expert refuses to perform such an evaluation for the abovementioned reason, there is a risk that members of the court will try to conduct this assessment themselves, basing their judgment mostly on experience and (unfounded) non-scientific reasoning (Biedermann et al. 2016). Hence, assigning probabilities acquired during laborious experiments provides more substantiated insights when assessing both the proposition at the activity level for the prosecution and defense (Biedermann et al. 2016). Nevertheless, forensic experts should always express the uncertainties that may confound the used probabilities. Moreover, the forensic expert should mention explicitly that the evidential value may change when new information surfaces (Gill et al. 2020).
Additionally, forensic and judicial experts have raised the issue of using data from different countries and under diverse study conditions to provide evidential support for probabilities, which should not be generalized. The generalization of this data for individual cases could be disadvantageous for the defendant (Prakken & Meester, 2017; Biedermann et al. 2016). For instance, the fact that 27% of gunshot-related deaths were suicides in the United States cannot be used as a gunshot-related suicide probability in the Netherlands due to differences in legislation (Naghavi et al. 2018). Likewise, forensic experts should not use transfer, persistence, or recovery probabilities from a study simply because the data is available. Instead, they should determine whether the experimental data is applicable to the individual case (van Oorschot et al. 2019).
Finally,there is no consensus regarding the acceptability criteria for assigning probabilities at the activity level. To date, probabilities are often assigned based on expert knowledge and experience. Some judicial and forensic experts believe that these assigned probabilities are unfounded, causing the calculated LR to become unfounded as well (Prakken & Meester, 2017). However, according to Biedermann et al. (2016), assigning probabilities based on forensic expert knowledge and experience does not necessarily mean that these assigned probabilities are unfounded. To assess the probabilities, the forensic expert will look at peer-reviewed scientific publications, databases, or conduct a case-specific experiment. The forensic expert is not supposed to fabricate probability values without proper scientific background information. However, the forensic expert may use known models or principles about transfer, persistence, and recovery to estimate probabilities for a specific case. Therefore, Biedermann et al. (2016) argues that there is indeed a scientific foundation that supports the assigned probabilities. However, in order to avoid a black-box situation where the court and defense are not able to assess and scrutinize the validity of the evidential value, it is necessary for the forensic expert to explain what kind of information is used to support the assigned probabilities (Biedermann et al. 2016; Gill et al. 2020). When the forensic expert has reinforced the used probabilities with known and accepted models and can explain the technical basis to the working method, consensus on the acceptability criteria for assignment of probabilities is no longer necessary. Consequently, it could be argued that the acceptability of the probabilities has to be assessed on a case-to-case basis (van Oorschot et al. 2019). The final decision whether the assigned probabilities can be used as evidence is then determined by the court, instead of the forensic scientist (Biedermann et al. 2016; Taylor et al. 2018).
Besides primary transfer, forensic experts also establish propositions for secondary transfer evidence. Secondary or indirect transfer refers to the transfer of DNA evidence through an intermediate, e.g. a person shakes hands with a second person, after which the second person transfers the first person’s DNA to the crime scene (Taylor et al. 2018). When establishing propositions at the activity level, the term ‘secondary transfer’ cannot be described within the proposition as this is a component considered by the forensic expert in the evaluation of the proposition. To clarify, if the term was used in the propositions, this would be interpreted as the probability of transfer given transfer, which would lead to a probability of one (Kokshoorn et al. 2017; Gill et al. 2020). Interestingly, there is controversy regarding whether the court or the forensic expert should assign the probability of secondary transfer. According to Twisk (2019), the judge – who eventually assesses the case at the offense level – has to evaluate whether secondary transfer has occurred or not by using case-specific information. Similarly, Gill et al. (2020) state that the court should assess if it is likely for secondary transfer to have occurred based on other obtained evidence and information about the case. Thus, the judge has to decide whether a particular alternative scenario is plausible or not by looking at the prior odds before the forensic scientist can perform the activity level evaluation. Although, in theory, the judge could use the case information to determine a probability for the forensic expert
prior to the trial, this assigned probability will not entail a scientific foundation and consider all possible elements that factor into the indirect transfer probability. The question is, however, whether you should expect judges to substantiate their decision scientifically since members of the court are often not used to assigning probabilities. Furthermore, the court may incorrectly use the evidential value obtained at the source level as the evidential value at the activity level when not consulting a forensic expert (Gill et al. 2020). Therefore, this raises the issue whether the judge or the forensic expert should determine the indirect transfer probability – an issue that is not resolved yet.
Altogether, when possible, forensic scientists should try to perform evidence evaluations at the activity level since these could undermine the high evidential strengths obtained from source level evaluations. Furthermore, in order for the court to properly assess and scrutinize the obtained evidential value of the scientific evidence, the forensic expert should fully elucidate how the probabilities are assigned and mention any uncertainty associated with them. Importantly, there is controversy regarding whether the court or the forensic expert should assign the probability of indirect transfer, which remains unsolved.
THE JUDICIAL APPROACH VERSUS THE FORENSIC APPROACH
Several forensic experts have posed important limitations and corresponding recommendations regarding (complex) DNA mixture evidence evaluations at the activity level (See section 2). This section will examine how these limitations and recommendations relate to the use and evaluation of complex DNA mixture evidence in criminal cases by members of the court by analyzing case law derived from section 1.
Out of the fifteen selected case studies, one case law contained a DNA mixture evidence evaluation at the activity level performed by the NFI and TMFI (See section 1). The investigative judge requested an evaluation at the activity level with non-exclusive propositions as both propositions included the presence of the suspect’s DNA and the activity of moving the victim’s body by the suspect after someone else had moved the body or not. The use of non-exclusive propositions is not in accordance with the rules of Bayes’ Theorem. However, the case law only mentions the evaluated hypotheses and corresponding conclusions, making further analysis of the interpretations at the activity level impossible. Nevertheless, it should be noted that the obtained evidential strength from the DNA mixture evidence at the activity level was relatively low and viewed in conjunction with other obtained non-scientific evidence. Therefore, to the author’s belief, the use of non-exclusive propositions had no negative influence on the findings in the case as a whole.
Thirteen out of fifteen case laws include a defense regarding the possibility of secondary transfer. Interestingly, in all thirteen cases, the judge assessed whether secondary DNA transfer was plausible – without additional support from a forensic expert (See section 1). Moreover, in these case laws, no established propositions nor elaborate argumentation were presented regarding the assessment of secondary DNA transfer, contradicting the forensic view concerning DNA mixture evidence evaluations at the activity level (See section 2). To illustrate, case law from the District Court of Gelderland19, includes a defense regarding secondary transfer since the DNA mixture profiles were obtained from movable objects, including a beer bottle and IKEA bag. However, no further statement was given by the defense as to how the defendant’s DNA could be present on those movable objects. Moreover, the court viewed the DNA evidence in conjunction with witness statements, which indicated that the suspect was trying to create a false alibi. Therefore, the court ruled that secondary transfer was not plausible. Similarly, in case law of the District Court of Zeeland-West-Brabant20,the defense argued that the defendant’s DNA could be present on the victim’s coat via secondary transfer. However, the defense did not substantiate their statement with information as to how secondary transfer could have occurred. The court considered the defense regarding secondary DNA transfer not plausible since merely stating the possibility of secondary transfer is not enough substantiation.
In both case examples, the court reviewed all gathered scientific and non-scientific evidence, while using logical inference to decide whether secondary DNA transfer is plausible or not – which concurs with the view of Twisk (2019) and Gill et al. (2020). Although both authors mention that the forensic expert can aid the judge in their decision-making process by providing additional expert knowledge regarding transfer, persistence, and recovery of DNA traces, the judge seems to be of the impression that additional aid is not necessary when it comes to determining the possibility of secondary DNA transfer. Interestingly, after reviewing case law from the Supreme Court21, the Court of Appeal of ‘s-Hertogenbosch22 stated that in cases where the defense presents an alternative 19The District Court of Gelderland, published 8 October 2019, ECLI:NL:RBGEL:2019:4472
20The District Court of Zeeland-West-Brabant, published 18 November 2020, ECLI:NL:RBZWB:2020:5682 21The Supreme Court in The Hague, published 16 March 2010, ECLI:NL:HR:2010:BK3359
scenario that is very unlikely, the judge is not obliged to give an explicit refutation. If there is a possibility that the alternative scenario could be true, the judge has to assess whether this alternative scenario is plausible considering all the acquired evidence in the case. Thus, the judge has to look at both exculpatory and incriminating evidence in order to evaluate whether (innocent) secondary transfer or (criminal) direct transfer has occurred – which concurs with the view of Robertson et al. (2016) and Kokshoorn et al. (2017), where the judiciary has to evaluate the evidence at the offense level and establish prior odds. However, this raises the issue as to whether thorough procedure is followed when the judge refutes an activity level examination and immediately moves on to the offense level. It is questionable whether using this type of procedure is the proper measure to establish whether secondary DNA transfer is plausible or not.
Another case example23 showed that the court mistakenly used the evidential value at source level as the evidential value at activity level, while assessing the possibility of secondary DNA transfer themselves. Although the defense stated that secondary transfer occurred, the court reviewed the obtained DNA mixture evidence – which contained a high evidential value at source level – and determined that secondary transfer was not plausible. Moreover, the court dismissed the possibility of secondary transfer even though multiple individuals had access to the crime scene. Thus, the court argued that the defense of secondary transfer was not plausible based on the source level evidence interpretation, suggesting that members of the court may sometimes falsely use the source level evidence to make decisions about activities.
Overall, it appears that the court requests additional expert knowledge from the forensic scientist when the defense presents an alternative scenario concerning direct DNA transfer. In this case, the court cannot distinguish between a legitimate or criminal activity without extra information regarding the transfer, persistence, and recovery of DNA mixture profiles, particularly when both types of activities are closely related. However, when the defense presents an alternative scenario regarding secondary DNA transfer, members of the court seem confident to address this matter themselves. While Gill et al. (2020) concur with this “judge”-based approach in which the plausibility of indirect transfer needs to be decided by members of the court, it raises the issue as to whether thorough procedure is followed when the judge refutes an activity level examination by looking at prior odds instead – especially since these prior odds are not expressed and supported by numerical values. Therefore, this approach is not entirely consistent with the forensic approach where scientifically substantiated probabilities are assigned to establish the possibility of indirect transfer. A possible explanation for this controversy is the dissociation between two completely different fields that have emerged separately and now have to collaborate on pressing legal matters. Therefore, it is crucial that agreements are reached on topics such as the decision-making of secondary transfer examinations. To pose the question differently; “who is the best person for the job?”
DISCUSSION AND CONCLUSION
Evidence interpretations are performed at different levels, including source, activity, and offense level. It is generally accepted that the forensic DNA expert performs evidence evaluations at the source level for acquired DNA mixture profiles, whereas offense level interpretations are solely performed by members of the court. However, according to Twisk (2019), there is still some controversy regarding the responsibility of activity level evaluations between the forensic expert and members of the court. Therefore, this literature review is set out to determine whether it is feasible for members of the court to perform the translation process from source to activity level evaluations of DNA mixture profiles themselves, given the complex nature of this type of evidence. In reviewing the case law, it appeared that in the majority of cases with complex DNA mixture evidence, the evaluation at the activity level was carried out by the judge himself. However, one exception was observed24, where the judge inquired additional expert knowledge regarding transfer, persistence, and recovery probabilities of DNA mixture profiles to assess whether the obtained DNA mixture sample was directly transferred through a legitimate or criminal activity. The defendant had the opportunity to review all obtained evidence during his previous trial in the district court and adjusted his alternative scenario accordingly. In this case, the judge appeared not to be confident enough to perform a DNA mixture evidence evaluation at the activity level when the activity described by the defense is closely related to the activity described by the prosecution. Therefore, in order to distinguish between the two closely related activities, it is valuable for the judge to receive additional scientific information regarding the DNA mixture evidence, including the shedder status of the victim or suspect, the degradation rate of DNA, and the influence of environmental factors on DNA – which is provided by the forensic DNA expert (van Oorschot, et al. 2019; Otten et al. 2019). Interestingly, members of the court, in general, do find themselves capable of evaluating the alternative statement of the defense regarding indirect transfer using logical inference and reviewing all the acquired scientific and non-scientific evidence in conjunction.
The members of the court seem to justify this degree of certainty in their evaluation skills by their ability to assess whether the explanation of the defense for DNA transfer is plausible given all other obtained evidence. If the defense presents an alternative scenario as to how secondary DNA transfer could have occurred, the court will assess whether the obtained evidence provides any grounds for this presented scenario. If the obtained evidence in the case contradicts the provided scenario by the defense, the court decides that the alternative scenario regarding indirect transfer is not plausible. Thus, this case law review showed that the court members do not perform a complete translation from source to activity level interpretations of the complex DNA mixture findings. Instead, the court conducts an offense level evaluation by weighing all exculpatory and incriminating evidence and establishing prior odds, which are not expressed or supported by numerical values. This is in line with scientific literature, which states that the possibility of indirect transfer should be determined by the court (Twisk, 2019; Gill et al. 2020). Case law from the Supreme Court25 reinforces this practice by stating that the court is not obliged to provide a refutation when the presented alternative scenario by the defense is deemed “unlikely”. Thus, in the opinion of the court, supported by scientific literature, the court may decide upon the plausibility of indirect transfer when the substantiation of the presented alternative scenario is insufficient. However, it remains unclear what the terms “unlikely” and “insufficient” exactly entail regarding the alternative scenarios given by the defense, and whether these definitions are universally applied within and between countries. As a consequence, this issue requires further research.
From a forensic perspective, the DNA findings have to be assessed given propositions regarding indirect or direct transfer. To evaluate the findings given the established propositions at the activity level, the forensic DNA expert needs additional scientific and case-specific information, including the number and type of activities that precede the initially deposited DNA, and the type of biological material from which the DNA mixture profile is recovered (Buckleton et al. 2016; van Oorschot et al. 2019). However, despite the increasing literature, not everything has been elucidated about the probability of transfer, persistence, and recovery of DNA mixture profiles – complicating the activity level evaluations (Taylor et al. 2018). Biedermann et al. (2016) offer a solution for this problem by pointing out that case-specific experiments can be performed to estimate probabilities for the obtained evidence given the activity level propositions. However, the case law shows that alternative scenarios are not always (completely) provided or detailed by the defense, which could complicate the activity level evaluations of the forensic DNA expert if experiments need to be outlined. The possibility of indirect transfer is often stated by the defense without further explanation as to how this indirect transfer could have occurred. Therefore, to the author’s belief, in order to examine the plausibility of a given explanation, it is the responsibility of the suspect to elaborate on this explanation – which is supported by the case law from the Supreme Court. As the suspect can provide an alternative scenario for indirect transfer, he must be able to explain this scenario in explicit detail in order for the forensic expert to construct an experiment proving or disproving the plausibility of the alternative scenario, which is in the advantage of both the court to achieve the truth and the suspect to prove his innocence.
24The Court of Appeal s’-Hertogenbosch, published 30 January 2020, ECLI:NL:GHSHE:2020:242 25The Supreme Court in The Hague, published 16 March 2010, ECLI:NL:HR:2010:BK3359
Consequently, the “judge”-based approach that assesses the plausibility of indirect transfer by the judge is not entirely consistent with the forensic approach based on the Bayesian framework (See section 3). This controversy can be explained by the dissociation between two extremely different fields that have emerged independently and now have to collaborate on important legal matters. Over the past decade, the forensic science community has been rigorously scrutinized by several institutes and their evaluative reports, including the PCAST report (Bright et al. 2019; PCAST, 2016). It is no secret that forensic scientists can make mistakes due to human errors and different types of biases, namely contextual or confirmation bias. When the forensic expert is given irrelevant case-specific information it could influence the expert’s decision-making and impartiality (Curley et al. 2020). However, when assessing activity level propositions, additional case-specific information is required, which may increase the risk of bias in the forensic examiner. As a result of all the criticism on forensic sciences, forensic experts are trying to minimize bias and improve the objectivity of their evidence evaluations by optimizing or creating statistical models and tools, including Bayesian Networks (BN). Bayesian Networks is a statistical tool that provides a graphic representation of the assigned probabilities used for evidence evaluations at the activity level (Hicks et al. 2018). However, these probabilities are still assigned by the forensic expert, and many probabilities for specific circumstances remain unknown, possibly leading to speculative LRs as argued by Prakken & Meester (2017). Even though assigned probabilities are based on known models or principles of transfer, persistence, and recovery, the acquired probabilities may differ between forensic scientists, consequently resulting in different LR values (Biedermann et al. 2016; Hicks et al. 2018). However, Bayesian Networks can be very useful in demonstrating the forensic scientist’s thinking process and evaluating the sensitivity of particular assumptions and probabilities on the evidential strength. Therefore, further research into the improvement of Bayesian Networks and the assignment of probabilities concerning transfer, persistence, and recovery is needed to make evidence evaluations at the activity level more objective for the forensic scientist.
Similarly, members of the court are no exception to possible biases. Judges receive a tremendous amount of contextual information, which can either be case relevant or irrelevant. A study conducted by Rassin (2020) revealed that judges are indeed sensitive to contextual information, which may affect the judge’s presumption of innocence as it appears that contextual information predominantly results in an unfavorable judgment toward the suspect’s case. Additionally, Twisk (2019) states that the judge has an overestimation of knowledge when assessing scientific evidence, making them less inclined to ask questions to the forensic expert. This is in line with van Straalen et al. (2020), who showed that members of the court are inclined to overstate the evidential value when high LRs are obtained, such as with source level interpretations of DNA evidence. Consequently, when members of the court use source level interpretations of DNA mixture evidence to assess whether indirect transfer is plausible or not, they might have overstated the evidential value of the DNA mixture evidence as seen in case law from the District Court of Rotterdam26.
From the case law, it is unclear whether members of the court have made use of the deployed forensic advisors. A possible explanation might be that this case law research does not provide complete information about the proceedings in- and outside the court, as judges have deliberations in private council chambers and discussions with forensic advisors verbally, which are often not entirely documented (Twisk, 2019). Nevertheless, the forensic advisor can play a pivotal role in advising the judge regarding DNA mixture evidence, and whether activity level interpretations are desirable or not. When additional information is needed regarding complex DNA mixture evidence, they can inform the judge of this necessity, after which the judge can ask for further explanation by the defense regarding their statement of indirect transfer. The performance of an activity level evaluation can provide a more scientific foundation for the decision of the court regarding the evaluation of indirect DNA transfer scenarios presented by the defense. With this in mind, the forensic advisor could play a more prominent role in advising whether activity level examinations should be performed or not since they possess knowledge of both law and science, which may bridge the gap between the law portion and the science portion in criminal casework. Nonetheless, the final decision rests with the judge, who could remove the scientific aspects of the LR values obtained by the forensic expert and convert it into a subjective expression such as "likely", “insufficient” and "unlikely". However, this is not reason enough to dismiss the evidence evaluations at the activity level performed by the forensic expert. Activity level evaluations could provide a better understanding of the evidential strength of complex DNA mixture profiles, consequently reducing the risk of overstating source level evidence interpretations and preventing the use of source level interpretations as activity level interpretations by members of the court. Similar to how the PCAST report was a wake-up call for forensic scientists, members of the court should be made aware of the possibilities in the forensic field with regard to indirect DNA transfer examinations as some may not fully comprehend the relevance and rather presume that their own knowledge is sufficient. Awareness among judges can be raised by seminars, symposiums, and presentations where both where both judges and forensic experts are present or presenting. Moreover, applied courses can be provided where judicial and forensic experts discuss relevant case law containing complex DNA mixture evidence and whether an activity level evaluation by 26The District Court of Rotterdam, published 15 July 2019, ECLI:NL:RBROT:2019:5623
the forensic DNA expert would have been desirable in those cases.However, it remains the suspect's responsibility to explain the alternative scenario for indirect DNA transfer to such an extent that the judge is able to assess the decision of further forensic examination. Moreover, it should be noted that the forensic expert should not be seated in the chair of the judge, and the judge always has the final say.
ACKNOWLEDGEMENT
I want to take this opportunity to express my gratitude towards Wanda Remijn, forensic advisor at the District Court of The Hague, who has helped and guided me in writing my literature thesis by having discussion meetings and providing tips and feedback. I would also like to thank Christian van Dam, forensic advisor at the District Court of Gelderland, for giving me the opportunity to write a literature thesis under the supervision of a forensic advisor at one of the courts in the Netherlands. This work would not have been possible without the support of both Wanda and Christian. I also want to express my gratitude towards my examiner of the University of Amsterdam, Dr. Maarten Blom, for reading and assessing my literature thesis.
REFERENCES
SCIENTIFIC PUBLICATIONS
Aben, D. (2018). Een kwantitatieve interpretatie van complexe mengprofielen. Expertise en Recht, 5, 169-171.
Biedermann, A., Champod, C., Jackson, G., Gill, P., Taylor, D., Butler, J. et al. (2016). Evaluation of forensic DNA traces when propositions of interest relate to activities: analysis and discussion of recurrent concerns. Frontiers in Genetics, 7 (215), 1-12.
Bille, T., Weitz, S., Buckleton, J.S., & Bright, J. (2019). Interpreting a major component from a mixed DNA profile with an unknown number of minor contributors. Forensic Science International: Genetics, 40, 150-159.
Bleka, Ø., Storvik, G., & Gill, P. (2016). EuroForMix: An open source software based on a continuous model to evaluate STR DNA profiles from a mixture of contributors with artefacts. Forensic Science International: Genetics, 21, 35-44. Boyko, T., Szkuta, B., John Mitchell, R., & van Oorschot, R.A.H. (2020). Prevalence of DNA from the driver, passengers and others within a car of an exclusive driver. Forensic Science International, 307, 110139.
Bright, J., Kelly, H., Kerr, Z., McGovern, C., Taylor, D., & Buckleton, J.S. (2020). The interpretation of forensic DNA profiles: an historical perspective. Journal of the royal society New Zealand, 50 (2), 211-225.
Buckleton, J.S., Bright, J., & Taylor, D. (2016). Forensic DNA evidence interpretation. CRC Press, Boca Raton. Budowle, B., Onorato, A.J., Callaghan, T.F., Della Manna, A., Gross, A.M., Guerrieri, R.A. et al. (2020). Mixture interpretation: Defining the relevant features for guidelines for the assessment of mixed DNA profiles in forensic casework. Journal of Forensic Science, 54 (4), 810-821.
Curley, L.J., Munro, J., Lages, M., Maclean, R., & Murray, J. (2020). Assessing cognitive bias in forensic decisions: A review and outlook. Journal of Forensic Sciences, 65 (2), 354-360.
Depauw, S. (2020). In search of a free movement of forensic evidence: towards minimum standards to determine evidence admissibility? Journal of Forensic and Legal Medicine, 74, 102021.
Fenton, N., Jamieson, A., Gomes, S., & Neil, M. (2020). On the limitations of probabilistic claims about the probative value of mixed DNA profile evidence. Queen Mary University of London, United Kingdom.
Gill, P. (2019). DNA evidence and miscarriages of justice. Forensic Science International, 249, e1-e3.
Gill, P., Hicks, T., Butler, J.M., Connolly, E., Gusmão, L., Kokshoorn, B., et al. (2018). DNA commission of the International society for forensic genetics: Assessing the value of forensic biological evidence - Guidelines highlighting the importance of propositions. Part I: evaluation of DNA profiling comparisons given (sub-) source propositions.
Forensic Science International: Genetics, 36, 189-202.
Gill, P., Hicks, T., Butler, J.M., Connolly, E., Gusmão, L., Kokshoorn, B., et al. (2020). DNA commission of the International society for forensic genetics: Assessing the value of forensic biological evidence - Guidelines highlighting the importance of propositions. Part II: Evaluation of biological traces considering activity level propositions. Forensic
Science International: Genetics, 44, 102186.
Hicks, T., Taylor, D., Biedermann, A., & Champod, C. (2018). A template for constructing bayesian networks in forensic biology cases when considering activity level propositions. Forensic Science International: Genetics, 33, 136-146.
