1

Public Trust and the Landscape of Law Enforcement Use of Advanced Forensic DNA Technologies

To better understand key considerations around law enforcement use of advanced forensic DNA technologies, the Committee on Law and Justice and the Computer Science and Telecommunications Board at the National Academies of Sciences, Engineering, and Medicine (National Academies) held a workshop titled “Law Enforcement Use of Probabilistic Genotyping, Forensic DNA Phenotyping, and Forensic Investigative Genetic Genealogy Technologies.” The workshop was organized in response to Executive Order 14074 (Box 1-1), issued in May 2022, and was held on March 13 and 14, 2024. The order focused on advancing effective, accountable policing, as well as criminal justice practices around algorithmic approaches to policing; it directed the National Academies to hold a workshop to explore the different approaches. The workshop, organized according to a statement of task (Box 1-2) and sponsored by the National Institute of Justice, focused on three specific advanced forensic DNA practices: probabilistic genotyping, forensic DNA phenotyping, and forensic investigative genetic genealogy (see Box 1-3 for working definitions of these terms).¹

ORGANIZATION OF THE PROCEEDINGS

This proceedings describes the workshop panel presentations and the discussion that followed each panel. The chapters are organized around

___________________

¹ Where possible, this proceedings refers specifically to forensic DNA technologies by name (e.g., probabilistic genotyping). At times, the three technologies in focus in this workshop are referred to more broadly as “advanced forensic DNA technologies.”

the key topics of the workshop, with some chapters including summaries of related content from multiple panels. Chapter 1 introduces these key technologies and gives an overview of their use in law enforcement, and discusses the ethics, equity, and accountability issues that have emerged as the technologies are being implemented. Chapters 2, 3, and 4 each focus on a specific technology and explore the unique aspects of the technology, the related risks and ethical issues, and considerations for implementation. Chapter 2 focuses on forensic investigative genetic genealogy, Chapter 3 focuses on probabilistic genotyping, and Chapter 4 focuses on forensic DNA phenotyping. Chapter 5 looks at forensic DNA technologies from a global perspective, exploring their use in Australia, Europe, the United Kingdom,

and Switzerland. Finally, Chapter 6 summarizes workshop discussions on research gaps and funding needs for emerging forensic DNA technologies and describes the key themes of the workshop as identified by individual speakers.

The full meeting agenda and biographical sketches of planning committee members and workshop presenters appear in Appendixes A and B, respectively. Appendix C provides a bibliography of resources mentioned by workshop speakers.

This proceedings has been prepared by the workshop rapporteur as a factual summary of what occurred at the workshop. The views contained in the proceedings are those of the individual workshop participants and do not necessarily represent the views of other workshop participants, the workshop planning committee, or the National Academies.

OPENING REMARKS

Alicia Carriquiry, Iowa State University and chair of the workshop planning committee, began the event with brief opening remarks. Carriquiry noted that the workshop was a direct response to Executive Order 14074 on advancing effective, accountable policing and criminal justice practices that enhance public trust and public safety (Exec. Order No. 14,074, 2022). She further noted that the executive order emphasizes the need to ensure public trust in law enforcement, and that the workshop would consider a range of considerations around implementation of advanced forensic genetic technologies, including risks associated with privacy, civil rights, and equity, which have the potential to erode public trust. Carriquiry recognized the planning committee, a multidisciplinary group of experts convened by the National Academies, who collaborated to develop the workshop agenda and select speakers.

Following these remarks, Lucas Zarwell, National Institute of Justice (NIJ), provided additional context for the workshop, stating that NIJ has been involved in forensic science for 50 years, with the mission of supporting state, local, and tribal advancements through the development of new technologies and the dissemination of best practices. NIJ utilizes a “listen, learn, and inform” approach and places great value in the collaborative spaces that the National Academies creates. Zarwell explained that the 2015 National Academies report Support for Forensic Science Research is still used by NIJ to guide its actions and strategic planning. The theme at the American Academy of Forensic Sciences meeting this year, noted Zarwell, was “Justice for All.” Workshops like this, he said, are necessary for exploring the power of new technologies while also considering how they can be used to ensure justice for all.

SETTING THE STAGE: ETHICS, EQUITY, AND ACCOUNTABILITY

Following brief introductory remarks by members of the workshop planning committee, the event began with a panel highlighting the topics of public concern at the center of the executive order that motivated the workshop. Speakers discussed issues of ethics, equity, and accountability related to the use of advanced forensic DNA technologies by law enforcement. Five panelists with expertise in civil and human rights, privacy, civil liberties law, biomedical ethics, public health, and community research offered remarks on these topics, with moderation by Sarah Chu, Perlmutter Center for Legal Justice at Cardozo Law.

Context and Framing

Chu opened the workshop by emphasizing the importance of discussing ethics, equity, and accountability in the context of law enforcement’s use of technologies like probabilistic genotyping, forensic DNA phenotyping, and forensic investigative genetic genealogy. She began the discussion by asking panelists what issues they viewed as central to considerations of law enforcement use of advanced forensic DNA technologies. Tierra Bradford, Leadership Conference on Civil Rights, pointed to historical and ongoing disparities in the criminal legal system and the potential for new tools to exacerbate or entrench them. She explained that she approached this discussion with a focus on civil rights, stating that the criminal legal system, as it currently exists, is unfair and inequitable. Bradford then expressed concern that the introduction of new technologies, if not properly verified, validated, and implemented, could exacerbate these inequalities. Furthermore, she pointed to the need to address the recurring issues of transparency and

accountability, public understanding of and engagement with the criminal legal system, and the complexity of private, for-profit vendors serving public institutions.

Jennifer Lynch, Electronic Frontier Foundation, explained that she approached the workshop through the lens of privacy and surveillance. Top of mind for her, she explained, was the extent to which advanced forensic genetic technologies could implicate those outside of the criminal legal system, including close relatives of suspects (in the case of forensic investigative genetic genealogy), and even the broader public (in the cases of forensic investigative genetic genealogy and forensic DNA phenotyping). Regarding forensic DNA phenotyping, Lynch noted that disseminating an image to the public of an individual suspected of perpetrating a crime has risks. For example, an image of a Black suspect could implicate an innocent member of the Black community, particularly given known challenges with eyewitness and cross-race identification. Lynch provided an example of these risks in practice, pointing to a recent case with the Edmonton Police Service in Canada. She suggested that law enforcement use of advanced forensic DNA technologies represents a significant change in how DNA is utilized for criminal investigations, because of its expanded scope of impact, and the implementation of the underlying science to law enforcement practice poses significant questions.

Drawing on her expertise in bioethics, Daphne Martschenko, Stanford University, stressed the importance of considering the ethical and social implications of genomic research and the need to include diverse voices in these discussions. Martschenko outlined her focus on the ethical, legal, and social implications of human genetic and genomic technologies; on innovations to expand the communities included in weighing risks and benefits of genomic research; and on the development of harm mitigation and benefit promotion strategies, each of which are necessary for responsible use of the technologies at the center of the workshop. She emphasized the necessity of grounding conversations about the use of advanced forensic DNA technologies in ethical and social considerations.

Finally, Krystal Tsosie, Arizona State University, introduced herself as a citizen of the Navajo Nation and an Indigenous geneticist-bioethicist, with expertise in Indigenous genomic data sovereignty.² She explained that her work is characterized by a strong commitment to ethical practices, Indigenous community engagement, and data sovereignty, and that this work directly intersects with law enforcement use of advanced forensic DNA technologies to identify missing or murdered Indigenous persons. Tsosie shared how Indigenous peoples and communities have sometimes been

___________________

² Indigenous genomic data sovereignty involves the right of Indigenous peoples to have agency and authority over DNA collection from their nations.

pressured to contribute their genetic information to databases to identify victims and rematriate³ their remains to their families and communities and reflected on how such pressures could be applied around use of technologies like forensic investigative genetic genealogy.

Expanded Impact of Advanced Forensic DNA Technologies Context and Framing

Lynch directed attention to the differences between advanced forensic DNA technologies and more traditional forensic approaches to DNA analysis and discussed the broader public implications of adopting new DNA technologies. In addition to the potential expanded scope of impact, she suggested that such tools represent a major shift from the previous system of comparing DNA from a crime scene with a database of convicted felons. Now, Lynch said, criminal investigations have the potential to implicate not just people directly involved in the criminal legal system, but extended family members, and even the broader public. Members of the public are involved when their genetic information from genealogy databases is used in a criminal investigation, or when a computer-generated image is released and might implicate members of an entire racial community. This expanded scope of impact, Lynch suggested, means that forensic investigative genetic genealogy and forensic DNA phenotyping represent a significant change from previous uses of DNA in the criminal legal system. She emphasized the importance of considering advanced forensic DNA technologies through the lens of their impact on individuals, communities, and privacy interests.

Law Enforcement Use of Proprietary Technologies

Central to the discussion was a recognition by the speakers of the role of private industry in the development and sale of genetic technologies for use by public institutions—in this case, law enforcement. Ethics in industry operates in a unique space, said Tsosie, as compared to the highly regulated areas of academic research and clinical practice. She outlined how companies involved in developing, marketing, and distributing advanced forensic DNA technologies acquire DNA as an asset, with databases and companies being bought and sold frequently. The question, she asked, is what is happening to the informed consent and the terms of use to which individuals

___________________

³ Rematriation is a holistic movement led by Indigenous women, focused on the revitalization of Indigenous cultures, knowledge systems, and sacred feminine relationships with lands and waters as a means of decolonization, healing, and restoration of balance (Sogorea Te’ Land Trust, 2021). Tsosie also qualified rematriation as a shift in power dynamics as it relates to data ownership, sharing that it is more restorative in equity than other approaches.

who contributed their DNA originally agreed? She expressed concern about the possibility that original privacy policies may not be honored by the new owner and that databases could become “open for mining” by other entities and uses to which contributors did not consent. Tsosie explained how this potential “data co-option” raises ethical and jurisdictional concerns, particularly for historically marginalized communities.

Lynch spoke to the reality that private companies are heavily involved in forensic sciences. Based on her professional experience, she described the difficulty of auditing the work of these companies, in part because of the complexity of the technology and in part because the source code and external verification and validation studies may not be readily available. This puts defendants at a disadvantage, remarked Lynch. She explained that while some defendants have gained access to the source code, they have obtained this information by hiring their own experts and agreeing to protective orders. She explained that information they uncover about the validity or accuracy of the technologies typically remains under protective order, so future defendants cannot access the information for their own defenses. Lynch emphasized that trade secrets protections for private companies also complicate the use of forensic investigative genetic genealogy and forensic DNA phenotyping by justice system actors because the data and the algorithms that companies use to analyze DNA are not readily available for the public or defendants to examine, or for independent validation or peer review. Bradford expressed her perception of the incongruity between a lack of transparency and the possibility that such information can have a substantial impact (in the form of state power) on individuals and communities.

Accuracy and Transparency

Bradford emphasized the significant negative impact that engagement with the criminal legal system can have on individuals and communities. Even a few days spent in jail can result in the loss of housing, employment, and/or custody of children. She indicated that transparency, accuracy, and accountability of advanced forensic DNA technologies are essential when weighed against the potential for negative impacts to individuals and downstream impacts on their communities. Bradford called for adequate training and education, risk management, continued monitoring, evaluations, and regulations around advanced forensic DNA technologies. She also stated that if these measures indicate that advanced forensic DNA technologies are found to have disparate impact, they should not continue to be used by law enforcement.

Advanced forensic DNA technologies are used to try to solve crimes that are “heinous” and may be associated with families who have been

seeking justice for a long time, said Lynch. She further noted that without transparency, oversight, and ongoing evaluation of law enforcement use of forensic DNA phenotyping, its use could lead to the implication of innocent persons. She explained that this is a particular problem for any technologies making use of facial predictions in cross-race crimes, because individuals typically have a difficult time identifying people of a different race. Furthermore, in crimes where an eyewitness has a memory of the perpetrator, a computer-generated image can influence and even override an individual’s original memory. This loss of the eyewitness perspective is a loss to the criminal investigation, she said.

Issues of accuracy and transparency lead to related concerns about the lack of standardization for advanced forensic DNA technologies, said Martschenko. She suggested that inconsistent or inadequate regulation of DNA samples leaves the door open for upstream and downstream misuse and abuse by actors in the criminal legal system, a fear born from real examples of scientific exploitation of historically marginalized populations. Martschenko argued that greater transparency can help to dispel the notion that genetic technologies are foolproof, objective, accurate, and without fault. She noted earlier the lack of incentives for researchers to consider downstream outcomes and called for mechanisms to ensure transparency and accountability in the use of all advanced forensic DNA technologies.

Regulation and Oversight

On the topics of privacy and consent, Lynch explained that she struggles with advising people to provide their DNA to any database, given that their information is most likely not explicitly protected by law, creating significant privacy concerns for users. Law enforcement can access non–law enforcement genetic genealogy databases via legal processes or under specific conditions, and Lynch noted that commercial genetic genealogy databases have effectively created a national, unregulated database that could enable identification of up to 95% of people in the United States with Western European ancestry. While no comprehensive federal regulations specifically govern the use of genetic genealogy databases by law enforcement, some states have enacted laws to regulate this practice and the U.S. Department of Justice (DOJ) has issued interim policies recommending the use of non–law enforcement genetic databases only for unsolved violent crimes where other methods have failed (Box 1-4). In contrast, said Lynch, criminal justice DNA databases (e.g., Combined DNA Index System⁴) have been the

___________________

⁴ The Combined DNA Index System, or CODIS, is a computer software program that operates local, state, and national databases of DNA profiles from convicted offenders, unsolved crime scene evidence, and missing persons (Federal Bureau of Investigation, n.d.a).

subject of much public debate, and many regulations are in place to control access to the data. Lynch pointed to Maryland Criminal Procedure Code § 17-102 as a leading example of state efforts to govern the use of forensic investigative genetic genealogy in criminal investigations by addressing the misuse of genetic information (see Box 2-4 in Chapter 2). She noted that the law balances public safety interests with privacy rights by prohibiting unauthorized disclosure of genetic data, establishing criminal penalties for violations, and allowing civil damages for wrongful disclosures.

Indigenous Communities

Turning to issues specifically impacting Indigenous people and communities, Tsosie explained that the rate of violence against Native populations and Indigenous people is several-fold higher than against other groups. “Genetics is a poor solution to violence,” said Tsosie, quoting an X (formerly Twitter) post from the prominent bioinformatician Larry Hunter (2019). She suggested that law enforcement should focus resources on efforts to reduce race-based violence from being perpetuated onto Indigenous people, rather than seeking post hoc measures such as the expansion of genetic databases to identify perpetrators and victims. Advanced forensic DNA technologies present several unique risks for Indigenous communities, said Tsosie, and these risks are still being identified and navigated.

Tsosie questioned the meaning of Indigenous in the context of DNA, noting that there are 574 federally recognized tribes in the United States, as well as state-recognized or unrecognized tribes, and they all have distinct cultures, languages, demographics, genetic histories, and phenotypes. She stressed that there is a need to question how using DNA as a signifier of race and ethnicity fits in with this reality. Using DNA to identify race opens the door to racial genomic profiling, said Tsosie. She explained that due to the comparatively large size of families in Indigenous communities, one DNA test could implicate a huge portion of the community. Furthermore, relying on DNA technologies to solve crimes puts the onus on victims, their families, and their communities to contribute their genetic information. Many Indigenous families, said Tsosie, hear the message that, because Indigenous peoples are not well represented in genomic databases, it is their fault that their missing and murdered relatives cannot be identified, which can be a form of victim-blaming and coercion (Tsosie et al., 2021). She emphasized the importance of protecting tribal data and ensuring informed consent, stating that Indigenous communities need the opportunity to develop and control their own resources and policy frameworks in this space to determine whether these advanced forensic DNA technologies are beneficial rather than harmful to the community.

Downstream Implications

Martschenko noted that with new technologies, there is often a considerable gap between research that occurs upstream and the downstream use of the new tools. She explained that there are few incentives or mechanisms in academia to encourage researchers to consider the broader implications of their research or to include the voices of those who will be impacted in the research process. She emphasized that the benefits and risks of a new tool are often identified from the perspective of those creating the tool,

rather than from the perspective of those who will be impacted. Marginalized communities have already been disproportionately harmed by the criminal justice system, said Martschenko, and advanced forensic DNA technologies have the potential to further these disparities. With emerging and evolving technologies, she said, it is often people who are already vulnerable who shoulder the potential harms.

A workshop audience member asked Martschenko to comment on how the concept of open science relates to the arena of genetic research. Martschenko acknowledged that there is a tension between the move toward open access, data sharing, and transparency, and the fact that studies can be used by actors with ill intent. For example, she explained how white supremacist groups have used social and behavioral genomics studies to justify violent, racist acts under the veneer of science. She said that there is a need to think deeply about the contexts in which data sharing may be appropriate and to what extent, and those in which sharing is not appropriate. For example, she explained that it may be inappropriate to share data collected via a years-long community engagement process with researchers who have no ties to the community and may want to use the data in ways that go against the values and goals of the community. Martschenko strongly encouraged researchers to think more critically about the context in which data are collected, for what purpose, and from whom, and to use this information to make decisions about when data sharing is appropriate. She also emphasized the value of implementing ethical frameworks for researchers that consider the social context of technology use.

Public Understanding

An audience member asked the panelists about the importance of educating the public on the uniqueness of genetic information and helping them to understand the implications of sharing their genetic information for various purposes. Tsosie answered that the public has a perception of DNA as being the “scientific arbiter of truth,” partly because of its representation in popular culture and media. She emphasized the importance of communicating the limitations of advanced forensic DNA technologies, particularly in a courtroom setting, as a more realistic representation of the capabilities of the field. Bradford explained that in her professional capacity, she and her colleagues often “work backwards” to educate the public about a new topic. For example, when a person’s civil rights are violated (e.g., they are wrongly incarcerated), she and her colleagues look back at what happened in the investigation and where mistakes were made, and then seek to understand the science behind the technologies that were involved in order to educate the public about the potential harms of certain technologies. She noted that this type of case-by-case approach is not ideal for large-scale

education efforts, and that there is a need for scientists and advocates to work together to proactively educate the public on emerging technologies. Martschenko echoed Tsosie’s emphasis on the importance of helping the public understand the limitations and ethical considerations of genomic technologies—namely, conveying the complexity of genetic information and moving away from the deterministic view of genetics. She then pointed to ongoing efforts to improve genetic literacy, which could be adapted and applied to a criminal legal context.

Martschenko provided two examples of organizations engaged in genetic literacy efforts. One such organization engaged in the promotion of genetic literacy is Personal Genetics Education & Dialogue, a nonprofit in Boston that creates curriculum for K–12 educators, holds public town halls, and partners with other groups to put on public-facing events on the ethical and social implications of genetics.⁵ Additionally, the National Science Foundation has funded research aimed at overhauling high school biology curriculum in order to shift toward a more complex view of human genetic variation. Preliminary assessments of these approaches to genetic literacy, she said, have shown that teaching people about the probabilistic, complex nature of human variation has an effect on how they think about themselves and others and reduces genetic determinism and racial essentialism. Martschenko said that while this work is not specifically focused on the use of genetic information by law enforcement, a similar approach could be beneficial in helping the public better understand new technologies and their limitations.

Erin Murphy, New York University, added commentary as an audience member, suggesting that there is a “fundamental misapprehension” on the part of the public that the criminal legal system is monitoring and regulating the use of genetic information. The public assumes that there are scientific standards for admissible evidence, and that judges will be making decisions about the use of advanced forensic DNA technologies. However, she said, these technologies are used primarily in the investigative phase. Law enforcement does not need to disclose that they used genetic information to identify a suspect, or that they collected genetic information from a nontarget person in order to help the investigation. “There will be no constitutional motions” on these issues because they are not relevant to the actual criminal trial. The public may believe that the legal system has oversight of advanced forensic DNA technologies, she said, but in practice this is generally not true.

Martschenko emphasized that although even the experts do not yet have a full understanding of advanced forensic DNA technologies or a complete picture of how they are being used, they should not hesitate to

___________________

⁵ For more information, see https://pged.org

bring the public into these conversations. It is important to be transparent about the uncertainty and lack of consensus among experts, and to allow the public to use their lived experiences to contribute to the conversation.

Ethical and Socially Responsible Implementation

Given all the challenges and risks associated with advanced forensic DNA technologies, said Chu, what would ethical implementation look like? Bradford said that it is critical to continually monitor and evaluate the use of the technologies and to not hesitate to put a ban or moratorium in place if necessary. For example, if assessments reveal that a technology has a disparate impact on a particular community, its use should be stopped. Martschenko added that while some technologies may eventually have benefits, using them while in their infancy presents greater risk of harm than benefit. If the tools are implemented while still untested, it can result in a situation where public mistrust is exacerbated, and the potential benefits of the tool do not get realized because of the damage already done. It is important to recognize the limited validity and utility of technologies and prevent them from being used inappropriately in high-stakes, life-or-death situations, she said.

Several speakers emphasized the need to think about public safety in the affirmative rather than focusing on solving crimes after they have occurred. Tsosie said that conversations and resources should be shifted toward preventing harm and violence in the first place. Bradford agreed and rhetorically asked what would happen if we invested more in health, education, violence intervention, and communities that are overpoliced.

Training of law enforcement and others using the technologies is key to ethical implementation, said Bradford, and agencies need to think about disparate impact and potential remedies, such as education and community-engaged participatory research. Communities that are most impacted—particularly communities of color—need to be brought into the conversation. These racialized groups are the “least considered [and] the most impacted,” said Bradford. Martschenko added that in addition to impacted communities, a variety of other stakeholders need to be involved in conversations about the use of advanced forensic DNA technologies, including law enforcement, victims, industry, and the media. These groups need to be in dialogue about the risks and benefits of technologies and how to implement them in an ethical way. Martschenko emphasized that ethical implementation of these technologies requires a recognition of the social context in which they are implemented. Technologies do not operate in a vacuum; it is the responsibility of all stakeholders, from developers to law enforcement, to ensure that technologies are used appropriately and responsibly and do not cause further harm.

SURVEYING THE LANDSCAPE

The second workshop session featured perspectives of advanced forensic DNA technology users working within the criminal legal system. Panelists with professional experiences including criminal investigation, policing, forensic science, forensic laboratory management, and criminal prosecution engaged in moderated discussion about their day-to-day experiences with forensic DNA technologies and the challenges and opportunities they present. Craig O’Connor, New York City Office of Chief Medical Examiner, served as moderator.

Scope of Use

O’Connor began the discussion with the observation that probabilistic genotyping (PG) and forensic investigative genetic genealogy (FIGG) are more commonly used in law enforcement than forensic DNA phenotyping (FDP), and asked panelists about their own experiences with these technologies.

Speakers collectively identified PG as a dominant tool in forensic DNA analysis, widely adopted by more than 100 law enforcement and state laboratories in the United States. Leigh Clark, Florida Department of Law Enforcement, explained that PG helps forensic DNA analysts interpret DNA found at crime scenes—even when the DNA is limited in quantity, damaged, or mixed from different people—and then compare it with DNA from potential suspects.

Panelists identified FIGG as an emerging forensic technique that combines advanced DNA sequencing with traditional genealogical research to generate investigative leads. Paul Belli, International Homicide Investigators Association, said that FIGG has gained prominence for its role in solving high-profile cold cases, such as the identification of the Golden State Killer. He explained that FIGG has been instrumental in solving cases where other investigative methods have failed, providing new leads and revitalizing stagnant investigations.

Panelists described FDP as less commonly utilized by law enforcement, when compared with PG and FIGG. Belli and Clark both noted that their encounters with FDP were limited to a small number of cold cases. Belli emphasized that FDP use requires a thoughtful approach and an acknowledgment that the results are not definitive, but instead are investigative leads aimed at generating conversation around a case. Ensuring that communities do not perceive FDP results as a definitive picture of the subject, Belli noted, will require additional work by law enforcement. Clark agreed that FDP can potentially help generate investigative leads, particularly when there is no information about what the perpetrator looked like. She shared an

example of a case where a behavioral profile suggested that the perpetrator was a middle-aged White man but phenotyping led the investigation in a different direction. Clark said that in contrast to Jennifer Lynch’s experience, she has found vendors to be forthcoming about the capabilities and limitations of phenotyping products in their terms of service. She added that while phenotyping can be useful for some attributes—such as freckles, which is controlled by one gene—it cannot account for changes to people over time (e.g., weight). She indicated that FDP may have use in unidentified human remains cases where only anthropology has been used to discern outward physical characteristics associated with race or ethnicity.

Opportunities and Benefits

Unsolved Cases and Victim Services

Ray Valerio, Queens District Attorney’s Office, opened the discussion of benefits with a reminder of the thousands of unsolved murders and sexual assault cases around the country. Thousands of victims, families, and communities are impacted by these cases, he said, and law enforcement has a duty to help solve these crimes. Valerio suggested that tools like PG and FIGG are invaluable in solving the most egregious crimes, including those committed against victims from historically marginalized communities, noting that in New York City, most violent crime victims are people of color. Mark Pooley, University of North Texas Center for Human Identification, noted that hundreds of unidentified human remains across the country have not been tested, representing a tragedy for families that do not know what happened to their loved ones. He posed that some families may never learn what happened to their loved ones unless advanced forensic genetic tools are used in their cases.

On the topic of unsolved cases, Pooley emphasized the pressure placed on detectives to solve cases. He offered a firsthand account of the experience of working on an unsolved homicide:

Pooley indicated that families appreciate knowing that detectives have used every tool at their disposal. On this note, Clark encouraged stakeholders to take a victim-centered approach to the implementation of advanced forensic genetic technologies.

Investigative Advantages

Another key benefit of advanced forensic DNA technologies, said Jeremy Triplett, Kentucky State Police Central Forensic Laboratory, is that they can provide more meaningful information than previous methods. For example, conducting probabilistic genotyping on a complex DNA mixture can help to exclude and include people from a sample that was previously uninterpretable. O’Connor explained that as forensic DNA technologies such as PG advance, forensic professionals can learn valuable investigative information from samples that were once too small or too low quality to interpret. Clark agreed that tools such as PG have provided significant enhancement to law enforcement’s investigative capacity. She explained that PG has proven particularly effective in allowing DNA results to be used and interpreted in criminal cases, for inclusion and exclusion purposes, including in cold cases and postconviction exoneration.

Valerio spoke to this benefit from the perspective of attorneys. To illustrate, he shared that since 2010, guidelines from the Scientific Working Group on DNA Analysis Methods have required that any positive DNA match—where the DNA evidence links a suspect to a crime—must include a statistical measure. He noted that this is also mandated by the Federal Bureau of Investigation’s Quality Assurance Standards. Valero remarked that this means when DNA evidence is presented in court, it must come with a statistical analysis that shows how likely it is that the DNA match is accurate. He said that including these statistics is crucial because it helps juries understand the strength of the DNA evidence presented and can help them evaluate all the evidence in the case and decide whether the suspect is guilty beyond a reasonable doubt.

Increased Accuracy and Reliability

In addition, the use of PG has improved the consistency of complex mixture interpretation. Triplett said that while PG results may differ depending on certain variables, the likelihood ratios generated within a laboratory are likely to be quite similar regardless of who is doing the work. He noted that the National Institute of Standards and Technology’s Organization of Scientific Area Committees for Forensic Science Registry (2018) has a standard for validating PG systems and that a standard for assigning propositions to likelihood ratios is being developed (Academy Standards Board, 2021). These standards will bring more consistency in PG results both within and between labs, Triplett said. Clark said that discussions around PG often overcomplicate the technology. She argued that PG “doesn’t really do anything that we weren’t already doing” with prior DNA mixture interpretation approaches; it simply does it quicker and better than

previous methods and considers multiple variables simultaneously. Laboratories conduct internal validation with their own known source data to ensure that the software functions correctly, she explained. Clark said that PG makes the interpretation within the laboratory more consistent from person to person, and it conducts the analysis without mathematical error and introduces much less subjectivity when statistical weight is calculated.

Technology’s Role in Investigative Work

Pooley outlined the practical reality that advanced forensic DNA technologies do not replace traditional investigative work but supplement it. Law enforcement must still “put in the [investigative] work,” but the information gleaned from advanced forensic DNA technologies can help point the investigation in a certain direction or provide new leads in a cold case. Belli concurred and said that advanced forensic DNA technologies are not used in a vacuum. Investigators do the work to get to the point of using one of the advanced forensic DNA technologies, and they use the information to continue the investigation. For example, the Golden State Killer case involved years of investigation, hundreds of interviews, and inclusion and exclusion of suspects before the case was ultimately cracked using FIGG. Valerio added that DNA is rarely the sole piece of evidence in a case, and forensic DNA technologies simply add information to the totality of the evidence for the jury to consider.

Clark said that the public sometimes mistakenly views forensic science as a silver bullet. Advanced forensic DNA technologies are enabling better utilization of the evidence that is collected, she said, but they are not magic. In order to make effective use of PG, for example, Clark explained that genetic material needs to be present in sufficient amounts at the crime scene, and the material then needs to be carefully collected and transported to the laboratory and analyzed. She emphasized that none of these steps operate independently of one another, explaining that each must be carried out successfully for PG to be utilized. While the integration of these technologies into investigative workflows has led to significant advancements in helping identify possible suspects, Clark suggested that there is room for additional education and training to improve law enforcement and lab execution of these processes from start to finish.

Challenges and Ethical Considerations

Regulation and Oversight

One challenge in the use of advanced forensic DNA technologies, said Belli, is inconsistent or lacking regulations among different jurisdictions.

As new technologies emerge, he said, it can be the “Wild West” until regulations catch up with practice. Belli noted that the conversations at this workshop, and the multiple perspectives represented, are critical to moving law enforcement use of advanced forensic DNA technologies forward in a thoughtful way. There is a need to find the middle ground between two distinct sides when considering the regulation of a new technology, said Dan Katz, Maryland State Police Forensic Sciences Division, describing his involvement with FIGG regulation in Maryland. Katz explained that in 2019, a bill was introduced to ban the practice of FIGG in the state. After this bill failed, relevant parties on both sides attempted to educate legislators about the issue, and progress was made but no bills were passed. In the third year, said Katz, the stakeholders came together to form a working group to find middle ground. Both sides conceded elements of what they wanted, he said, but everyone was relatively satisfied with the final regulations. Katz shared that his experiences working on Maryland Criminal Procedure Code § 17-102 have given him hope that relevant parties can collaborate productively as new technologies emerge.

In addition to legislation, several groups are working on best practices and guidelines for advanced forensic DNA technologies, said Valerio. The DOJ (2019) published an interim policy on FIGG in 2019 that lays out the requirements for the use of FIGG by law enforcement. The National Technology Validation and Implementation Collaborative has brought together stakeholders to develop guidelines and programs for new technologies; they recently published a guideline for FIGG programs in Forensic Science International: Synergy (Wickenheiser et al., 2023; see also Box 2-1 in Chapter 2). Courts are also weighing in on the appropriate use of advanced forensic DNA technologies, Valerio said. PG has been the subject of numerous evidentiary hearings across the country, and this process is building legal standards for the use of advanced forensic DNA technologies.

Using Results at Trial

Advanced forensic DNA technologies can also present unique challenges for attorneys attempting to address findings in legal proceedings, said Valerio. For example, he noted that it can be difficult to convey the statistical significance of the information provided by PG. Valerio suggested that lawyers are not generally comfortable with statistics, a subject area that is not taught in law school and does not appear on licensing exams or in continuing legal education courses. Despite these challenges, Valerio explained that prosecutors are responsible for presenting information in a way that appropriately upholds the scientific conclusions and does not mislead the jury. Prosecutors and defense attorneys sometimes try to simplify information for the jury, said Valerio, but this approach can result in a misstatement of the scientific conclusion.

Preparation and training of attorneys is thus critical to the accurate conveyance of the results of advanced forensic DNA technologies in the courtroom, Valerio explained. He noted that select jurisdictions require prosecutors to complete training in forensic evidence, pointing to the National District Attorneys Association as one source of this training for prosecutors. Valerio concluded by calling for attorneys to increase their understanding of PG and statistics, while recognizing that the scientist on the stand is the expert.

Clark agreed that attempts to simplify the findings of advanced forensic DNA technologies usually result in overcomplication. It is the expert scientist’s responsibility to educate the judge or jury, she explained, and attorneys’ attempts at simplification can necessitate further clarification from the scientist. Clark pointed out that this issue is compounded by a training gap within the forensic science community. Biologists and DNA analysts are not inherently statisticians, and they too may misstate results when attempting to simplify. Additionally, varying laws and practices across jurisdictions mean that evidence must be presented differently, requiring attorneys and scientific experts to be well informed and prepared for these requirements. Clark underscored the importance of pretrial collaboration among attorneys and experts to ensure shared understanding of forensic results, and as a space with room for improvement in the implementation of advanced forensic DNA technologies.

Audience Questions

During the Q&A section, Martschenko responded to a proposal for a “victim-centered” approach to implementing advanced forensic DNA technologies. She suggested that “victims” could also include those who are unnecessarily and adversely impacted by the criminal justice system; she asked stakeholders to consider that while the success stories of using advanced forensic DNA technologies to solve crimes are easy to find, it is harder to see the hidden victims who have been unjustly arrested or incarcerated. Clark responded that while wrongful involvement in the criminal legal system is a risk, these technologies can also be used to exonerate innocent individuals. She suggested that using these technologies more broadly outside of the crime investigation arena—such as to identify missing persons or to exonerate—could help illustrate the benefits of the technologies and indicated that more funding could be warranted to expand the scope of access of these tools to populations that have historically faced adversity within the criminal legal system.

During the question-and-answer session, Erin Murphy, New York University—a workshop panelist and audience participant—questioned the statement that PG results are more accurate and consistent than previous

methods. She said that through her legal work as well as reviews of the literature, she has observed large variation across laboratories and across analytical threshold choices. She noted that there are many choices to make in using PG, including which software to use, whether an allele is drop-in or drop-out,⁶ and which population frequency files to use, and suggested that these choices can result in very different output. Murphy noted that these choices could be made based on the desired result, for example, setting a threshold so that it doesn’t exclude the suspect. She asked the panelists for their thoughts on this observed variation. Clark responded that analytical thresholds are very dependent on the physical environment and specific instrument that are used and set as a product of each laboratory’s validation, which makes interlaboratory comparisons difficult. Some variables and choices are unlikely to make a big difference to the results—such as choice of population frequency database used—while other variables have not yet been addressed by the forensic community. For example, the qualitative interpretation of a result could vary between analysts; one might conclude that a result provides “strong evidence” in support of a hypothesis, while another concludes that it is “very strong evidence.” The onus is on the forensic science community, said Clark, to conduct more training on how to explain results not as absolutes but on a continuum.

___________________

⁶ An allele is a variant form of a gene at a specific locus on a chromosome. More specifically, it is one of two or more versions of a DNA sequence that can occur at a given genomic location. Individuals inherit two alleles for each gene, one from each parent. These alleles can be identical (homozygous) or different (heterozygous; National Human Genome Research Institute [NHGRI], 2024). Alleles contribute to the genetic diversity within a population and can influence various traits, including physical characteristics, disease susceptibility, and responses to environmental factors. The interaction between different alleles can result in dominant, recessive, or codominant expression of traits (NHGRI, 2024). Allele drop-in and drop-out are phenomena that can occur during DNA amplification and analysis, particularly in forensic genetics. Allele drop-out refers to the failure of an allele to amplify during a polymerase chain reaction, resulting in its absence from the final DNA profile, which can lead to false homozygous results or partial profiles (Shestak et al., 2021). Allele drop-in, on the other hand, is the appearance of an additional allele in a DNA profile that does not belong to the true contributor(s) of the sample, often caused by contamination, artifacts in the amplification process, or stochastic effects in low-template DNA analysis (Gill et al., 2012).

This page intentionally left blank.