2

Forensic Investigative Genetic Genealogy

OVERVIEW OF FORENSIC INVESTIGATIVE GENETIC GENEALOGY (FIGG)

In this session, panelists shared presentations on FIGG and spoke to various aspects of its use by law enforcement, including its process, benefits, challenges, ethical considerations, and the need for standards and regulations (see Box 2-1). With moderation by Heather McKiernan, RTI International, the session featured panelists with expertise in forensic science, law, medical ethics, and health policy.

Ray Wickenheiser, New York State Police Crime Lab System, began the session by reiterating earlier commentary from the law enforcement perspectives panel, calling attention to how FIGG is meant to be used by law enforcement only once other investigative methods have been exhausted on major violent crimes like homicides and sexual assaults where the suspect is still at large. He explained that after developing a DNA profile from crime scene evidence and getting no hits in Combined DNA Index System (CODIS) databases, the DNA is sent out for single nucleotide polymorphism (SNP) analysis to search commercial genealogical databases for potential familial matches. Echoing previous law enforcement representatives, Wickenheiser suggested that surviving victims and family members be considered in deliberations around the use of advanced forensic DNA technologies. FIGG also may be used in cases of missing persons or unidentified human remains, said Wickenheiser. This can generate investigative leads by building family trees from distant relatives, he explained. Wickenheiser argued that FIGG is scientifically and technically sound, with limitations, and that its use by law

enforcement to solve crimes is ethical if done with appropriate guardrails and informed consent from database participants.

Process of FIGG

Claire Glynn, University of New Haven, described FIGG as a multidisciplinary process, because it utilizes multiple technologies from the areas of genetic genealogy, consumer DNA testing, genealogical research, and forensic science to analyze matches and build family trees. Glynn noted that millions of people have sent DNA samples to consumer DNA websites. However, Glynn clarified that for the purposes of FIGG, only a small percentage of these samples are available for analysis by law enforcement. She explained that in the wake of the use of FIGG in the Golden State Killer investigation in 2018, several commercial DNA testing companies changed their terms of service and privacy policies to explicitly prohibit the use of their databases for law enforcement. However, she noted that both FamilyTreeDNA and GEDmatch allow the use of their databases for FIGG; together they have DNA information for about 4 million people. Both sites allow consumers to opt in or out of allowing their data to be used for law enforcement, said Glynn, so the number is somewhat smaller when accounting for those who have opted out.

Wickenheiser and Glynn went into further detail on the multidisciplinary FIGG process. First, evidence from the crime scene is analyzed by an internal or external laboratory that conducts SNP testing, said Wickenheiser. There are a few options for SNP testing, Glynn said; one option can be conducted in house and specifically eliminates the discovery of any health-related genetic information. Wickenheiser noted that traditional sample analysis that is used to look for direct matches in the CODIS uses short tandem repeat (STR) analysis, while SNP analysis is used to look for kinship matches in a genealogical database.¹ The DNA file is uploaded to one of the genealogy websites using a law enforcement–specific portal, said Glynn, and an algorithm compares the data with user data and looks for matching SNPs. The results give information about the amount of shared DNA, and a statistical calculation determines the potential relationship type to the person of interest (e.g., second cousin). Results may also provide information about the potential ancestry of the person of interest, she

___________________

¹ SNPs and STRs are both genetic markers used in forensic DNA analysis, but they differ significantly in their structure, applications, and advantages. STRs are the primary markers used in forensic DNA profiling for human identification, paternity testing, and linking suspects to crime scenes. They are the basis for national DNA databases like CODIS (Butler, 2006; Federal Bureau of Investigation, n.d.-b; National Institute of Justice, 2011). SNPs are useful for analyzing highly degraded DNA samples, predicting phenotypic traits (e.g., eye or hair), and providing information on biogeographical ancestry (Weir & Zheng, 2015).

explained. When kinship matches are identified, investigators can use these relative names to search publicly available information and begin to build family trees to identify a person of interest.

Glynn emphasized the importance of documenting every step of the FIGG process and using best practices and standards for genealogical research. Documentation serves as evidence that “the result was achieved in an ethical and responsible manner within the boundaries of all the laws and policies that exist,” she said. The result of a FIGG analysis, Glynn continued, is the identification of a potential suspect and the use of traditional STR testing to exclude or include the person. If the STR profile from the crime scene does not match the person of interest, the investigator continues building the family tree and looking for other possibilities.

Before using FIGG, said Glynn, an investigator must ensure that the case type meets the criteria set by the U.S. Department of Justice (DOJ) guidelines and relevant state law (see Box 1-4 in Chapter 1). In addition, the case must meet the criteria for the use of the two genealogy databases. For the latter, the investigator must have already conducted traditional forensic DNA analysis consisting of STR testing and looking for matches in CODIS. If state law allows familial searching using STR profiles in the state DNA database, this search must also have been conducted. If there are no hits in CODIS, and no other investigative leads, FIGG may be conducted, said Glynn. She called for investigators to follow established genealogical standards, create quality assurance protocols specific to FIGG, increase education and awareness to encourage more people to opt in to databases, and balance protecting public safety with individual privacy through responsible and ethical use. Glynn offered the National Technology Validation and Implementation Collaborative guidelines for establishing FIGG programs as a source of valuable information on responsible and ethical implementation of FIGG programs (Wickenheiser et al., 2023). She also encouraged increased uploading of people’s DNA into FIGG databases, so long as they have an informed understanding of what this process entails, considering the huge potential for FIGG to help prevent and solve crimes.

Distinct Characteristics of FIGG

Wickenheiser said that while FIGG involves many steps and interactions between multiple proven, traditional forensic and scientific technologies, it is essentially a new process of obtaining an investigative lead, which can eventually be confirmed through follow-up investigation and standard STR direct comparison. Erin Murphy, New York University School of Law, disagreed with this statement, suggesting that FIGG is “totally different” from traditional investigative methods and that these differences create real risks. Murphy described some of the major differences between FIGG and

traditional forensic DNA analysis (Table 2-1). She explained that the DNA analyzed is different—traditional analysis looks at noncoding “junk” DNA, while FIGG uses SNP analysis that reveals many information-rich genes. Traditional forensic analysis is done in regulated government laboratories, whereas FIGG relies on largely unregulated technologies, she said. Traditional analysis implicates only the target suspect and a few closely related individuals, whereas FIGG brings potentially thousands of people into the investigation.

An individual’s DNA contains information that is relevant not just to the moment that it is collected; it can also reveal information about prior generations and unborn descendants, said Murphy. She noted that this makes DNA different from other types of crime scene evidence, which do not provide investigators with such information. Furthermore, she said, as technology advances, a DNA sample gains more capacity to reveal sensitive information. If misuse by bad actors occurs, a person cannot revoke access to their or their ancestors’ DNA. Finally, Murphy echoed Tsosie’s comments from the opening session in noting that, unlike other types of crime scene evidence, DNA is monetizable and has tremendous value beyond crime-

TABLE 2-1 Differences Between Traditional Forensic DNA Analysis and Genetic Genealogy

NOTE: bp = base pairs; CODIS = Combined DNA Index System; GTCA = guanine, thymine, cytosine, and adenine; QA/QC = quality assurance and control; QAS = quality assurance standards; STR = short tandem repeat.

SOURCE: Presented by Erin Murphy on March 13, 2024.

solving. To this point, Murphy noted that outside of law enforcement, a robust conversation is occurring about the risks of commercial DNA analysis; for example, the military has advised members not to use recreational DNA testing because of potential privacy and security risks.

Public Perspectives on FIGG

There is strong public support for FIGG, said Christi Guerrini, Baylor College of Medicine. Guerrini told workshop participants about her research on public and expert perceptions of FIGG and the policy implications of these perceptions. She presented empirical research from a national, representative sample showing strong and consistent public support for using FIGG to identify violent perpetrators, but less support for nonviolent crimes. She then explained that FIGG fundamentally relies on consumer DNA databases and these databases exist only if people contribute genetic samples and consent to their use by law enforcement. If database participants begin to have doubts about the use of FIGG, they might withdraw their consent or remove their DNA profiles. Experts are very aware of the need for public support, said Guerrini, and they support the creation of guardrails to ensure public trust in the technology.

Guerrini’s study found that over 90% of the public support the use of FIGG to identify violent perpetrators, unidentified human remains, and unidentified babies, or to exonerate wrongfully convicted individuals (Figure 2-1). However, the level of support for FIGG drops to about 50% for its use to solve nonviolent crimes. There is broad public consensus around the need for regulation and oversight of FIGG, Guerrini reiterated.

Guerrini’s research project also included input from a “policy Delphi,” with 34 expert participants with backgrounds in law enforcement, forensic science, genealogy, law, ethics, and victim advocacy. In the policy Delphi, the experts came together several times over the course of a year to discuss, debate, and identify priority issues. The group identified nine priority issues in four areas (Figure 2-2), and then explored policy options in these areas. Guerrini provided details on two of the priority issues and the associated policy options. The Delphi experts agreed that patchwork governance is a major concern and identified four policy options: (a) a federal FIGG law, (b) state model laws, (c) a finalized DOJ policy, and (d) conditioning grants on best practice compliance. The second area of concern was law enforcement participating in databases against their terms of service. To address this issue, policy options included an outright ban on the practice, a standard database consent approach (e.g., all opt-in or all opt-out), and the development of a database that is used only by law enforcement that is populated with genetic profiles donated specifically for this purpose. Guerrini emphasized that the policy Delphi process included relevant parties from a broad

CONSIDERATIONS FOR USE OF FIGG BY LAW ENFORCEMENT

Opportunities

Wickenhauser and Glynn mirrored the previous panel in emphasizing the significant benefits of FIGG in solving cold cases and enhancing public safety (see Box 2-2). Wickenhauser noted that FIGG has been instrumental in identifying perpetrators who might otherwise remain at large, thereby preventing future crimes. Glynn highlighted the success of FIGG in identifying a perpetrator in more than 1,000 cases, demonstrating its potential to resolve many more with increased public participation and funding. Before speakers addressed the ethical challenges and scientific limitations of FIGG, Wickenheiser clarified what he considered to be ethically sound components of FIGG. He argued that individuals have no right to privacy regarding forensic evidence discarded at a crime scene and reminded the audience that these forms of forensic evidence have been collected and analyzed to solve crimes for decades. In addition, he noted that databases and public records

spectrum of fields. She echoed discussions from the previous two panels when reiterating that any policymaking in this area should similarly be informed by a diverse group.

are routinely used to solve crimes. What is new with FIGG, he said, is the use of databases with information on relatives. With informed consent, however, Wickenheiser said that public databases can be used ethically in FIGG. He added that people have the right to make any use of their own DNA, including helping law enforcement solve crimes.

Challenges and Ethical Considerations

Reliability and Limitations

One major risk of any new technology, said Wickenheiser, is whether it is reliable and accurate. The process of FIGG encompasses several technologies and approaches, so its reliability is dependent on the reliability of these underlying approaches. There may be limitations to the utility of FIGG if the sample at the crime scene has issues with quantity, quality, or mixtures of samples. If crime scene DNA is of limited quantity or quality, said Glynn, it may not be appropriate to utilize all of it for FIGG. Wickenheiser said that both STR and SNP analysis are well-established and reliable methods of analyzing DNA. He explained that kinship matching via DNA is a scientifically sound practice, although genealogical and public databases that are used to find relatives may have errors or omissions. Wickenheiser emphasized that any leads gathered through this process must ultimately be confirmed with STR direct comparison, which he said is an established, accurate, and reliable method of forensic evidence analysis.

Murphy disagreed with Wickenheiser’s assertion that FIGG and its underlying testing are always reliable, raising concerns about the legitimate risks of FIGG use. She said that DNA testing always involves the risk of wrongful arrest and detention, and she illustrated this reality by outlining a case in Washington where Joao Monterio was wrongly arrested for a 25-year-old cold case based on a genealogical profile built using DNA from the crime scene (Carter, 2024; Monterio v. Cormier). He was released after spending three-and-a-half years in jail, during which time he lost his job and his wife died (Carter, 2024). In another case, Murphy highlighted, a man was charged with murder because his DNA was found on the fingernails of the decedent (Worth, 2018). Because it was a death penalty–eligible case, his attorney accessed his medical records to look for mitigation factors. He found that the man had been in a detox center at the time of the murder and the DNA had likely been transferred via a pulse oximeter (Worth, 2018). Murphy said that these cases demonstrate that advanced forensic DNA technologies, even with “reliable” STR confirmation, can result in innocent people being arrested. There are other anecdotes like these, said Murphy, but it is difficult to know the true extent of mistakes because of

a lack of systematic information about the use of advanced forensic DNA technologies.

Informed Consent

Murphy pushed back against Wickenheiser’s conceptualization of informed consent, suggesting that using an informed consent model analogous to that used for research is misguided in the case of FIGG. She noted that when one uploads their DNA to a website, they are consenting to the matching of their own DNA—not their brother’s, children’s, or cousin’s. However, she advised that the nature of DNA data creates opportunities for FIGG investigations to become overbroad and intrusive, when considering factors such as their ability to trace distant relatives, the realities of routine rule-breaking by genealogists with little consequence, and the surreptitious DNA sampling of third parties. The consent model is a fiction, she said, because FIGG is used specifically to find people who are not in the database and who did not consent.

Disparate Impact

Another risk of FIGG and other forensic DNA technologies, said Wickenheiser, is the potential for it to be used in an inequitable way or to have a disparate impact. Greater than 80% of crimes are conducted by individuals within the same socioeconomic group as the victim, he said. When choosing to use limited resources on tools like FIGG, it is important that cases are chosen in an equitable fashion, and resources are deployed to solve crimes in disadvantaged areas. Furthermore, Wickenheiser said, there is an ethical imperative to use FIGG and other technologies to find perpetrators who are still at large, as well as to exonerate wrongfully suspected or incarcerated individuals. Advanced forensic DNA technologies should be applied to all unsolved cases equitably and quickly, he said, because unsolved cases are a major public safety issue.

Scope of Use and Impact

The use of FIGG brings up concerns about overbroad and intrusive investigations, said Murphy, with a lack of accountability and incentives to break rules and push ethical boundaries. The limited rules that have been put into place have been broken with little consequence, she said, and law enforcement officials have engaged in behavior that many would consider unethical, including

• going beyond terms of service to get access to data;
• using a fake name and account to access data;
• surreptitiously sampling many nonsuspects and suspects in order to speed up the investigation; and
• retaining elaborate family trees in law enforcement databases.

Murphy added that using forensic DNA technologies like FIGG presents the risk of spillover effects, such as undermining public health benefits of DNA analysis. People may be hesitant to have their DNA analyzed for health risks if they fear that their information will be used for law enforcement purposes; this risk is particularly relevant in communities of color where there is historic mistrust of both the health care system and law enforcement. There are also risks to public safety, said Murphy. For example, a rape victim may refuse to report her rape and do a rape kit for fears that her genetic information will be used to implicate herself or a relative.

Considerations for Implementation

Like all powerful technologies, said Wickenheiser, FIGG should be implemented using established best practices. He suggested that FIGG best practices would include accreditation to ensure quality of testing, a system of oversight, terms of service for databases with penalties for misuse, and the proper handling of DNA samples from all parties. Wickenheiser pointed to the published National Technology Validation and Implementation Collaborative (NTVIC) guidelines that describe appropriate guidelines, policies, and procedures for the use of FIGG (Box 2-3; Wickenheiser et al., 2023). In order to follow these best practices, he said, training and education will be necessary for investigators and other parties within the legal system.

Many people have described FIGG as the “Wild West” of emerging forensic DNA technologies, said Glynn, but she said that there are now “some reins on the horse.” For the first few years, forensic scientists were “building the plane while they were flying it” because there were no rule book or protocols to follow. Now, she said, there is a road map of how a FIGG case should be carried out, who should be involved, and the NTVIC guidelines to follow (Wickenheiser et al., 2023). Since forensic DNA was first used in 1984, training, education, and method validation have been critical to ensuring its reliability and appropriate use, said Glynn. With FIGG, the same process should be applied. Adhering to policies, laws, and standards is essential for fulfilling the potential of the technology, as well as for maintaining public trust and protecting the privacy and safety of the public.

Several speakers in this and the previous session referred to Maryland Criminal Procedure § 17-102 (2024; see Box 2-4), with both Murphy and

Dan Katz, Maryland State Police Forensic Sciences Division, commending the collaborative process that led to its development and adoption. Murphy said that while she would have preferred an outright ban on FIGG, she views both the process and the outcome as good. The law is a comprehensive regime, and it includes the voices and perspectives of many stakeholders. In response to a question, Murphy described some of the components of the law. A warrant is required to begin the process of FIGG, which means there is judicial oversight from the beginning. Under the law, FIGG is treated as a “last resort as opposed to first option,” said Murphy. It requires following the rules of databases, and explicit notice and informed consent of nonsuspects or nontargets. There are public reporting and data requirements, and a route allowing for the defense to get access to the technology. Murphy said that law enforcement needs to provide information about return on investment when the public is paying for advanced forensic DNA technologies; tracking metrics about the use of these technologies is critical for understanding how and when they are used, and these are “not super complicated questions.” The Maryland law also requires that laboratories and genealogists be licensed, said Murphy, but the licensing system does not yet exist. Some stakeholders have complained that this is going to “put the brakes” on FIGG because it will take a long time to ramp up; Murphy said that taking time to do things properly is how it should be when implementing new technologies.

Currently, many rules around FIGG are simply “the fox guarding the hen house,” said Murphy. Instead of these self-regulations, she called for laws to be developed through a democratic process with the involvement of relevant parties; she indicated that these laws should have accountability mechanisms to incentivize proper utilization and disincentivize misuse.

In response to a question from a workshop audience member, panelists identified several essential components of appropriate and ethical implementation of FIGG, many of which are directly reflected in Maryland Code, Criminal Procedure § 17-102 (2024) and Utah Public Safety Code 53-10-403.7 (2023):

• a diverse set of stakeholders to set policies (Wickenheiser & Glynn);
• a unified system of overarching guidelines, with room for state differences (Wickenheiser);
• laboratory accreditation (Wickenheiser);
• support from the federal government to bring together stakeholders (Glynn & Guerrini);
• data transparency (Murphy);
• acknowledgment of the power systems at play (Tsosie); and
• education and engagement with the public (Guerrini).

REFLECTIONS

Following the four panelist presentations, Krystal Tsosie, Arizona State University and the discussant for this session, offered her reflections on the discussion around FIGG. The use of FIGG, she said, is not devoid of complex cultural, ethical, societal, and legal concerns. Recalling a statement by Wickenheiser about the high percentage of crimes in which the victim and the perpetrator are from the same economic class, Tsosie said that it is easier for tribal people to justify the use of forensics when violence is perpetrated onto Indigenous women by nontribal members. However, when violence is perpetrated by members of one’s own community, it creates “uncomfortable questions” about the use of the community’s genetic information to solve crimes. Another issue that stood out for Tsosie was the discussion around storing DNA indefinitely to use it in the future with more advanced technologies. She noted that every time the technology improves, the risks to privacy for individuals and communities increase. “We need to answer questions about how long it is appropriate to store DNA, who has access to it, and what standards are used for future testing. For genetic information that is held by commercial DNA testing companies, what safety measures are put in place to ensure that the DNA is used for the purposes for which consent was given? What happens when companies are bought and sold? Do the privacy protections and terms of service follow the data?”

A variety of people is involved in the process of FIGG, said Tsosie, and it is not clear what training they have, whether they are certified, and if their expertise is validated. Who are considered “FIGG experts,” she asked, and what communities do they or do they not represent? Are there FIGG ethical and professional standards, and if so, who is writing these standards? Murphy noted during her talk that there are multiple instances of people involved in FIGG breaking their own rules with no consequence.

Finally, Tsosie addressed the fact that genetic databases do not reflect the general population. Incarcerated individuals are disproportionately likely to come from communities of color (National Academies of Sciences, Engineering, and Medicine, 2023). As a result, they are also overrepresented in criminal DNA databases, so the use of such databases can exacerbate racial injustices. Genealogy databases have less information on historically marginalized populations, in particular Native Americans. This is the result of complex sociopolitical processes, and it is not going to be solved by simply encouraging people to provide their DNA to the system, she said. Tsosie asked how to ensure that advanced forensic DNA technologies ameliorate rather than exacerbate inequities, and how to ensure that they are used with respect for individual and community autonomy. She noted that there are no easy answers to these questions, but that the discussions at the workshop generate necessary conversations around these issues.

This page intentionally left blank.