Incidental Findings in Forensic Genetics

DNA is designed to build organisms, store information, and forensically speaking to answer questions. But what unfolds when it reveals secrets nobody intended to ask?

UPHAWU asked me to write about how forensic DNA analyses preserve identity and respect: restoring names, supporting the families of missing persons, protecting dignity and strengthening justice across Africa. It is a subject I care deeply about, and one I have spent much of my career trying to make more possible.

I just returned from the 31st Congress of the International Society for Forensic Genetics (ISFG), held in Montréal, Canada, where I had the privilege of co-leading a pre-congress workshop with my colleagues on the Forensic Databases Advisory Board on emerging ethical challenges in forensic genetics.

My session focused on incidental findings in forensic DNA profiling. The conversations that emerged from that workshop stayed with me. They raised questions that I thought were particularly fitting for this inaugural issue. Preserving identity is more than finding a name, and with all its benefits to society sometimes DNA reveals far more than we intended to know.

At first glance, incidental findings may appear to be a relatively narrow issue within the broader landscape of forensic genetics. Yet as forensic DNA technologies become increasingly powerful and databases more widespread, the boundary between what a forensic analyst can identify and what they are ethically, professionally or legally expected to report is becoming increasingly difficult to define.

WHAT IF?

IF. Two simple letters, with two meanings.

IF herein stands for incidental findings. But it also poses the question at the heart of this discussion: What if?

What if forensic DNA reveals something no one sought?

What if DNA tells us more than we asked it to?

What if that information changes someone’s life?

During the workshop, what particularly struck me was the number of forensic practitioners who described a sense of responsibility when confronted with an incidental finding: information that was not part of the original forensic question, but which could potentially have significance for an individual, their family, or even a wider investigation. For many, the instinct was clear: if we see something that could matter, surely, we have a responsibility to do something with it. But should we?

This question becomes particularly important in forensic genetics because the purpose of forensic DNA analysis typically is fundamentally different from that of clinical or research genetics. A forensic analyst is usually asked to answer two specific questions that are two sides of the same coin: whether the source of the crime-scene sample is the same as that of the reference sample or is someone else the source of the crime-scene sample? The analysis is therefore undertaken within a defined evidentiary and investigative framework. Yet advances in DNA technologies, public genetic genealogy databases, and artificial intelligence (AI) mean that the same data may disclose information about that individual extending far beyond that original question.

This outcome creates an uncomfortable ethical territory.

When does an observation become a finding?

When does a finding become an obligation?

Who decides what the analyst’s responsibility should be?

In the context of forensics, those results are most often handed directly to the investigator, and then the prosecutor, court, and only later the individual concerned. The weight of that responsibility, and the impact of those findings, weighs heavily on forensic analysts, which was predominantly conveyed to me. Forensic analysts are not trained as genetic counsellors. But they understand, inherently, the nature and impact of the finding. One analyst spoke of finding a case of incest based on STR profiles. This finding goes beyond a complicated kinship association. It now speaks to criminality.

The discussion at the ISFG workshop made clear that IF is not simply a technical question. It sits alongside several other emerging ethical challenges the workshop set out to map: the relational nature of genetic data when searching forensic investigative genetic genealogy databases, the use of forensic genetics in immigration control, the application of AI methods in forensic practice, and the ethical implications of forensic epigenomics for equity-seeking population groups. It also raises a fundamental question about the role of the forensic scientist: is the analyst responsible only for answering the question that is posed, or also for acting upon incidental information uncovered along the way?

WHEN DNA ANSWERS QUESTIONS NOBODY INTENTED TO ASK

IFs are one of the oldest, and today one of the fastest-growing, ethical challenges in forensic genetics. While they have existed since the earliest days of parentage testing, technologies such as forensic genetic genealogy and AI are increasing both their frequency and their complexity.

How do we define Incidental findings?

An IF is genetic information revealed during DNA analysis that falls outside the purpose of the investigation (Caenazzo, Tozzo & Dierickx, 2020). In other words, DNA answers a question nobody intended to ask and the data may not be relevant to the initial tasking.

Clinical genetics has long dealt with IFs, but forensic genetics is different. In medicine, there is usually informed consent, an ongoing clinician–patient relationship and opportunities for counselling. In forensic genetics, DNA is analysed for an investigative purpose, not to provide personal genetic information. The distinction fundamentally changes the ethical questions we must ask.

WHAT CAN WE ACCIDENTALY DISCOVER?

IFs are not confined to one type of forensic case. They can arise wherever DNA is analysed: crime scene investigations, kinship testing, molecular autopsies, decedent identification, or reference samples used for validation studies and quality assurance practices. An analyst working from a routine STR profile might notice a non=paternity of an assumed father in a missing persons case, which is an IF but potentially opens up issues of infidelity.

Advances in forensic genetic genealogy, searchable databases and AI allow for deeper scans of the human genome and interpretations beyond human identification purposes. Thus, opportunities for discovery and the potential for IFs have dramatically increased and eventually law enforcement may want to exploit Ifs to solve cases.

More importantly, genetic information no longer remains solely within the laboratory. Once genetic data enter the public domain which may occur during judicial proceedings, others may draw conclusions that extend well beyond the original purpose of the investigation. In other words, once data leave your hands, it does not stop generating IFs.

Tone should pause and take notice of the significance of the reach of IFs: an IF does not require a dramatic case.

A routine staff sample used to validate a new protocol may trigger an IF. That alone is sufficient to not treat IFs as a fringe problem, and it is why the ethical frame matters even before entering results in courtroom proceedings.

IFs cannot be dismissed as rare or exceptional. They are an inherent possibility whenever forensic DNA is analysed, one that every forensic laboratory should anticipate, and every laboratory should implement policy to address.

WHEN TECHNOLOGY BECOMES AN ‘IF’ AMPLIFIER

Not all forensic DNA technologies create the same risk of IFs.

Traditional STR profiling generates relatively limited personal biological information. However, IFs can still occur, for example through sex-marker analysis or kinship testing. But IFs are comparatively uncommon.

Massively Parallel Sequencing (MPS) or Next Generation Sequencing (NGS) fundamentally change that picture. Instead of analysing a relatively small number of markers primarily for identification purposes, these advances-capability technologies can generate information relating to ancestry, externally visible characteristics, age, health and biological relationships (Kayser, Branicki, Parson & Phillips, 2023).

Forensic genetic genealogy takes the issue further still. Here, discovering genetic relationships is no longer incidental. It is the investigative strategy itself.

AI and publicly available genetic and genealogical data mean that all information is no longer confined to the laboratory and all sorts of questions can be pursued without having the requisite expertise traditionally needed to mine complex genetic data. The same data may generate new findings months or years later, potentially by entirely different people.

Yet despite these expanding capabilities, forensic genetics still has no widely accepted framework for managing secondary or IFs (Sajantila & Budowle, 2016; Miller et al., 2023).

WHEN FINDINGS EXTEND BEYOND THE INDIVIDUAL AND BECOME STRUCTURAL

Forensic genetic genealogy takes the problem to another level.

One search can implicate not only a person of interest, but a much wider network of living relatives. The key ethical distinction is between people who voluntarily uploaded their DNA and relatives identified only because someone else provided a reference sample. The second group did not submit DNA, consent to a search, or anticipate becoming involved in a criminal investigation.

Thus, relatedness is not merely an incidental by-product of forensic genetic genealogy; it is structurally built into the method. Governance therefore needs more than access controls. It requires clear limits around proportionality, validation, documentation and how far investigators should extend a kinship network. Budowle and colleagues argued in 2024 for disciplined hypothesis testing and transparency (Budowle, Baker, Sajantila, Mittelman & Mittelman, 2024). These measures cannot remove the network effect, but they can make its use more accountable, transparent, and documented.

DOES “ACTIONABLE” MEAN “WORTH DISCLOSING”? [The Meaningfully Actionable Threshold and its limits]

Clinical genetics has developed a relatively clear approach to IFs. Disclosure is generally justified when a finding indicates a serious condition for which prevention, treatment or clinical management is available.

Forensic genetics is different.

There may be no treating clinician, no genetic counsellor and, in some cases, no living subject. The question therefore becomes: who decides whether an IF should be disclosed?

More importantly, not all significant IFs are clinically actionable. Some may have little or no medical significance, yet profoundly affect an individual’s privacy, identity, dignity or expose them to discrimination.

Consider XXY Klinefelter syndrome. It is a marker of infertility, but its discovery can also reveal, incidentally, that a father is not the biological parent of children in his own family information with no clinical urgency at all, and enormous personal consequence (Lynch, Heathfield & Budowle, 2025).

This information exposes an important limitation of the traditional clinical model. A framework based solely on serious, treatable disease does not adequately address serious non-clinical harms, informational, reputational and dignitary.

That premise is why a second framework for serious non-clinical harms is needed.

At the same time, the opposite situation also arises. A molecular autopsy may identify a heritable cardiac variant that allows surviving relatives to reduce or prevent future harm to themselves. In those circumstances, withholding the information may itself be unethical. The same tension shows up with an inherited disease that, if caught early in a child, could be prevented. Is failing to disclose that, in and of itself, unethical? Does the overriding principle of do no harm override non-disclosure?

The challenge, therefore, is not simply whether an IF should be disclosed, but what kinds of harm, and benefit, should be considered when making that decision.

WHO DECIDES WHAT SHOULD BE DISCLOSED? [Intentional Access to Unintentional Information]

Once an IF exists, it does not move directly from the laboratory to the affected person or family. It passes through an intermediary chain: expert, police or prosecution, defence, judge, and only then potentially the individual or family.

A molecular autopsy illustrates the tension. A result intentionally generated for the court may become an IF for surviving relatives once it enters the public domain.

Existing frameworks, such as UNESCO, the Oviedo Convention and GDPR, recognise privacy and informational rights. But these frameworks were developed to govern traditional genetics, medical ethics and data protection, and not the modern sequence of events enabled by technologies like FGG and AI.

If we look at who has access to this information, i.e. the intermediary chain, it creates three distinct decisions rather than one general question about disclosure.

1. Investigative relevance

Is the finding material to identification, attribution, or the alleged offence?

2. Judicial confidentiality

Can it be shared consistently with legal duties, court controls, and procedural fairness?

3. Individual and family rights

Could disclosure create irreversible harm to privacy, dignity, identity, family relationships or diversity equity?

These duties can point in different directions. A finding may be scientifically valid but still lack evidential value, or be legally restricted, or create disproportionate non-clinical harm. A defensible framework should separate these decisions, define who has authority at each stage, and document the reasoning behind them.

THE GOVERNANCE GAP

The governance gap is longstanding.

In 2005, the mitochondrial DNA disease-marker debate ended with a recommendation that ISFG and SWGDAM develop formal guidance (Budowle, Gyllensten, Chakraborty & Allen, 2005). In 2016, a forensic secondary-findings framework was again requested (Sajantila & Budowle, 2016). In 2020, pathogenic-variant filtering renewed the reporting debate.

Yet there is still no binding international standard, and national approaches remain fragmented. Switzerland, for example, takes a restrictive approach to disclosure of unintended findings in paternity and forensic contexts (Zieger, 2025).

This issue is therefore not a new dilemma created by the latest technology.

The technology did not create the governance gap; it has simply made it impossible to ignore.

This gap creates a clear opportunity to lead the conversation before a high-profile failure produces reactive regulation from outside the field. A practical framework needs three components.

First, define categories of IFs: clinical, relational, identity-related, and other serious non-clinical harms.

Second, decide thresholds for relevance, validation, escalation, disclosure, and non-disclosure.

Third, govern the process through consent rules, treatment of relatives, right-to-know and not-to-know principles, documentation, and oversight.

FROM AD-HOC DECISIONS TO GUIDANCE

These issues are no longer theoretical.

MPS and FGG are already routine, the guidance is not addressed.

It is time to move from ad-hoc, case-by-case decisions to principled, field-wide guidance.

I thought a good way to present some of these issues would be to share my speaker notes, alongside a few of the slides I prepared for my presentation.

It does not resolve the issues; it raises them up for discussion.

The first step is awareness.

The next step is toward policy that regulates actions regarding IFs.

I invite you to join the conversation.

ACKNOWLEDGEMENTS

I am grateful to Prof. Bruce Budowle for his generous guidance and input in the lead-up to this presentation. Much of the thinking here was sharpened through conversation with him.

REFERENCES

1. Caenazzo L, Tozzo P, Dierickx K. New Frontiers and Old Challenges: How to Manage Incidental Findings When Forensic Diagnosis Goes Beyond. Diagnostics. 2020;10(9):731.

2. Gabriele A, Chierto E, Gino S, Inturri S, Aneli S, Robino C. Privacy and ethical challenges of the Amelogenin sex test in forensic paternity/kinship analysis: Insights from a 13-year case history. Forensic Science International: Synergy. 2023; 7:100440.

3. Kayser M, Branicki W, Parson W, Phillips C. Recent advances in Forensic DNA Phenotyping of appearance, ancestry and age. Forensic Science International: Genetics. 2023; 65:102870.

4. Lynch V, Heathfield LJ, Budowle B. Disclosure of biological sex may impact individual privacy. Forensic Science International: Genetics. 2025; 76:103213.

5. Miller DT, Lee K, Abul-Husn NS, et al. ACMG SF v3.2 list for reporting of secondary findings in clinical exome and genome sequencing: A policy statement of the American College of Medical Genetics and Genomics (ACMG). Genetics in Medicine. 2023;25(8):100866.

6. Sajantila A, Budowle B. Postmortem medicolegal genetic diagnostics also require reporting guidance. European Journal of Human Genetics. 2016;24(4):481-482.

7. Budowle B, Baker L, Sajantila A, Mittelman K, Mittelman D. Prioritizing privacy and presentation of supportable hypothesis testing in forensic genetic genealogy investigations. Journal of Forensic Sciences. 2024;76(9):425-431.

8. Zieger M. Dealing with incidental findings in Forensic Genetics. Proceedings of the 30th Congress of the International Society for Forensic Genetics. Universidade de Santiago de Compostela; 2025:69-76.

9. Budowle B, Gyllensten U, Chakraborty R, Allen M. Forensic analysis of the mitochondrial coding region and association to disease. International Journal of Legal Medicine. 2005; 119:314-315.