Written by Virginia Barron | Criminal Defense Attorney & DNA Forensic Litigator
What recent forensic oversight reports say about using DNA evidence to determine how, when, or through what activity DNA was deposited
A DNA result can be scientifically accurate and still be used to support a conclusion the science does not actually answer or support.
A laboratory may generate a strong statistical result supporting the proposition that a person contributed DNA to a sample. The statistic may be valid. The analyst may have correctly interpreted the DNA profile. But those conclusions do not automatically tell us how the DNA arrived, when it arrived, whether the person touched the item, or whether the DNA was deposited during the alleged crime. (Texas FSC Final Report, p. 8; Massachusetts FSOB Updated Recommendations, p. 13.)
I have written before about touch DNA and secondary transfer, and about why not every DNA result should be accepted at face value. What is especially important now is that forensic oversight bodies are confronting the same boundary in increasingly direct terms: the difference between identifying whose DNA may be present and evaluating what activity produced it.
Two documents make that development particularly worth watching. The first is the Texas Forensic Science Commission’s July 2024 final report arising from testimony in the Kaitlin Armstrong murder trial. The second is an April 30, 2026 updated memorandum from a subcommittee of the Massachusetts Forensic Science Oversight Board addressing a separate Tiffany Roy complaint involving fingernail DNA and testimony that the amount of DNA was “inconsistent with casual contact.” The cases are different, but the scientific problem at their center is strikingly similar. (Texas FSC Final Report, Executive Summary, pp. i-vi; Massachusetts FSOB Updated Recommendations, pp. 1-3.)
The full Texas Forensic Science Commission Final Report on Complaint No. 23.67 is available here. It is unusually useful for lawyers because it does not treat “DNA evidence” as a single question. It separates the DNA comparison from the activity-level conclusion and asks what scientific safeguards should exist before an expert moves from one to the other.
The First Question: Whose DNA May Be Present?
Most forensic DNA comparison work is directed at what the Texas Commission describes as the “who” question. At the sub-source level, the analyst evaluates the DNA results under competing propositions about who may have contributed to the profile. Modern probabilistic genotyping can produce a likelihood ratio expressing how much more probable the observed DNA results are under one contributor proposition than another. (Texas FSC Final Report, pp. 8-10, 17-21.)
That is important evidence. But it is evidence about the DNA comparison. It is not a statistical answer to what happened in the underlying event.
A source-level result and an activity-level conclusion are not the same thing.
A large likelihood ratio may provide strong support for a proposition about who contributed DNA. It does not automatically provide strong support for a proposition about how the DNA was deposited or what the person did.
The Armstrong Case Shows Why the Distinction Matters
In State of Texas v. Kaitlin Armstrong, DNA was recovered from the seat and handlebars of the murder victim’s bicycle. The Texas DPS analyst interpreted both samples as three-person mixtures. For the bicycle seat, she reported that the DNA results were 224 billion times more probable if Wilson, Armstrong, and an unrelated unknown person were contributors than if Wilson and two unrelated unknown people were contributors. For the handlebars, the reported likelihood ratio was 49,400 under the corresponding propositions. The Commission noted that there was no dispute at trial about the DNA mixture interpretation reported by Texas DPS. (Texas FSC Final Report, Executive Summary, p. i; report p. 8.)
The difficult question was what those results meant about the bicycle. The prosecution’s theory was that Armstrong had handled and removed it around the time of the murder. The defense elicited testimony about possible indirect transfer pathways, including a shared motorcycle and helmet. In rebuttal, Dr. Timothy Kalafut offered an activity-level evaluation and testified that the DNA evidence was much more likely if Armstrong had directly interacted with the bicycle than if the DNA had arrived through a series of indirect activities. (Texas FSC Final Report, Executive Summary, pp. i-ii; report pp. 12-16.)
That shift is the heart of the issue. The question was no longer simply whether the DNA findings supported Armstrong as a contributor. The question had become whether the findings supported one alleged activity over another.
A Big Likelihood Ratio Does Not Become an Activity-Level Number
This is one of the most important points for lawyers and jurors to understand. A source-level likelihood ratio does not simply carry upward into the activity question. The Texas Commission expressly warned about the risk that a factfinder may misunderstand a DNA comparison statistic as though it also answers how or when the DNA was deposited. The Commission also noted that activity-level likelihood ratios are typically far lower than sub-source likelihood ratios. Activity-level evaluations must also account for additional uncertainty involving transfer, persistence, prevalence, recovery, background DNA, case assumptions, and competing activities. (Texas FSC Final Report, pp. 22-23, 52-54.)
In other words, the impressive number the jury hears may be answering a much narrower question than the one that actually decides the case.
What the Texas Forensic Science Commission Found
The Texas Commission did not reduce the issue to a simple rule that DNA experts may never discuss transfer. In fact, the report repeatedly acknowledges that activity-level evaluation is an established area of forensic practice in parts of Europe, the United Kingdom, and Australia. The Commission’s concern was whether the scientific, procedural, training, reporting, and quality-assurance framework necessary for that work exists in the United States. (Texas FSC Final Report, Executive Summary, pp. ii-vi; report pp. 28-30.)
The report identified ten unresolved issues that it said require attention before activity-level evaluations could be successfully integrated into forensic laboratory practice in the United States. Those issues include the formulation of balanced propositions, the need for written reporting, suitability assessment, transparency and traceability, the sufficiency of TPPR literature, the use of numerical likelihood ratios, expert education and training, and quality assurance. (Texas FSC Final Report, Executive Summary, pp. ii-iv.)
The Written Report Matters
One of the Commission’s clearest observations concerned documentation. Dr. Kalafut had provided prosecutors with an “evidence reconciliation” document and papers of interest, but the materials were treated as notes rather than a formal expert report and were not provided to the defense before the testimony. The Commission concluded that the case would have been better served by a written report and emphasized that written reporting is a cornerstone of international guidance on activity-level evaluation. (Texas FSC Final Report, pp. 38-39, 59-60.)
That concern is not merely procedural. A written report forces the expert to identify the propositions being evaluated, the assumptions being made, the data relied upon, the model used, and the reasoning connecting the data to the final opinion. It also gives the opposing party a meaningful opportunity to review and challenge that reasoning before the jury hears it. (Texas FSC Final Report, pp. 38-46, 59-60.)
The Data Must Fit the Question
The Commission also identified a genuine scientific problem with the fit between published research and a specific case. It observed that none of the studies cited by Dr. Kalafut involved the actual activity at issue, the movement of a bicycle, and that the written analysis did not explain why those studies could reliably be extrapolated to the facts being evaluated. More broadly, the Commission described a near-complete absence in the United States of guidance for determining how to assess published TPPR data and reliably assign probabilities in casework. (Texas FSC Final Report, pp. 47-49, 59-60.)
This is crucial. A study showing that DNA can transfer under one set of experimental conditions does not automatically tell us the probability of obtaining the evidence under a different set of real-world circumstances. The closer an expert gets to offering a case-specific probability, the more important it becomes to explain exactly where the probability came from and how well the underlying research fits the case.
Transparency and Traceability Are Not Optional Details
The Texas report repeatedly returns to transparency. If an expert relies on published studies, case-specific experiments, professional experience, or some combination of those sources, the basis for the probability assignment must be identifiable and capable of review. The Commission stated that when an expert cannot provide a detailed justification for the evaluation, that may be a sign the evaluation is not appropriate for the case. (Texas FSC Final Report, pp. 49-51.)
The Commission’s recommendations therefore call for written, balanced, logical, transparent, and robust evaluations; clear disclosure of assumptions and statistical models; traceability to the sources used for probability assignments; technical review; proficiency testing; bias-mitigation measures; and other quality-system protections associated with accredited laboratories. (Texas FSC Final Report, Executive Summary, pp. v-vi.)
Training for DNA Comparison Is Not the Same as Training for Activity Evaluation
Another central point is expert qualification. The report makes clear that expertise in DNA profile interpretation does not automatically make someone an expert in activity-level evaluation. The work requires additional knowledge of TPPR, Bayesian reasoning, model construction, proposition formulation, probability assignment, sensitivity to case assumptions, and the limits of the available data. (Texas FSC Final Report, pp. 9-10, 54-58.)
The Commission was particularly direct about short conference workshops. It described a day or two of exposure as insufficient preparation for performing activity-level evaluations in forensic casework. That matters in court because a witness can be highly qualified in one area of DNA science and still lack the specialized training needed for a different scientific question. (Texas FSC Final Report, pp. 55-56.)
The Massachusetts Memo Reaches the Same Boundary From a Different Case
The April 2026 Massachusetts memorandum arose from a different murder case and a different kind of DNA testimony. The complaint involved DNA recovered from beneath a victim’s fingernail and testimony that the amount of DNA detected was inconsistent with casual contact. The complaint also challenged efforts to connect visually observed tissue to the DNA profile itself. (Massachusetts FSOB Updated Recommendations, pp. 2-4.)
The memo is significant because the subcommittee did not consider the issue in isolation. It expressly situated the complaint alongside the 2024 NIST human-factors report, the Texas Forensic Science Commission’s 2024 report, and SWGDAM’s 2025 position statement on formal activity-level reporting. It also incorporated input from laboratory representatives and experts working on developing future guidance. (Massachusetts FSOB Updated Recommendations, pp. 1, 8-12.)
The subcommittee declined to recommend case-specific findings about the older testimony because much of the testimony predated the more recent published guidance. Instead, it focused prospectively on what DNA reports and testimony should communicate going forward. (Massachusetts FSOB Updated Recommendations, pp. 1-2.)
Five Cautions the Massachusetts Subcommittee Wants DNA Reports to Communicate
The most concrete part of the Massachusetts memo is its proposed reporting guidance. The subcommittee recommended that DNA comparison reports communicate five core limitations: (Massachusetts FSOB Updated Recommendations, pp. 10, 13.)
DNA interpretations occur in the context of case information, and a change in that information may require a new evaluation.
DNA comparisons can help address whose DNA may or may not be detected.
The DNA interpretation should not conclude that a person is the source of the DNA simply because of the magnitude of the statistical result.
DNA testing by itself does not provide information about how or when the DNA was or was not deposited.
DNA testing by itself does not identify the biological source of a DNA profile, such as blood, semen, or saliva.
Those cautions are not anti-DNA. They are an attempt to define what a DNA comparison actually establishes and, just as importantly, what it does not establish.
NIST and SWGDAM Show Why This Area Is Still Developing
The Massachusetts memo also relies heavily on the 2024 NIST report on human factors in forensic DNA interpretation. Chapter 7 separates “who” questions from “how and when” questions and warns against allowing an ordinary DNA comparison to be interpreted as though it answers both. A majority of the NIST working group recommended that DNA analysts not opine about the possibility or probability of direct or indirect transfer in a case and emphasized that “how” and “when” evaluations require different policies, methods, training, and expertise. (NIST IR 8503, Chapter 7; Massachusetts FSOB Updated Recommendations, p. 8.)
But the scientific community is not unanimous about the best path forward. The Massachusetts memo notes dissent from the NIST recommendation by Lynn Garcia and Dawn Boswell, who agreed that “how” and “when” testimony can be misleading but questioned a categorical prohibition before the reliability of TPPR opinion testimony, expert qualifications, and quality-system safeguards are more fully developed. The Texas Commission similarly supported the NIST direction at a philosophical level while emphasizing the need for significant analyst and legal-system training before imposing a blanket expectation on courtroom answers. (Massachusetts FSOB Updated Recommendations, p. 8; Texas FSC Final Report, pp. 28-30.)
SWGDAM’s 2025 Position Statement on Formal Activity Level Reporting adds another important layer. SWGDAM concluded that formal activity-level reporting, in its present form, is not suitable for use in the U.S. criminal justice system and identified scientific, practical, and legal barriers to implementation. At the same time, SWGDAM said it would continue monitoring the field and exploring alternatives because analysts will continue to face TPPR questions in court. (SWGDAM Position Statement on Formal Activity Level Reporting, pp. 2-5; Massachusetts FSOB Updated Recommendations, p. 9.)
The important takeaway is not that transfer science has no place in court.
Answering “how did the DNA get there?” is a different scientific task from answering “whose DNA may be present?” A reliable activity-level opinion requires a separate foundation.
What Defense Attorneys Should Ask When DNA Testimony Moves From “Who” to “How”
For a defense attorney, these reports suggest a very practical way to evaluate testimony. The moment an analyst or expert moves from describing the DNA comparison to explaining what activity produced the DNA, the questions should change.
What level of proposition is the expert actually addressing: sub-source, source, or activity?
Is the expert merely describing general transfer principles, or offering a case-specific opinion about what happened?
If the opinion is case-specific, where is the written evaluation?
What competing activity propositions were evaluated, and are they balanced and mutually meaningful?
What case information and assumptions were used? What happens if one of those assumptions changes?
What TPPR studies support the probability assignments, and how closely do those studies resemble the circumstances of the case?
Can the expert trace each important probability or judgment back to published data, case-specific experimentation, calibrated experience, or another disclosed source?
Was the method validated? Was the work technically reviewed? Is there proficiency testing or another quality-assurance process?
What education and training qualifies this witness specifically to perform activity-level evaluation, as opposed to ordinary DNA comparison?
Is a large source-level likelihood ratio being used, expressly or implicitly, as proof that the defendant performed the alleged act?
These questions go directly to whether the expert is offering a scientifically supported interpretation or whether a valid DNA result is being stretched beyond the question it was designed to answer. (Texas FSC Final Report, pp. 38-60; Massachusetts FSOB Updated Recommendations, pp. 11-14.)
Why These Reports Matter in Minnesota
Neither the Texas report nor the Massachusetts subcommittee memorandum is binding Minnesota law. They do not decide what a Minnesota judge must admit in a particular case. Their importance is scientific and practical: they identify the kinds of foundational questions that arise when expert testimony crosses from DNA comparison into activity-level interpretation.
Minnesota Rule of Evidence 702 requires expert opinion to have foundational reliability and, when novel scientific theory is involved, requires general acceptance in the relevant scientific community. Minn. R. Evid. 702 therefore makes the quality of the foundation, the expert’s qualifications, and the scientific reliability of the methodology central questions when specialized testimony is offered. (Minn. R. Evid. 702.)
Minnesota Rule of Evidence 403 separately permits exclusion of relevant evidence when its probative value is substantially outweighed by dangers that include confusion of the issues or misleading the jury. Minn. R. Evid. 403 That concern has obvious relevance when a powerful source-level statistic risks being heard as proof of an activity the statistic does not actually evaluate. (Minn. R. Evid. 403; see also Texas FSC Final Report, pp. 22-23.)
Whether those rules support exclusion, limitation of testimony, a foundational reliability hearing, cross-examination, or simply a clearer explanation will depend on the facts of the individual case. The reports do not provide a one-size-fits-all litigation answer. They do, however, provide a framework for identifying when the courtroom meaning of a DNA result has moved beyond the laboratory question that produced it.
The Science Can Be Accurate and the Inference Can Still Go Too Far
There is a tendency in criminal cases to divide forensic evidence into two categories: reliable or unreliable, good science or bad science. DNA evidence often requires a more careful analysis.
The laboratory may have correctly detected a person’s DNA. The mixture interpretation may be sound. STRmix may have produced a valid likelihood ratio. None of that means the evidence answers every question the prosecutor may want it to answer.
A source-level DNA result can be accurate while the inference drawn from it about conduct, timing, possession, contact, or participation goes beyond what the DNA result itself reliably establishes. That is precisely why the distinction between “who” and “how” matters so much.
The Texas and Massachusetts documents are valuable because they show forensic oversight bodies grappling openly with that boundary. They also show that the solution is not simply to replace one oversimplified answer with another. Activity-level interpretation requires careful propositions, relevant data, transparent reasoning, appropriate training, written disclosure, and quality assurance. Where those safeguards do not exist, the limitations need to be made clear. (Texas FSC Final Report, Executive Summary, pp. ii-vi; Massachusetts FSOB Updated Recommendations, pp. 8-14.)
For defense attorneys, that means a large DNA statistic should not end the inquiry. Sometimes it is where the more important inquiry begins.
For Defense Attorneys Handling DNA Evidence
If DNA is central to a prosecution theory, I review more than the final laboratory report. I look at the underlying data, the assumptions used to interpret it, the statistical model, the laboratory procedures, and the way the result is being translated into a courtroom narrative. When testimony begins to move from whose DNA may be present to how, when, or why it got there, that distinction deserves specific attention.
I consult with criminal defense attorneys throughout Minnesota on forensic DNA evidence, including STRmix, complex mixtures, low-template DNA, transfer issues, contamination concerns, and the limits of expert testimony. The goal is not to challenge DNA simply because it exists. The goal is to determine what the science actually supports and whether the legal meaning being assigned to it goes further than the evidence can reliably establish.
If you are handling a Minnesota criminal case in which DNA is being used to prove not only who may have contributed to a sample, but what the person allegedly did, that is a question worth examining carefully.
Call or text Ginny Barron at 507-822-5735 to discuss whether a forensic DNA review may be useful in your case.
Frequently Asked Questions About DNA Evidence and Activity-Level Interpretation
-
Not by itself. A DNA comparison can provide information about whose DNA may be present, but it does not automatically establish how the DNA was deposited, whether the person directly touched the object, or what activity occurred. Determining how DNA arrived requires a separate evaluation of transfer, persistence, prevalence, recovery, case circumstances, and competing explanations.
-
No. A large likelihood ratio can provide strong statistical support for a proposition about who contributed DNA to a sample, but it does not automatically provide the same level of support for a proposition about how the DNA got there. A source-level likelihood ratio should not be treated as proof that the person touched the item or participated in the alleged crime.
-
Activity-level DNA interpretation evaluates DNA findings in the context of competing explanations about what activity may have produced them. Instead of asking only, “Whose DNA may be present?” it asks questions such as whether the findings are more expected if someone directly handled an object or if the DNA arrived through another transfer pathway. That is a different scientific question from ordinary DNA profile comparison.
-
STRmix is a probabilistic genotyping program used to help evaluate DNA mixture evidence and calculate likelihood ratios concerning possible contributors. A valid STRmix result can provide important information about the strength of a DNA comparison, but it does not by itself determine how the DNA was deposited, when it was deposited, or what the person was doing when the DNA was left.
-
TPPR refers to transfer, persistence, prevalence, and recovery. These factors become important when the question changes from whose DNA may be present to how biological material may have moved, how long it may have remained, how commonly it may be found in a particular environment, and whether it would be expected to be recovered under the circumstances. Activity-level opinions therefore require a different scientific foundation than a basic DNA comparison.
-
Yes. A laboratory may correctly detect a person’s DNA, correctly interpret a mixture, and correctly calculate a likelihood ratio while the inference later drawn from that result goes beyond what the DNA analysis itself establishes. Problems can arise when evidence answering “whose DNA may be present?” is presented as though it also answers questions about conduct, timing, possession, contact, or participation.
-
The attorney should determine exactly what scientific question the expert is answering, what competing propositions were evaluated, what assumptions were used, what research supports the opinion, and whether the analysis was documented and technically reviewed. The expert’s training in activity-level interpretation also matters because expertise in DNA profile comparison does not automatically establish expertise in evaluating how DNA was transferred.
-
No. The Texas Forensic Science Commission report and the Massachusetts Forensic Science Oversight Board recommendations do not control what a Minnesota court must admit. They are important because they reflect current scientific concerns about the boundary between DNA comparison and activity-level interpretation and can help identify foundational issues that may matter under Minnesota’s evidentiary rules.
-
An independent review may be particularly important when DNA is being used to prove more than simply whose DNA may be present. If the prosecution is relying on DNA to establish touching, possession, timing, transfer, or a particular activity, reviewing the underlying data, assumptions, laboratory procedures, statistics, and scientific basis for that interpretation can reveal issues that are not apparent from the final laboratory report alone. Virginia “Ginny” Barron of Barron Law Office consults with Minnesota criminal defense attorneys on STRmix, complex DNA mixtures, low-template DNA, transfer, contamination concerns, and the limits of forensic DNA testimony.
Sources and Further Reading
Massachusetts Forensic Science Oversight Board Subcommittee, Updated Recommendations in Response to Tiffany Roy Complaint (April 30, 2026). Source document reviewed for this article.