Emerging Issues in Genomics and the Courts

While DNA has been a forensic tool in the courts since the 1980s, recent advances in the ability to generate, store, and analyze genetic data have broadened the application of DNA technologies in the federal judiciary. Emerging issues for the courts include the broader use of genetic information for identification, concerns over privacy and discrimination stemming from disclosure of genetic data, and the use of genetic data to make predictive claims about traits, health, and behavior. Overall, these materials aim to familiarize judges with the different genetic technologies discussed and the legal questions that may arise with their use in court.

Here are some key takeaways.

New DNA genotyping and sequencing technologies provide more information than traditional forensic DNA methods. The courts are familiar with DNA genotyping using short tandem repeats (STRs) through decades of forensic DNA evidence. New genome-wide genotyping and sequencing technologies, however, provide dramatically more information about an individual than traditional forensic DNA methods. This has important implications for an individual’s privacy and what information can be learned about their health, behavior, and appearance. When presented with DNA evidence, it is important to clarify what type of genetic information is being discussed: whole-genome or whole-exome sequencing, targeted gene sequencing, SNP-array genotyping, or STR genotyping. Equally important, different genotyping and sequencing technologies carry different error rates, and these can be quantified through validation experiments.

Genetic data are passed across different domains with varying levels of protection. Genetic data can be collected for different domains of use: research, healthcare, commercial, and forensic. The data collected for a domain are intended for a specific use and are protected accordingly, but it is common for genetic data to be passed across these domains. An individual’s genetic data may move from commercial to healthcare, or healthcare to research, or commercial to law enforcement domains. As data are passed between domains with different regulations and levels of protection, this creates the risk for lapses in protection, exposure of private information, and discrimination. Certificates of Confidentiality, granted to many human research studies, are one mechanism of protection that strictly limit the disclosure of information about research study participants. But even in cases where this protection is waived, judges can still impose limitations on the extent of the disclosure.

Genetic algorithms need to be validated for their intended use. In assessing the reliability and utility of an algorithm for interpreting or clarifying genetic data, it is important to closely examine whether the algorithm is being used in the same context in which it was validated. For example, polygenic risk scores (PRS) are built from genetic data sets that are predominantly from individuals with European ancestry. This may bias PRS in nonwhite populations. Similarly, many probabilistic genotyping systems have been validated across narrowly constrained parameters—for example, with a limited number of contributors or a few specific laboratory conditions. However, these systems are often engaged in less constrained circumstances.

The field of genomics is developing rapidly, and our understanding of the links between genetic variants and traits changes with new information. Genetic data from millions of individuals are collected with the help of new DNA sequencing methods that allow us to learn more about the effects of genetic variation. Assigning an effect to a genetic variant has become a more centralized process involving large databases of collected and curated clinical reports (e.g., ClinGen and ClinVar) and guidelines for their interpretation from national organizations like the American College of Medical Genetics and Genomics (ACMG) and the Association of Molecular Pathologists (AMP). As genomics research continues, variants and their clinical significance continue to be reclassified. In the absence of regulation, it remains unclear how often, if at all, clinical testing laboratories need to reevaluate their variant classifications and recontact patients if reclassifications change test results. It is important to recognize that the field of genomics is developing rapidly, and that the understanding of a genetic variant’s effects may change with new information.

Genetics is not deterministic, and traits are also affected by environmental and behavioral factors. Most traits are complex, mediated by multiple environmental, behavioral, and genetic factors. Even very strongly determined genetic traits, like height, are controlled by multiple genetic variants and environmental conditions. Each genetic variant offers a little nudge to affect a trait, and it is the accumulation of these little nudges, along with other pressures such as an individual’s environment, life history, and behavioral choices that shape the overall effect. Predictions about how a pattern of genetic variants may affect a trait are probabilistic. It is inaccurate to suggest that genetic information alone is sufficient to make any predictions about an individual or group.