BMC Med Ethics. 2026 May 28.
BACKGROUND: Reanalysis - the process of re-examining a person's existing genomic data based on new knowledge or technology - has the potential to greatly increase diagnostic yield for people living with a genetic, rare or undiagnosed condition. Routine reanalysis has not been widely implemented, largely because of the labour-intensive nature of doing this manually. Yet, the more genomes that undergo sequencing, the greater the likelihood that we are missing opportunities to reanalyse previously undiagnosable cases. Automating reanalysis of genomic data could increase efficiency and accuracy, subsequently providing an opportunity to increase the availability (and frequency) of reanalysis to all people living with an undiagnosed condition. Despite these significant potential benefits, concerns include a lack of transparency and accountability, bias, risk to data security, and the subsequent impact on trust in automated systems. However, the public's perspectives on these ethical issues are yet to be examined.
METHODS: Twenty-three Australians living with - or caring for a person with - a genetic, rare or undiagnosed condition participated in one of seven 1.5-hour focus groups conducted online. Focus groups explored participants' preferences and values regarding key characteristics of automated reanalysis of genomic data. Focus groups were recorded, transcribed and analysed using inductive content analysis.
RESULTS: Discussions demonstrated strong support among people living with, or impacted by, a genetic, rare or undiagnosed condition for the automation of genomic reanalysis, emphasising its potential to reduce both practical and psychological burdens associated with manual processes. Participants underscored the need for transparent communication and continued human oversight to build trust in automated systems. However, widespread misunderstandings about reanalysis, concerns around data privacy, and patient preferences for return of results remain significant barriers that must be addressed to ensure ethical and effective implementation.
CONCLUSIONS: Our findings suggest that people living with, and caring for someone with, a genetic, rare or undiagnosed condition perceive significant potential benefits of automated genomic reanalysis. Participants anticipated that automation could reduce some of the practical demands associated with initiating reanalysis and provide reassurance that their genomic data would continue to be reviewed as scientific knowledge evolves. However, successful implementation will require careful attention to governance frameworks, informed consent processes, transparent data practices, and sustained human oversight to maintain trust and ensure equitable access.
Keywords: Artificial intelligence; Australia; Focus groups; Genomics; Patient perspective; Qualitative