For centuries, Polynesian navigators crossed thousands of miles of open ocean, using their knowledge of stars, currents, birds, clouds and the sea to settle islands across the Pacific. Now, a new University of California San Diego study shows the history of voyaging is also reflected in the genomes of Polynesian populations today.
Published Oct. 8 in Science and featured on the journal's cover, the study analyzes whole-genome sequences from 1,050 individuals across Polynesia. The findings offer genomic evidence of the extraordinary skill, knowledge and achievement of Polynesian voyagers, whose mastery of long-distance ocean travel helped shape the genetic history of the Pacific.
Using DNA to trace how Polynesian communities were formed, the study also shows that the same migration history may help explain why certain health-related genetic variants are more common in some island populations.
As Polynesian voyagers left established island communities to settle new ones, each group carried only a portion of the genetic diversity found in the population it left behind. For example, a group departing Mangareva for Hawaiʻi or Rapa Nui would have carried only a smaller genetic selection of the people living in Mangareva. When later groups set out from newly settled islands, they carried an even smaller subset of that genetic variation with them.
The authors compare the pattern to a set of Russian nesting dolls: The largest population contains the most genetic variation, while each successive population contains a smaller genetic subset within it.
"As we go from one island to the next, we're selecting a subset of a gene pool, and that is shaping our genome over time," said Keolu Fox, an associate professor of anthropology at the UC San Diego School of Social Sciences and a corresponding author of the study.
Reconstructing a history of voyaging
The study examined genomic data from populations in French Polynesia, including the Austral, Society, Marquesas and Tuamotu islands, as well as Mangareva. The researchers also incorporated data from Rapa Nui, Native Hawaiians, Samoans and Tongans.
The analysis found that Hawaiʻi and Rapa Nu i — two of the most geographically distant Polynesian populations — are also the most closely genetically related among the remote island populations studied.
The islands are separated by approximately 7,000 kilometers and each is more than 3,000 kilometers from the nearest inhabited Polynesian island. The researchers say their genetic similarities suggest that both populations were ultimately settled by voyagers whose deeper ancestral roots trace back to Mangareva.
The findings support the existence of an ancient culture capable of repeated, long-distance voyaging across the Pacific. The researchers refer to this hypothesized tradition as Holomoana Nui, meaning "great ocean voyaging."
The pattern was not created by a single, one-way migration. Polynesian voyaging involved repeated movement among island populations, with voyagers carrying provisions, knowledge and genetic material across the ocean. Over time, the repeated settlement of new islands produced increasingly concentrated founder effects.
In Polynesia, the researchers found evidence of multiple founder effects — when small groups establish new populations, carrying only a portion of the genetic diversity of the larger population they left behind. They also discovered bottlenecks — when disease or another major event sharply reduces a population, leaving survivors with less genetic variation than before — accumulating over the course of settlement. The result is one of the strongest cumulative founder effects observed in human population genetics.
What the findings could mean for health
Founder effects can influence health because rare genetic variants carried by a small founding population may become more common in later generations.
The researchers identified three variants associated with autosomal recessive diseases at relatively high frequencies in French Polynesia. One variant in the FAN1 gene, associated with karyomegalic interstitial nephritis — a rare condition that can lead to kidney failure — was found in more than 11% of the French Polynesian cohort. The variant was absent from gnomAD, a major international database containing genomic data from more than 800,000 individuals.
The findings demonstrate how medically important variants can be common within a specific founder population while remaining invisible in large global genomic databases.
That gap matters because much of modern genomic medicine has been developed using data from populations of European ancestry. Without more detailed information about Polynesian populations, clinicians may not know which variants to screen for or how frequently certain disease-associated variants occur.
The study also underscores why broad population categories can be inadequate for clinical genetics. Grouping Native Hawaiian and Pacific Islander populations together may obscure important differences among individual island populations.
Fox said the findings could eventually help researchers develop more precise approaches to screening, diagnosis and treatment, including for diseases such as cancer.
The study itself does not establish new clinical guidelines or treatments. Rather, it identifies genetic information that could guide future research and clinical work.
A study designed around trust
Fox, who is the first Native Hawaiian to earn a Ph.D. in Genome Sciences, has a longstanding interest in Polynesian history and the health needs of Pacific Islander communities.
During a 2017 visit to French Polynesia, a conversation about the legacy of French nuclear testing led Fox to consider how genomic research might help clarify disease risks in communities exposed to environmental hazards and underserved by modern medical infrastructure.
The French government conducted 193 nuclear tests in French Polynesia between 1966 and 1996. Fox said the experience prompted him to think more deeply about how precision medicine could serve communities that have not always had access to comprehensive cancer screening, genetic testing or specialized care.
But the study was also shaped by a second question: How could the research be conducted in a way that protected Indigenous communities and allowed them to share in the benefits of scientific discovery?
The answer became Variant Bio, a biotechnology company that Fox joined as a senior advisor while conducting the research. It develops partnerships with populations that have historically been underrepresented in genomic research. He said the company was created to give participating communities greater control over how their data are collected, stored and used.
"We want to build trust first," Fox said. "Making you a partner and not a subject in this allows us to expedite the development of these projects."
That approach treats community members as partners in research rather than simply as sources of biological material. It includes explaining the potential uses and risks of genomic data, building relationships with community leaders and involving local health professionals in the research process.
The study includes contributions from Indigenous and Pacific Islander communities in Hawaiʻi, Rapa Nui and French Polynesia. Its authors also acknowledge the French Polynesian clinicians, nurses and community health workers who helped make the research possible.
Fox said the inclusion of local clinicians reflects a central principle of the project: The people who understand the health needs of a community should be involved in research conducted there.
"None of this is possible without the contributions of these partners," Fox said.
Sharing in the benefits
Variant Bio's model includes a formal benefit-sharing commitment of 4% of the company's revenue and 4% of equity-derived value that are designated for partner communities.
The arrangement is intended to ensure that communities contributing genetic data share in the value created by research and future commercial partnerships. The funds can support priorities identified by the communities themselves, including health care, education, infrastructure, food sovereignty, energy projects and data infrastructure.
For Fox, benefit-sharing is not separate from the science. It is part of the research design.
"If you want to recruit communities into research, you have to make them stakeholders in the technology you are building," he said.
Fox, who spent nearly a decade on the research endeavor, says the paper in Science reflects a broader vision for the future of genomic medicine — one in which Indigenous communities help shape the research questions, govern the use of their data and participate in the benefits of discoveries made from their genomes.
"It is remarkable to say that our achievements as voyaging people have shaped our genomes over time," Fox said. "And to have that recognized on the cover of Science is incredibly validating."
Read the full study, "Message in a Bottleneck: Nested Founder Effects from French Polynesia to Rapa Nui and Hawaiʻi" in Science.
