The potentially dubious science of commercial DNA ancestry tests has serious implications for identifying Indigenous Australian heritage. By Andrea Booth.
Indigeneity and DNA ancestry tests
In the past decade, DNA ancestry testing has grown into a $500 million global industry. With just a cheek swab, testing companies offer to trace your ancestry back generations and across continents. But as the direct-to-consumer testing has boomed, scientists have become increasingly concerned about its methodologies and accuracy. “The field in general has a poor reputation in my opinion,” said Dr David Balding, a professor of statistical genomics at the University of Melbourne. “Commercial concerns outweigh any attention to scientific rigour by a big margin.”
Balding said the results were “largely uncheckable by the customer or even an expert, and there seems no penalty for low integrity to the extent of even just making up results, which is rumoured to occur”.
Dr Peter Gunn, associate professor in forensic biology at the University of Technology Sydney’s Centre for Forensic Science, has raised issues about aspects of the methodology used by one United States-based company, DNA Tribes, which offers a test many others don’t – to determine whether someone has Indigenous Australian ancestry. “The whole method that they have used to interpret results is based on what I would call a logical fallacy,” he told The Saturday Paper.
Gunn analysed the results of two tests from DNA Tribes, one taken by me and the other by my NITV colleague Rachael Hocking, a proud Warlpiri woman. Neither of us received the results we were expecting. Mine showed high matches with Poland and Veneto in one historic migratory category and Poland, Veneto and European–Aboriginal Australia in another. This is completely at odds with my understanding of my British and East Asian heritage. Meanwhile, Rachael’s report said some of her top matches across both categories were with Portugal and Basque – her Indigenous heritage was nowhere to be seen.
“Learning that my results showed no Aboriginal ancestry, I immediately assumed there had been a mistake,” said Hocking, who will explore ancestry testing of Indigenous Australians more on next week’s episode of The Point on NITV. “Even though I doubted the veracity of DNA Tribes’ testing, doubts which had been backed up by experts, my immediate response was, ‘Oh, they’ve mixed up the samples’, [because] Andrea’s DNA showed some connection to the central desert in Australia – my grandmother’s country.”
Gunn said, regardless of a potential mix-up, the issue remains that “the interpretation that DNA Tribes has used is wrong”. To illustrate, he used the report addressed to me, which estimates the profile of the person tested to be about 1800 times more common in the Polish population in comparison with the rest of the world.
“What that number actually says is, if Andrea Booth is Polish, we are 1800 times more likely to see this profile than if she is of some other ethnic origin … What it definitely does not say is that Andrea Booth is 1800 times more likely to be Polish than something else,” explained Gunn. “It’s sort of like saying, ‘If Spot is a dog, Spot will have fur.’ It’s not the same thing as saying, ‘If Spot has fur, then Spot is a dog.’ ”
Another concern Gunn had is that the tests we ordered examined the part of human DNA called autosomal short tandem repeats (STRs). “They are powerful in identifying particular individuals,” Gunn said, adding that these tests are often used in forensic crime scene investigations. “But they’re not necessarily powerful in identifying your ancestry.” Analysing STRs can still help detect close relatives such as siblings or parents but it can’t be sure of much more than that.
“You get half of your DNA from your mother and half from your father,” Gunn said. “That means you’re getting a quarter from each of your grandparents or an eighth from your great-grandparents or a 16th from your great-great-grandparents. The further back you are looking from your ancestral link, the less and less of that STR DNA that’s available to help you determine if that link really does exist.”
Dr Dennis McNevin, a professor in forensic genetics at the Centre for Forensic Science, says the parts of human DNA known as single nucleotide polymorphisms (SNPs) are typically regarded as more effective in determining people’s origins.
“[SNPs] are generally considered better [ancestry informative markers] because they have lower mutation rates and so introduce less ‘noise’ into ancestry predictions,” he said.
However, it can actually be more difficult to detect Aboriginal and Torres Strait Islander heritage with SNPs, according to Dr Emma Kowal, a professor in anthropology at Deakin University who researches the use of genetics in Indigenous Australians.
“AncestryDNA admits that they don’t have any Aboriginal Australian samples, and they actually advise their customers if you have Aboriginal ancestry, it will most likely come out as South-East Asia or Oceania,” Kowal told The Saturday Paper.
Part of the issue is that there’s relatively little relevant genetic information to compare Aboriginal or Torres Strait Islander people against. As a 2016 study published in Nature found, while 80 per cent of participants in genetic research globally were Europeans, only 0.05 per cent were Indigenous peoples.
This is compounded by the fact there are high levels of genetic diversity between the different cultures of Australia’s First Peoples. As Dr Anna-Sapfo Malaspinas, the co-author of the first genomic study of Indigenous Australians, said in 2016: “Because the continent has been populated for such a long time, we find that groups from south-western Australia are genetically more different from north-eastern Australia, than, for example, Native Americans are from Siberians.”
According to Kowal, genealogy companies would need reference databases for each group in order to conduct reliable DNA analysis.
The intergenerational effects of colonisation and the Stolen Generations are also part of this conversation. As many Indigenous Australians have some European ancestry and because most genetic testing companies have only European reference databases, these tests can pick up only that connection.
“All a genetic test can ever do is compare your DNA – certain parts of the DNA – to a reference sample,” Kowal said. “If somebody gets a match with one of these services, what it could actually reflect is a match between their European ancestry and shared European ancestry with the Aboriginal people in their sample.”
Dr Lyndon Ormond-Parker, a descendent of the Alyawarr people in the Northern Territory and a research fellow at the Melbourne School of Population and Global Health, told The Saturday Paper that even an accurate genetic test wouldn’t, on its own, determine whether someone is an Indigenous Australian.
Australia’s legal definition of Indigeneity has three components: that someone is of Aboriginal and Torres Strait Islander descent; that someone identifies as an Aboriginal or Torres Strait Islander person; and that they are accepted into their community. “All three must work together, and that’s how you have an identity of being Aboriginal and Torres Strait Islander under Australian law,” Ormond-Parker said.
“What also concerns me is when people go out and collect DNA from Indigenous people with or without their consent and then that becomes publicly available for organisations.”
The Saturday Paper has so far found forensic anonymised genetic information of Indigenous Australians in three scientific articles that are available to the public. In accordance with Australia’s Privacy Act 1988, federal bodies can share people’s health information for reasons such as assisting research in the public interest and safety if they remove their personal information. The states and territories have similar requirements.
Australian human rights ethics committees also review research proposals involving human participants to ensure they are ethically sound, including not identifying participants unless they consent.
At the Australian National University, the National Centre for Indigenous Genomics (NCIG) is trying to develop the gold ethical standard to conduct research that involves Indigenous Australians.
NCIG was born out of concerns around how ANU would treat thousands of blood samples the university collected from Indigenous Australians between the 1960s and 1990s for scientific research. After protest from Indigenous communities, ANU placed a moratorium on using the samples and they went into storage.
The centre, which is creating one of the first Indigenous Australian reference genomes, says samples will only be examined with the consent of the families to whom the samples belong and to benefit First Peoples, such as to design medicine tailored for them.
“It is clear from the huge success of [ancestry testing providers] Ancestry, 23andMe and the like, that there is an enormous demand for information about ancestral identity,” NCIG’s director, Professor Simon Easteal, told The Saturday Paper. “My view is that it is more useful to spend effort developing a culturally appropriate and scientifically robust way of meeting that demand.”
The development of more ethical research practices around Indigenous DNA testing is a positive step. But Rachael Hocking says she’s concerned about the profound impact that giving imprecise predictions about Indigenous Australian heritage could be having.
“To know that someone, like myself, might contact this company to discover their ancestry and be given results we know not to be true is, frankly, terrifying. I feel for our brothers and sisters who were part of the Stolen Generations, perhaps looking for some closure, only to be given results … which say, ‘You are not black’,” she said.
An Australian Competition and Consumer Commission spokesperson told The Saturday Paper that overseas-based companies, which sell to the Australian market, must abide by the Australian consumer law. “Under the Australian consumer law, businesses are not allowed to make statements that are incorrect or likely to create a false impression,” they said.
Asked for comment about the source of its genetic data, as well as Gunn’s concerns about its methodology, DNA Tribes did not respond.
However, Hocking says questionable results don’t factor into her understanding of her identity. “Dodgy DNA results can’t take away what I know about my language, my Jukurrpa [Dreaming], my family,” she said. “And that is because I already identified as Aboriginal, as Warlpiri.”
This article was first published in the print edition of The Saturday Paper on Oct 20, 2018 as "Indigeneity and DNA tests".
A free press is one you pay for. In the short term, the economic fallout from coronavirus has taken about a third of our revenue. We will survive this crisis, but we need the support of readers. Now is the time to subscribe.
Letters & Editorial