The DNA ancestry craze from The Stream

DNA testing was featured this week in a two-part TV programme on The Stream, an online social-media-oriented TV show hosted by the Al Jazeera media network. The guests on these programmes were: Julie Granka, a population geneticist with AncestryDNA, Joseph Graves, an Evolutionary Biologist at North Carolina A&T State University, and Kevin Jones Giddins, an AncestryDNA user who had found his mother through DNA testing. Questions were submitted via social media and there was a very good and balanced debate on the pros and cons of DNA testing, with a particularly valuable contribution from Professor Joseph Graves. He provided some interesting insights into the methodology used for admixture tests – the reports received from the testing companies that give you the percentages of DNA that you share with different populations. These tests cannot provide the granularity that the companies claim. They can distinguish between populations at the continental level but cannot generally provide meaningful breakdowns at the country level in Europe, Africa or elsewhere. (This might change in future. The new test from Living DNA provides regional breakdowns within the British Isles though we have not yet seen any customer results.)

The programmes mostly focused on results from AncestryDNA. It would have been helpful if they could have shown results from 23andMe and Family Tree DNA as well, and asked scientists from these two companies to contribute to the debate.

There was one matter of concern when one of the presenters, Malika, showed some results that her father had received from a company called African Ancestry back in 2008. You could see from the screenshots that her father had only taken very low-resolution tests (an HVR1 mtDNA test and a 9-marker Y-DNA test) yet the family were told by the company that they were descended from specific tribes in Chad, Cameroon and Ghana. When so few markers are used for a Y-DNA or mtDNA test, the genetic signature could potentially be shared with thousands of people from many different countries, and it is simply not possible to infer an origin in a specific country. Better resolution can be obtained from a full mtDNA sequence test or a comprehensive Y-chromosome sequencing test, but even then it's not generally possible to pinpoint a specific country of origin. An additional problem is that most of the people in the company databases are Americans and Europeans, and there simply aren't enough reference populations from Africa to make such inferences. Unfortunately these Y-DNA and mtDNA results were glossed over in the programme and the limitations of the African Ancestry tests weren't discussed.

However, these are minor quibbles in two otherwise very interesting programmes. It was also good for once to see more of an emphasis on using DNA testing for genealogical matches rather than a mistaken attempt to discover "where you're from". Each programme lasts for just 25 minutes, and they are both well worth watching. Part 1 covers The DNA ancestry craze and Part 2 looked at The science and security of DNA testing. You can click on the videos below to watch the programmes.

Living DNA – a new genetic ancestry test providing comparisons with the People of the British Isles dataset

Over five years ago I attended a one-day conference in Cardiff in Wales on Ancient Britons, Europe and Wales. At the conference Professor Sir Walter Bodmer presented the first results from the People of the British Isles (POBI) project which were hot off the press having only been analysed three weeks previously. The audience were completely blown away by the results. For the first time researchers had been able to detect regional differences between the people of Britain based on their DNA. Sir Walter shared a map with us and you could even see that some counties, such as Devon and Cornwall, stood out as distinct regions in their own right.

I don't have a photograph of the map that we saw that day in Cardiff but it was probably very little changed from the map that appeared in the published paper in Nature in 2015 and which you can see here on the Wellcome Trust website. I remember thinking at the time that this research opened up the tantalising possibility of being able to receive an admixture report from a genetic ancestry company which would allow you to compare your results with the POBI data and see how much of your ancestry came from Devon or Cornwall and other regions of Britain.

I am very pleased to report that that day has now arrived with the launch of a new DNA test from a company called Living DNA! This is the first genetic ancestry test to incorporate data from the POBI Project and to give customers the percentages of DNA that they share with people from different regions in Britain.


Disclosure. I should declare here that I have been involved in informal discussions with DNA Worldwide over the development of their product. I have not receive any payment but, in return for my advice, I will receive a free Living DNA account. The company have had access to my raw data files from AncestryDNA, 23andMe, Family Tree DNA for testing purposes. Yesterday I had a Skype briefing about the new test from David Nicholson of Living DNA. I also received a preview of my own results, though I don't yet have access to my online account. There is still some tweaking going on behind the scenes and the results I've received are likely to be changed, so I will report on my results once I have my account.

The original POBI map, which formed the centrepiece of the Nature paper, featured 17 regions. For the new Living DNA test the POBI data has been re-analysed using improved methods and the data has been clustered into 21 areas across the British Isles.

The Living DNA test also includes a standard ancestry report providing admixture percentages. This analysis is based on 80 worldwide regions. (Note that the test does not currently report Jewish or Aboriginal ancestry though hopefully these reference populations will be added in the future.)


A unique feature of the test is the "back in time" map which allows you to see how your DNA matches with populations across a range of historical dates. Historical population statistics have been used for the mapping. All the existing tests on the market are trying to pick up on the small and very recent differences in our DNA but this means that the bigger picture of our universal relatedness gets lost. Recent studies have shown that you only have to go back about 5000 years or so before we reach a point where we all share the same ancestors. I hope that this new test will go some way towards educating people about our common ancestry and the knowledge that there is no such thing as "race".

Here are some sample reports which have been provided by the company. The first map shows "the areas of the world where you share genetic ancestry in recent times" at the continental level.

This second map provides regional resolution.

On the regional map you can click through and see how your ancestry is broken down within the different regions. If you have ancestry from Britain you will be able to see your shared percentages of DNA in comparison with the local populations. The test has also been built so that people with mixed ancestry from other countries get more detail than ever before.

 

Here is the family ancestry visualisation.

The Living DNA test is currently available as a new standalone test but they are also planning to accept imported data from other companies for a small fee. It is hoped that this feature will be available by the end of the year. I understand that GedMatch are preparing to accept uploads of raw data from Living DNA. I hope that Family Tree DNA will also be able to accept the raw data as this test is likely to attract a lot of interest and we want to ensure that people who are interested in using their results for genealogical matching are able to do so.

The Living DNA test was offered to the public for the first time last weekend at the New Scientist Live show at the ExCeL arena in London. The test can be ordered from the website at www.livingdna.com and the kits will be shipped in mid-October.

Living DNA is a partnership between the British company DNA Worldwide and the European company Eurofins. Here are the company details from the website:

Living DNA is a collaboration of over 100 world-leading scientists, academic researchers and genetic experts. The team is led by DNA Worldwide Group, a DNA testing company, whose services are used by every court in the UK. The company is run by David Nicholson and Hannah Morden who saw an opportunity to show humanity that we are all made up of all of us, dissolving the concept of race. It was launched in 2016 after two years of intensive development but its parent company DNA Worldwide Group has been operating since 2004, and employs over 35 people from its head office in Somerset, UK.

Although Living DNA is a for profit company they also intend to use their test for educational purposes to show how we are all connected to each other. They are working with an organisation called Show Racism the Red Card, a UK charity which works to combat racism. The company are working with a school in East London and ten schools around Europe. They also have PhD students working with them on different projects.

The following video provides further information about the company and the philosophy behind the test.

Who are Living DNA from Living DNA on Vimeo.

Technical details
Here are some of the technical details which I have gleaned from the website or which have been given to me by David Nicholson during my Skype briefing:

• The tests are being done on a new custom Illumina Global Screening Array chip. Living DNA are the first company in the world to have access to this chip. This chip is a replacement for the Illumina OmniExpress chip which is in the process of being phased out. The new chip has been designed specifically for imputation and achieves a very high accuracy rate of 99.9%. (The datasheet for the Illumina chip can be found here.)
• Medical SNPs are included on the chip and there is considerable overlap with the 23andMe chip. (I note from the Illumina press release that 23andMe have signed up to the Global Screening Array chip and I wonder if a v5 23andMe test might be in the pipeline.)

• The following SNPs are on the Living DNA chip:

– 638,000 autosomal SNPs

– 17,800 X-chromosome SNPs

– 22,500 Y-SNPs

– 4,700 mtDNA SNPs

• The Y-DNA SNPs have been selected from the 1000 Genomes Project and research papers, and checked and reviewed on the ISOGG SNP tree, The Y-SNP test has been developed in collaboration with Professor Mark Jobling and Pille Hallast at the University of Leicester.
• The mtDNA SNPs have been selected by a team of scientists at Innsbruck University in Austria. They have used Phylotree and the latest tree builds to select the most useful mtDNA SNPs.
• The autosomal component of the test has been developed by Daniel Lawson, Grady Weinberg, Daniel Falush, Garrett Hellenthal and Simon Myers. Lawson, Falush, Hellenthal and Myers were the team who developed the fineSTRUCTURE program used for the analysis of the POBI data and they were all authors on the POBI paper. The Living DNA test uses a refined version of that program upscaled for commercial use. fineSTRUCTURE is a haplotype-based method which takes advantage of the fact that SNPs travel together, and provides a more detailed analysis.
• The autosomal DNA data is phased prior to processing.
• White papers will be published soon with more information.
• It will be possible to share results on the platform with friends and family and to manage multiple kits from a single account.
• The test is done using a mouth swab. There is no liquid involved. This method has been chosen to facilitate international shipping. 
• The developers have worked hard to ensure that the new chip is backwards compatible with other chips so that data can be accepted from all companies. The intention is to make the transfer process as easy as possible, preferably by interfacing with company APIs where available.
• Customers will be able to download their raw data. Raw data will be imputable so that it is backwards compatible with older chips.
• The test is available worldwide but prices will vary from country to country depending on local taxes and shipping rates. Negotiating global shipping rates and regulations is an extremely complicated process and prices are subject to change. There will be no hidden extras and the cost of shipping (including return shipping) will be included in the advertised price. The website currently shows prices for the UK, USA, Canada, Australia and Europe, and a standardised international price for other countries.
• Participants will have the opportunity to opt into academic research studies.

Future plans

• It is planned to add in more data as it becomes available. It is hoped in particular to get some samples from Ireland.
• Updated ancestry reports will be provided free of charge and there will be no subscription fees to receive the updates.
• There are plans to incorporate ancient DNA data.
• A chromosome browser view is in the pipeline.
• There is a long feature list in development, and the company hope to receive feedback from the community and will develop the tools that people want.

Privacy

• There is a major focus on privacy.  The customer decides what happens to their data.
• All the processing is done within the European Union.
• Data will not be sold on to pharmaceutical companies.
• Care has been taken to ensure that people understand the implications of testing.
• Children under the age of 18 can take the test but will be required to re-consent once they reach the age of 18.
• There will be an option to nominate a beneficiary.
• For full details read the company's page on Privacy and Security.

Conclusion

This is a very exciting new test. It is a very pleasant change to have a test that specifically caters for the British and European market. The admixture tests currently provided by the big three genetic genealogy companies are primarily serving the very different needs of the American market. Access to the POBI dataset is what we've all been waiting for and I'm sure that everyone is going to be keen to see how their results compare. I'm also particularly pleased to see the back in time map which I hope will be a useful educational tool and will help to improve our understanding of our shared human ancestry. This is a very welcome addition to the genetic ancestry marketplace, and I look forward to seeing some interesting new developments in the future.

Update 10 October 2016
The Living DNA team are hoping to improve their Irish reference panel. To this end they would like to recruit people with four grandparents born within 80 km (50 miles) of each other in Ireland. If you have already tested with 23andMe, Family Tree DNA or AncestryDNA you can transfer your results free of charge (23andMe data is preferred if available). If not, you can benefit from a discounted rate for a new test. Here is the link to transfer your results: https://www.livingdna.com/en-gb/irish-dna-research-project

Further reading
- See Chris Paton's blog for the official press release.
- For further commentary see John Reid's blog post More on the Living DNA genetic ancestry test and plans.
- Razib Khan provides some insights on admixture in his blog post on Scale of time and space in admixture.

© 2016 Debbie Kennett

Rockstar Genealogists 2016 - the voting is now open

John Reid, who writes the Anglo-Celtic Connections blog, is once again organising the annual Rockstar Genealogists poll.

There have been a few changes this year to try to overcome the problem of multiple voting, and there are also changes in the way that the votes will be reported, so it will be interesting to see if there will be any surprises in store.

I'm pleased to see that so many of my fellow members of the Guild of One-Name Studies are on this list including our Chairman Paul Howes who is a very welcome first-time addition. I have fond memories of the recent Guild of One-Name Studies conference where Paul volunteered to fill a hole in the programme at the last minute and held the audience spellbound with an inspiring, unscripted and Powerpoint-free talk on some of the success stories of his one-name study.

All the respected UK genealogy speakers and writers, such as Nick Barratt, Peter Calver (who writes the Lost Cousins newsletter), Colin Chapman, Else Churchill, Audrey Collins, John Hanson, Celia Heritage, Janet Few, Michael Gandy, Emma Jolly, Chris Paton, Alec Tritton and John Titford, are included in the list. I feel very honoured to be included in such illustrious company.

Genetic genealogy is well represented with all the big names from the US receiving nominations. I'm one of four genetic genealogists flying the flag for the UK. I am in excellent company with Maurice Gleeson, Chris Pomery and Geoff Swinfield.

John Reid's intention is to provide a resource to help volunteers select speakers for their events. The nominations on the list can serve as a useful starting point but it's also a good idea to get feedback from other genealogists in each country who have worked with the individuals concerned before booking a speaker or commissioning an article or book.

Before you vote I urge you to read Janet Few's blog post on Rockstars and unsung heroes. Janet very eloquently and humorously sums up my thoughts on the contest, and says all that needs to be said.

Thank you to John Reid for organising the poll.

Make sure you cast your vote here. You can vote for as many people as you want but make sure you use your votes wisely.

Update
Thank you to everyone who voted for me. I was honoured to be placed third in the list for England. Scotland and Wales though I'm still very uncomfortable with the whole idea of holding such a poll.

Here is the full list of gold medallists: http://anglo-celtic-connections.blogspot.co.uk/2014/09/rockstar-genealogists-gold-medalists.html

Here is the list of silver and bronze medallists;

http://anglo-celtic-connections.blogspot.co.uk/2016/09/rockstar-genealogists-2016-silver-and.html

John Reid published a further blog post in which he published the comments received by the voters. There were some interesting comments raised some of which, if true, are a cause for concern for the genealogy community: http://anglo-celtic-connections.blogspot.co.uk/2016/09/rockstar-genealogist-2016-comments.html

Kerry Scott at the Clue Wagon blog questions the validity of holding such a contest and I share her concerns: http://www.cluewagon.com/2016/09/the-rockstar-genealogist-contest-is-a-fiesta-of-awkward-lets-change-this/

The new GPS Origins test from DNA Diagnostics Center – caveat emptor

We've had a few questions in our ISOGG Facebook group about a new GPS Origins test  so I thought I'd take a look at it. I'm not aware of anyone who has tested with the company and received results. At present the test is only available in the US and Canada during the initial launch period but there are plans to make it available in other countries in 2017.

You can either order a new test from scratch for an introductory price of US $149 or you can transfer your raw data from AncestryDNA, Family Tree DNA or 23andMe to order a GPS Origins report for $59. The transfer is not currently active and is advertised as "coming soon".

Here is a description of the test from the home page of the company website:

The GPS Origins™ (Geographic Population Structure) ancestry test uses the latest genetic research and a new ancestral tracking technique to pinpoint much more precisely where your DNA was formed. The GPS Origins™ test indicates the town or village where groups of your ancestors from different cultures met - building a rich picture of the migration journeys that formed your deep genealogical heritage.

There is further information in the How it Works section of the website:

The GPS Origins™ test then traces the migration route of your DNA back to where it originated from and dated the age of your DNA signature. It does that for both your maternal and paternal lineages indicating where your DNA began. Your results are detailed in a report that reveals your ancestral origins. 

Your personalized report identifies your top three Ancestral Origins (the Gene Pools or ancestral communities that contributed significant portions of your genetic makeup) and shows the percentages of DNA you inherited from each. The report is much more detailed than an estimate of ‘ethnicity’.

Your story starts with the shared origin of all humankind, and then builds into a vibrant picture of where and how your ancestors lived, and the conditions that led them to migrate. Your report contains maps illustrating the two most important migration journeys and describes how your ancestors’ circumstances changed as they crossed continents to find better lives. The report concludes with a summary page of helpful links to discover additional information to reveal your ancestral origins...

The GPS Origins™ test is a revolutionary ancestry test that enables you to trace where your DNA was formed over 1,000 years ago, along with its migration routes, down to the nearest village or town.

Current ancestry DNA tests locate where your DNA formed within countries or continents. Typically you find that some of your ancestries come from Western Europe, Africa or South Asia, and given broad estimates of your ethnicity. These tests generally cannot identify your origins to particular countries or locations.

It would be very exciting to have a DNA test which could provide this level of accuracy but unfortunately the reality is probably not going to live up to the hype. The GPS Origins test appears to be an updated version of a test originally offered by a company called Prosapia Genetics. The Prosapia test was based on algorithms described in a paper by Elhaik et al Geographic population structure analysis of worldwide human populations infers their biogeographical origins (Nature Communications, 2014). Customers were given a co-ordinate indicating their supposed place of origin 1000 years ago. There were a lot of complaints from customers, many of whom discovered that they had an aquatic origin in the middle of an ocean or river! See my blog post Driving in the wrong direction with a dodgy DNA satnav. Although the Prosapia Genetics website is still live there has not been any activity on their Facebook page or Twitter account for a long time, and it is not clear if the test is still on sale.

The new GPS Origins test has been developed by Dr Eran Elhaik, the lead author of the 2014 Nature Communications paper, and is using the same algorithms. The company claim in their FAQs that "The accuracy of GPS was demonstrated by identifying the DNA signature of ancient Ashkenazic Jews and their formation approximately 1500-2000 years ago." They are referring to a recent study by Das et al in Genome Biology and Evolution on Localizing Ashkenazic Jews to primeval villages in the ancient Iranian lands of Ashkenaz. Eran Elhaik was one of the co-authors of this paper. Both of these papers are highlighted on the GPS Origins website in a section on Research behind the GPS Origins™ test.

Unfortunately the company has failed to mention that the Das et al paper has been heavily criticised and that the original research on which the GPS test is based has been called into question. See the paper by Flegontov et al in Genome Biology and Evolution (2016) on the Pitfalls of the geographic population structure (GPS) approach applied to human genetic history: a case study of Ashkenazi Jews.

Here is the abstract from the paper:

In a recent interdisciplinary study, Das and co-authors have attempted to trace the homeland of Ashkenazi Jews and of their historical language, Yiddish (Das et al. 2016. Localizing Ashkenazic Jews to Primeval Villages in the Ancient Iranian Lands of Ashkenaz. Genome Biology and Evolution). Das and co-authors applied the geographic population structure (GPS) method to autosomal genotyping data and inferred geographic coordinates of populations supposedly ancestral to Ashkenazi Jews, placing them in Eastern Turkey. They argued that this unexpected genetic result goes against the widely accepted notion of Ashkenazi origin in the Levant, and speculated that Yiddish was originally a Slavic language strongly influenced by Iranian and Turkic languages, and later remodeled completely under Germanic influence. In our view, there are major conceptual problems with both the genetic and linguistic parts of the work. We argue that GPS is a provenancing tool suited to inferring the geographic region where a modern and recently unadmixed genome is most likely to arise, but is hardly suitable for admixed populations and for tracing ancestry up to 1000 years before present, as its authors have previously claimed. Moreover, all methods of historical linguistics concur that Yiddish is a Germanic language, with no reliable evidence for Slavic, Iranian, or Turkic substrata.

The Das et al paper was also criticised from a linguistic perspective by Marion Aptroot in a paper in Genome Biology and Evolution (2016) entitled Yiddish language and Ashkenazic Jews: a perspective from culture, language, and literature

Here is the abstract from her paper:

The typology of Yiddish and the name Ashkenaz cannot serve as arguments to support the theory put forward by Das et al. (2016). (Localizing Ashkenazic Jews to primeval villages in the ancient Iranian lands of Ashkenaz. Genome Biol Evol. 8:1132–1149.) that the origin of Ashkenazic Jews can be located in ancient Iran. Yiddish is a Germanic, not a Slavic language. The history of the use of the term Ashkenaz from the Middle Ages onward is well documented. Ashkenazic Jewry is named for the Hebrew and Yiddish designation for Germany, originally a Biblical term.

The new GPS Origins test is therefore based on an unproven methodology, which can't be replicated and which does not produce the village-level or even country-level accuracy that has been claimed. It is merely giving you geographical co-ordinates which represent an average of your closest matches in a reference database of modern populations. The whole concept of the test is fatally flawed because it is simply not possible to use the DNA of living people to identify at an individual level where your DNA originated 1000 or more years ago. Similarly we cannot trace the individual migration journeys of our ancestors from modern DNA. Ancient DNA is providing interesting new insights about past populations but these inferences apply to everyone's ancestors and are not unique to an individual.

The new test does seem to have some "improvements" compared to the old Prosapia test. Rather than providing one co-ordinate to represent the origins of your ancestors it would appear that the new test will provide a place of origin for both parents. Presumably this opens up the possibility of customers having origins in two rivers or oceans instead of just one!

If anyone does get results from this test I would be interested to hear from you and to see a sample report.

Further controversies
It should be noted that Eran Elhaik has been at the centre of a number of controversies in recent years.

In 2013 Elhaik published a paper in Genome Biology and Evolution on The missing link of Jewish European ancestry: contrasting the Rhineland and the Khazarian hypotheses. His findings were refuted by a leading team of Jewish researchers in a paper entitled No evidence from genome-wide data of a Khazar origin for the Ashkenazi Jews (Behar et al, Human Biology, 2013).

In 2014 Elhaik was the lead author of a letter in the European Journal of Human Genetics on The ‘extremely ancient’ chromosome that isn’t: a forensic bioinformatic investigation of Albert Perry’s X-degenerate portion of the Y chromosome. This letter was a critique of  a paper by Mendez et al entitled An African American paternal lineage adds an extremely ancient root to the human Y chromosome phylogenetic tree (American Journal of Human Genetics, 2013). Elhaik even published a mocking video on YouTube explaining where he thought the authors had made mistakes:

Mendez et al published a robust rebuttal Reply to 'The extremely ancient' chromosome that isn't' by Elhaik et al (European Journal of Human Genetics, 2013) pointing out that the authors' criticisms resulted from "a misunderstanding of population genetic theory, as well as a misrepresentation of the methodology of Mendez et al". They also discussed the "technical and conceptual flaws" that undermined the claims.

Update 26th August 2016
There is an ongoing thread on the Anthrogenica forum about the GPS Origins test. Some people have shared screenshots of their test results.

Update 30th November 2016
I have now tested with GPS Origins. See my blog post: A review of the GPS Origins test: four ethniticites and four reports.

Update 1st December 2016

I have discovered that the linguist Asya Pereltsvaig has published two critiques of Das et al 2016 on her blog:

• Is Yiddish to be traced to Turkey? 29 April 2016
• Turkey, Shmurkey! — A Response to Das et al. (2016) 10 June 2016

Update 18th July 2017

I had the opportunity compare tests done on the GPS Origins chip compared with transfers from AncestryDNA and 23andMe. See my blog post The GPS origins test - the DREAM chip compared with AncestryDNA and 23andMe transfers.

© 2016 Debbie Kennett