Figure 2: Outline tree for the claimants in the Pringle case, illustrating the line of descent from the 1st Baronet disputed in the case. © John Cleary
When using DNA samples from relatives, new acquaintances or unknown people there are two big principles you should keep in mind. Firstly there is the sensitivity of DNA, which carries delicate information about people – health and disease risks, ancestry or potentially nonancestry, and ethnic origins. Everyone has the right to privacy over their DNA, even when taking part in a hobby that involves sharing and publishing parts of their DNA sequences. If you are organising a family DNA project, you need to act to maintain the levels of privacy that each of your test-takers would feel comfortable with.
The second thing to bear in mind is that you will be leading a research project that involves other people. Your discoveries could in some cases be unexpected, embarrassing or otherwise difficult to accept. Most people will probably enjoy hearing about what you discover, but there will always be a small number of discoveries that conflict with memories or expectations and potentially upset relations between family members. When conducting research involving other people, it is essential to use ‘do no harm’ as the basic principle to underpin every decision that you make. Anyone you invite to take part in your research by giving a DNA sample has a right to know in advance why you are taking the sample and exactly how you will use it – before they give that sample.
If fully informed in this way, the people you approach to test their DNA can give informed consent. Respecting their ownership of their own DNA sample and results, and their right to determine the limits for its use and public exposure, is the basis of ethical research in genetic genealogy.
Privacy is a concept growing in importance in today’s data-driven society; because DNA encodes information about individuals, contemporary beliefs and anxieties about privacy are deeply connected with it. While some years ago the term ‘privacy’ would be used by most people to refer to their domestic world, separate from the public worlds of work, entertainment or politics, today it captures a tension between the public sharing of data about ourselves via social media, the collection of information about the details of our personalities and lives by public and private organisations, and the anxieties we feel about being drawn into these webs of information.
With genealogical DNA testing, many test-takers keenly feel these tensions, and the people who work with you will have differing degrees of comfort about the public sharing of their results. Their willingness to share results and their comfort in doing so is affected by the gains they perceive from it (for their own research and for the general good of helping others with theirs), and the control they feel they have over who will see their data and what use they will put it to. How privacy feelings are developing in a data-sharing society has been explored in a recent article by Lee Rainie and Maeve Duggan, ‘Privacy and Information Sharing’, a thought-provoking follow-up piece on the broader topic.
Considering genetic genealogy and DNA testing, you must first decide your own attitudes towards privacy. You will need to consider that large parts of your DNA are shared by your family members, distant relatives and even complete strangers. If you are testing the Y chromosome or mtDNA, then almost identical Ys will be found in all your patrilineal male relatives, and mtDNA in your matrilineal ones – publishing your data means you are publishing theirs too. There are relatively few sensitive regions in the Y data that you are likely to publish, but there are several significant health-related sites on the mtDNA coding region, which even test-takers with an open attitude to data-sharing lean towards keeping private.
Autosomal DNA is shared in progressively smaller amounts as relatives become more distant from you, but your siblings, parents, children and other close relatives out to second cousins share large amounts of your sequence. Here is the dilemma: it is these shared sequences that allow us to find our unknown relatives – but they are not only ours, they are theirs too, and belong to others who have not tested, or might not want to test and share. Some argue that we are becoming aware of privacy just as we are losing it – and that the world of the near future may expect it less, as we are captured in ever-bigger data. Younger people already seem more comfortable with lower degrees of data privacy, as evidence of social media use may suggest. But there are potential minefields here that need to be navigated if you test other people and share their data – and especially if you begin to run larger projects with multiple test-takers, either informally or as an administrator of an established project (see Chapter 8).
Here are some elements of privacy in respect of managing DNA test results that you may want to build into your practice, when working with other people’s DNA, or when permitting others to make use of yours:
• Control of your information – which platforms may it be uploaded to?
• Right to limit (or permit) access to your information– who may see it under what conditions? Which levels of visibility may it have on public platforms?
• Freedom from surveillance or being identified (the ‘right to be let alone’) – confidentiality, including whether and how your name or identifying information is, or is not, linked to your data.
• Security from misuse of your data – it must not be possible to identify information about you that you had not chosen to reveal, or for your personal contact details to be farmed by external agents.
• Protection against secondary uses of your data you did not consciously permit – like reselling your data or re-parcelling it for other research purposes, unless you had explicitly given consent for it.
These aspects overlap to a certain extent, but underlying them all are control (over who accesses the data, when and how); confidentiality and security (of your data when it is in the hands of others); and use (what was agreed to and nothing else that was not openly agreed to).
Many countries have legislation in place to regulate data privacy and protect citizens. In May 2018 the General Data Protection Regulation (GDPR) of the European Union took effect, implemented and supported in the UK by the Data Protection Act, 2018. GDPR and DPA 2018 grant strong powers to the Information Commissioner to protect the personal information that organisations hold on individuals resident within the EU and UK, with stringent fines for misuse. While individual genetic genealogists are unlikely to fall foul of GDPR, the Regulation and Act put the onus on ‘data processors’ to demonstrate they are applying maximum care and confidentiality over the private, personal information of the people they work with – and this is best practice for any genetic genealogist who handles the personal and DNA data of another person.
Linked to privacy rights over data is ethical conduct of research. In universities and professional research organisations, research proposals are vetted by ethics committees to ensure that people involved in research or impacted by it are protected from risks arising from their involvement. A hobby genealogist working with documents has probably had limited concerns over the ethical impact of their pastime, apart from considering the effect a shocking family story may have on a family member. But with testing relatives’ DNA becoming a common activity, the close involvement of other people forces the researcher to think about the ethics of research more in the manner of a professional researcher.
As well as the privacy issues discussed above, a number of risks to the wellbeing of your research partners are created by testing DNA. The ‘non-paternity event’ (NPE) (discussed in Chapter 5) can be a colourful family story when it is a few centuries back in time, but when it turns out to be someone’s grandparent or parent that was the child of an unknown past affair, this can be upsetting to a person and undermine their happiness. If the objective is to investigate a rumoured, suspected NPE in recent generations that directly impacts upon a living person, it should be prepared carefully, with the involvement of the person(s) who will be most affected, so they are not harmed by the discovery. See p20 for a case in which the revelation of a recent NPE by a DNA test had profound consequences for the people involved.
DNA tests bought as presents have revealed unknown children or half-siblings, and such events can tear families apart. There is a substantial cottage industry of adoption searching in the genetic genealogy world, with both professional genetic genealogists and volunteer ‘search angels’ offering help to adoptees or donor-conceived children to find their birth parents. Even done with sensitivity and accuracy, this can be a traumatic experience for discovered parents who were promised anonymity, and for adopted children who must face parents who gave them up as children; if done incorrectly or insensitively, there is a tremendous prospect for harm. The more people impacted by a study, or the greater the risk of traumatic information being revealed, the more the genetic genealogist should think about adopting ethical procedures used by professional research specialists.
Here are some basic principles in research ethics. These are adapted for genetic genealogy from the Key Ethics Principles put forward by the Economic and Social Research Council (ESRC) of the UK in its The Research Ethics Guidebook (website). Though intended for students and academic researchers, the Guidebook is worth exploring for further information on research ethics. ‘Study’ here refers to any project you may develop using DNA to answer genealogical questions, and ‘partners’ is meant generally to include anyone who collaborates with you, helps raise funds, carries out interpretation of results, or donates their DNA for you to sequence.
• Design your study with care to ensure its quality and integrity. This makes sure it is worth doing, and will not waste the time, resources or money of you or your partners.
• Inform your partners fully of what you want to do, how their participation will help you, and be upfront about any risks. This is one component of obtaining informed consent.
• Guarantee the confidentiality of your partners’ data, and preserve the anonymity of your partners’ DNA sequences. Only ever publish their real names if it is necessary to do so and if they have given prior informed consent for you to do so.
• Participation by your partners must be voluntary and not coerced. They have the right to withdraw from your study at any time and to withdraw their data from the study if they do so. This is the second component of obtaining informed consent.
• Do no harm. Analyse risks in advance and ensure your partners are aware of them.
• Do good. Be independent of, or open and honest about, any conflicts of interest.
The ethical research principles in the last section are underpinned by the process of obtaining informed consent from research partners. Imagine you have been introduced to a possible distant relative, or someone who has your surname of interest from a line you have not tested before. Maybe this person is not especially interested in genealogy and knows little about it, but wants to help you and agrees to test. You need to explain in full what you are trying to discover, how their DNA may help, and what the process of sampling, testing and analysis involves. Sometimes it can be difficult to cover all this in a conversation – there may be other people or things to do that can distract. Your partner may be generally interested, but not particularly interested in the details. They may be confused by any risks you explain, if suddenly the process starts to sound risky or dangerous. How best to obtain informed consent in this type of situation?
Professional researchers are under the oversight of ethics committees, and will usually need to collect written and signed evidence of consent. Consent forms generally contain two parts:
• An information statement: a concise summary of the goals of the research, with key details on how the person will be involved, what they will need to do, and of any risks.
• A consent statement: a pro forma which the person will sign to signal their agreement to the process as they understand it; and to acknowledge they have been informed of the right of withdrawal.
Professional researchers may also provide a third party, who is not the researcher, who can be contacted in case the research partner feels the need to discuss an issue with someone who is not the investigator.
It is arguable that a genetic genealogy project, particularly a smallscale one, would not need to follow all the steps above. Informed consent statements can be long and use complex or legalistic language, which can be off-putting in the world of pastime genealogy research. For genealogists, their partners are often family or possible family, and a bureaucratic process can get in the way of building relationships with them, seeming cold and impersonal – at odds with the personal nature of investigating shared ancestry.
In fact, such misgivings are often felt by professional researchers like anthropologists or sociologists too, as their research with people is built around trust, exploring shared values and the appearance of friendship. Anthropologists accept that informed consent can be given in different ways, including verbal assent in conversation. Genealogists need to develop ways of achieving consent that suit their needs. Receiving a verbal agreement may well be enough, or a short email that confirms the partner has thought about what was said and is happy to agree with it. An informal written information statement to leave with people who have tested, or a short email laying out the key facts, will help to meet the informed component, as well as giving the partners something to refer to if they forget the details. Figure 1 gives an example of how this could look.
Thanks very much for agreeing to take part in my study. I really appreciate it. Here’s what I’m trying to do – our surname Squitch is very unusual, and it only seems to be found in Cheshire. Since you live there, you might be from one of the oldest branches of the name to live continuously in the county, and we might be distantly related.
We will ask the testing company to read 37 markers on your Y chromosome, called STRs. We will compare these with the same 37 STRs in my test, and some other Squitches who have Cheshire roots. Our test results have a lot of similarities with some interesting shared differences, or mutations. We will see if you have the same patterns in your STRs which would suggest we could be related.
The testing company will take about 2–3 months to do the sequencing. When it is done, your results will appear in the results table online I showed you. We agreed your results will be on public view, but let me know if you want to change that as we can do it easily and set it for other members only to view. I will help you understand your results and show what they mean for our shared genealogy, and I will answer any questions you have. And don’t forget – if you feel uncomfortable about any aspect of this, do remember you can withdraw from our study and stop the test, or withdraw your results, or make them private, at any time. Let me know and I’ll show you what to do.
Figure 1: One way an informal and friendly information statement, sent by email after a meeting with a testing partner, could look.
Family members and testing partners introduced as friends could feel uncomfortable signing consent statements, but finding an appropriate way to receive and record that consent is important. You may find an email works well, or if you plan to do any oral history interviewing of people that you test, you can ask a question about consent on the recording. If you do testing of a more formal type, for example associated with a sensitive issue like adoption searching, or for professional work within genealogy, then formal consent forms become more advisable. Often it is a judgement call, but the baseline principle is that anyone who joins your study should receive full information in an appropriate way, and assent freely to what you plan to do. And if you change or add to what you plan to do with that partner’s test results, then you should obtain informed consent once again for those new research plans, before commencing. See the next section for a case in which this became deeply relevant.
An important judgement was made by the Judicial Committee of the Privy Council, which has jurisdiction over titles awarded by the British Crown, in June 2016. On the death of Sir Steuart Pringle of Stichill, 10th baronet, in 2013, two men claimed the heirship to the title – the 10th baronet’s son, Simon, and his cousin, Murray. The reason for the disputed succession lay in the discovery that the 10th baronet’s father was not the biological son of his own father. The 9th baronet was apparently a legitimate son to his father the 8th baronet, as he was born in wedlock (though just over 7 months after his parents’ marriage). His legal legitimacy was not challenged in the case, but one of the claimants had something else never used before in a case of this type – DNA evidence.
Figure 2: Outline tree for the claimants in the Pringle case, illustrating the line of descent from the 1st Baronet disputed in the case. © John Cleary
The baronetcy of Stichill was granted by Charles II in 1683 to Robert Pringle of Stichill and the male heirs of his body (ac heredibus masculis de suo corpore in the words of the grant). When a baronet dies, the passage of the title is not automatic, but has to be claimed by an heir whose claim is compatible with the terms of the grant.
Sir Steuart Pringle had donated a Y-DNA sample to a DNA Surname Project for the Pringle surname which had been set up to find relatives of the last Pringle Clan Chief, who died in 1738. This revealed that he did not match the Y chromosomes of other Pringles from the family. This is familiar to genetic genealogists as a common type of NPE (nonpaternity event, or ‘not the parent expected’), when one child in a family turns out to have had a different biological father, and in most cases it is of no legal consequence.
Simon Pringle, son of Steuart, did not contest the implications of the DNA results, but he contested their admissibility as evidence. The defence was put that his father’s DNA sample had been given for a particular purpose, researching the Pringle chief’s genetic family, but it was a breach of his right to confidentiality to then apply that DNA test result to the different purpose of questioning his right to his title. The court judged that confidentiality had not been breached in this case, since Sir Steuart had voluntarily given his sample to Murray, and that investigating the descent of the Clan Chief and holding a heritable title were both genealogical purposes. Sir Steuart should have been aware that information revealed by Y-DNA on the one subject would do the same on the other, so the court was unwilling to accept there was evidence ‘that Murray had obtained Sir Steuart Pringle’s DNA on a false premise.’
The court judged that the value of the evidence for resolving the rival claims superseded any concerns over whether confidentiality or data protection law had been breached. As a consequence of the Privy Council’s decision the baronetcy was awarded to Murray Pringle.
What does this all mean for you and others like you who may want to collect DNA samples for genealogical testing? There are both specific and general consequences. Specific consequences may be felt by anyone who would like to invite a holder of such a title to take a DNA test.
Approaching a titleholder for a DNA sample is not as unlikely as it may sound: many genealogists researching surnames are interested in the lines with the longest historically verifiable pedigrees and many of these may lead to an invitation to a living titleholder to test to establish the DNA of a ‘main line’. It is likely that some clan chiefs or the like may feel that taking a DNA test poses a risk to their interests. A genealogist who wanted to pursue this kind of research would need to offer full explanatory warnings to the potential test-taker of the risks and stringent safeguards for how they would be avoided, and that the data from their test would be used in the most confidential and specific manner.
This is partly an issue of informed consent, as discussed in the first part of this chapter, which is the general consequence arising out of the Pringle case. The court to some extent drew a veil over whether personal data collected for one stated purpose had been applied to another to the disadvantage of the donor. This could be seen as an example of where law and ethics part from each other. While the court took the view that the DNA evidence should be accepted because it existed, because of its quality to resolve the case, and because the way in which it had been obtained broke no law, it is clear that both Sir Steuart and Simon Pringle felt that their DNA data was not being used in the manner they believed it should.
This is an important lesson for all genetic genealogists who are spreading their attention beyond their immediate families towards asking their distant relatives or people they do not know to donate DNA samples. Whatever the legal rights and wrongs of the Pringle case, genealogists running DNA projects must be open and set clear limits around what they will do with the information they obtain from their donors’ DNA. If those purposes change, then they must return to their donors to receive clear, informed consent for any additional testing or new forms of analysis. If the donors are actively involved in the project, this is less of a risk, but many people are passive donors giving their DNA to genealogists to help with their research – and their willingness to contribute should be rewarded with the protection of fully committed ethical conduct from the researcher.
Having reviewed this case, now go back and look at the ethical principles for running a DNA project discussed on p16–20. The legal aspects were for the court to decide, but what do you feel about the ethical handling of the DNA sample submitted by Sir Steuart Pringle to the Pringle Surname Project?
Chain of custody
People new to taking ancestral DNA tests are often heard to ask questions like:
• Can my test be used as legal evidence (e.g. of paternity or connection with a crime)?
• Can law enforcement/police use my test results to link me to a crime?
The answer to the first one is a clear no. DNA sequencing undertaken for forensic purposes, such as part of a criminal investigation, is subject to strict chain of custody conditions. There are also strict standards governing how DNA is taken from suspects or witnesses to ensure the samples are clean and verifiably given by the person documented as the donor. Chain of custody implies two elements – the process that must be followed to guarantee the integrity of the sample; and the documentation of each stage that validates the evidence to be admissible in a court.
Chain of custody documents all the stages at which a piece of evidence is transferred between persons and places. This may include:
• Seizure of the evidence sample
• Custody of the sample
• Transfers – as few as possible
• Access – authorised and on strict basis of need
• Secure storage
• Analysis at an approved and licensed lab
• Disposition e.g. provision of accurate results to courts and counsel
Now pause to consider that test you administered to your great aunt or uncle in their home one Saturday over tea, biscuits or something stronger. What steps did you take to make sure the sample was theirs, and not contaminated by your DNA or DNA in the general environment? What official forms did you submit to document and witness the taking of the sample according to due process, and to document the legal standing of the agent to whom you turned Aunt Paula’s test over, and the legal standing of the lab they passed it on to in turn? Well, I am sure you were careful enough and the test will be perfectly good enough for your goals of comparing her DNA with yours for ancestral research. But what if a canny lawyer was to demand that you prove you did not slip a little kit ‘you’d prepared earlier’ into Paula’s test envelope while she was putting the kettle on, with the DNA of another person entirely inside? Why would you – of course. But in court, in some countries, people are sentenced to death on DNA evidence and miscarriages of justice are too common. So for this reason, your ancestral DNA test can never be used as formal legal evidence that can incriminate you or a family member.
The answer to the second question is also generally no – but the situation may be less clear, especially in jurisdictions with more open regulation over the searching of public databases. As DNA is shared to greater or lesser extents between family members, a technique known as familial searching can attempt to identify a suspect by looking for matches not with the suspect’s own DNA, but with the DNA of possible family members who may in turn match with a suspect’s sample. This will be discussed in the following Case Study on p26. It is important to remember though that DNA in ancestral databases, whether public or controlled by company Terms and Conditions, will never satisfy chain of custody rules to be admissible in court (the decision in the Pringle case was before a special kind of court, and the DNA results themselves had not been challenged by the defence). They can however provide leads to investigators, which can later be tested by a full legal process. In this way, genealogists need to be aware that chances are growing that family members could be traced from samples that they have put in ancestral databases. Set against this, it is also the case that miscarriages of justice can be set right and wrongly convicted people exonerated by use of the same methods. New technological applications of ancestral DNA databases are appearing fast, and the challenge for current times is to make sure this process is properly regulated and protected against misuse.
Forensic science has developed its own testing methods for DNA, including specialised testing kits (which in part inspired the kits now used for ancestral testing) and databases. Genealogists are often surprised by how few markers are tested in forensics. The standard test to identify someone uses fifteen autosomal STRs (recently expanded to over twenty), and Y tests use seventeen STRs (expanded in newer tests to twenty-three or twenty-seven, still far fewer than the thirty-seven plus genealogists are used to – and often still find ambiguous). Forensics aims to establish the probability that a DNA sequence in a crime scene sample is DNA left by a particular person, the suspect or accused. To do this, it is not enough to find that a haplotype is identical to the suspect’s. The forensic evidence must also be able to quantify the possibility that the two haplotypes could be identical by chance, and to do this, a number of forensic haplotype databases have been built.
We will come back to forensic databases in Chapter 10, but first let’s look at a major event in 2018 that may have turned the world of forensic searching upside down.
On 24 April 2018, a seventy-two-year-old man was arrested in Sacramento and charged with being a serial killer and rapist who had been active from 1974–86. Not caught at the time, he was given the label the ‘Golden State Killer’, or GSK. Californian police revealed he had been identified through a technique that is coming to be known as familial searching of genetic genealogical databases. Though this was not actually the first time it had been used, it was the first case of its kind to be made public, creating a media splash, drawing genetic genealogists’ attention to the technique and the general public’s to the existence of genealogical databases.
To Catch a Killer
There had already been searches made of Y chromosome databases in pursuit of unapprehended violent criminals – though many of these had led to flawed investigations and had been unsuccessful, leaving doubts about the technique. Because Y chromosomes across large kinship groups differ very little from each other, Y-DNA is not a sufficiently powerful discriminator to pinpoint suspects reliably.
However, the new methods used in spring 2018 to crack this landmark case are ground-breaking, and may change forensics and genetic genealogy fundamentally. In early 2018 a sample was found in one of the counties where the GSK had been active, stored in a freezer for thirty-seven years. It was found to be in very good condition and a genome-wide SNP chip test could be carried out successfully.
The autosomal DNA tests discussed in Chapter 4 use microarrays that hold a selection of SNPs from the autosomes (often referred to as SNP-chips). The standard chips used today have around 600,000-700,000 SNPs on them, a tiny sample of the full genome, but powerful for comparing genomes on a one-to-one basis for establishing identities and close family relationships. What the investigators did next was to create a profile from the test result that looked like one generated from the standard genetic genealogy companies’ SNP-chip tests. This generated DNA profile was then uploaded to the public genealogy sharing database, GEDmatch.com.
This profile matched with numerous distant relatives of the killer. There was no immediate revelation for the investigators, but they found themselves in a position now very familiar to that branch of genetic genealogy that does genealogical research on autosomal matches in unknown parent searches. In this case, they worked up to the great-great-great-grandparents of promising matches, reaching almost back to 1800, before beginning to work forwards again to search for leads among the other descendants of the shared ancestral couple.
Ultimately, a suspect was identified, surveillance was carried out and his DNA was taken from discarded items, which is legal under Californian law, to make a formal legal identification against the crime scene DNA. It is important to stress that the familial searching technique could not provide evidence that would be admissible in court, but is aimed at creating leads that can be tested by legal forensic methods with a full chain of custody. Once the suspect was identified, and his modern DNA taken (surreptitiously in this case), a legal identification could be made using the FBI’s forensic markers, and this is what a court eventually will see. It is not known how many suspects had ever been placed under surveillance by Californian law enforcement, though at least one further person is known to have been a suspect up as far as the stage of having his DNA tested.
The apprehension of the GSK suspect came hard on the heels of another celebrated solving of a cold case, the ‘Buckskin Girl’, an unidentified murder victim (see Chapter 10). In this case, whole genome sequencing (WGS) methods were applied to the sample and then the genealogy companies’ file formats were ‘spoofed’ and uploaded to GEDmatch.
The technical work in identifying the GSK and Buckskin Girl was certainly impressive, and will before long impact on genetic genealogy, where whole genome sequencing is starting to gain traction with testtakers as the prices fall dramatically. But the cases threw up some major ethical issues and at the time of writing they are still being discussed. There has been a visible impact on the availability of databases, with two ageing search platforms, Ysearch and Mitosearch, being closed down, mainly because of GDPR requirements, but also in the shadow of the fallout from the GSK. The biggest impact has been on the Terms and Conditions and Privacy Policies of the platforms used by genealogists to share DNA results.
It is hard to argue that something that led to the apprehension of a violent murderer is not a good thing, and anecdotal evidence from popular forums suggests that the majority of opinions from active users of the GEDmatch site were supportive. However, some voices have pointed to darker ethical issues that users should be aware of, in particular the question of where lines may be drawn. Does searching GEDmatch with criminal samples breach the privacy rights of the members? Is this a secondary use that members have not given informed consent to? It also throws the spotlight onto many activities that have become accepted by the genetic genealogical community, such as unknown parent searching. How can it be judged a breach of privacy to upload a profile that might lead to a serial killer being apprehended, while it is judged not a breach of privacy to confront birth or donor parents with their biological children? There have been cases in which the latter activity breached the ‘do no harm’ principle. On the other hand, if there should be a serial killer currently active, surely it would be a gross ethical breach not to use all means available to catch them, including familial searching of genetic genealogy databases without alerting the public that this was in process.
Set against the privacy and consent issues is the power of DNA testing to ‘do good’. The techniques used to find the GSK can also be used to exonerate wrongly accused people and set right miscarriages of justice. In November 2017 a Los Angeles man convicted of multiple murders was exonerated and released after thirty-nine years in jail when DNA evidence proved his innocence – as of spring 2018 the murders were being re-investigated within the GSK case. Statutory governmental databases have strict rules on who may search them and what kinds of searches may be permitted, to protect the individuals within them, largely offenders, who have lost the privacy rights of the wider public not to have their DNA stored. As genetic genealogical databases are non-statutory they can only be protected by the actions of the companies and organisations which created them.
As with other aspects of genetic genealogy discussed in this chapter, you should always read carefully the Terms of Service (ToS) given by any platform you wish to use before uploading your data, and make sure you are comfortable with them. If you are uploading data of family members or other people with their permission, make sure the consent they give is fully informed of the risks as you see them – or let them read the ToS themselves to make their own decision. In May 2018, GEDmatch produced a new set of ToS to address these issues, while still committing themselves to open sharing and to have users take responsibility for their understanding of the risks. (Other testing companies have also been clarifying their ToS under the twin shadows of GDPR and the GSK case). The new ToS or Site Policy includes (with our emphases):
When you upload Raw Data to GEDmatch, you agree that the Raw Data is one of the following:
• Your DNA;
• DNA of a person for whom you are a legal guardian;
• DNA of a person who has granted you specific authorization to upload their DNA to GEDmatch;
• DNA of a person known by you to be deceased;
• DNA obtained and authorized by law enforcement to either: (1) identify a perpetrator of a violent crime against another individual; or (2) identify remains of a deceased individual;
• An artificial DNA kit (if and only if: (1) it is intended for research purposes; and (2) it is not used to identify anyone in the GEDmatch database); or
• DNA obtained from an artifact (if and only if: (1) you have a reasonable belief that the Raw Data is DNA from a previous owner or user of the artifact rather than from a living individual; and (2) that previous owner or user of the artifact is known to you to be deceased).
‘Violent crime’ is defined as homicide or sexual assault.
By registering for GEDmatch and using the Site, you agree that you will not upload Raw Data that does not satisfy one of these categories. If you have previously uploaded Raw Data that does not satisfy one of these categories, you hereby agree that you will remove it immediately.
The responsibility remains with the user to obtain the necessary consents to upload, though the requirements are now spelled out more clearly. GEDmatch has also addressed a major ethical conundrum: if some law enforcement uses are acceptable, then which such uses are they? They have defined this as violent crime and given a definition of that. By definition then, using GEDmatch for solving other crimes such as drug offences, theft or motor offences is not to be permitted. What powers GEDmatch has to enforce this remains to be tested, but as of 2019 an uneasy balance exists between the desire of genetic genealogists to share their data openly and mutually and their need to control who has access to that data and for which purposes.
In 2013 a group of active, mainly American, genetic genealogists selforganised into a committee to draft a set of standards for the field, to answer accusations that it was unregulated and open to poor practice. The group were keen to make clear that they did not represent any company, organisation or interest group linked to the genetic genealogical field, and the Standards were an initial attempt to agree a consensus set of guidelines on good practice. After an open period of consultation the Standards were launched in January 2015 at a colloquium in Salt Lake City, and published online for interested parties to refer to.
The set of twenty-one Standards, which can be viewed at http://www.geneticgenealogystandards.com, addresses many of the concerns dealt with in this chapter, and more. While not official, they are being followed by many practitioners providing professional services and can be consulted by anyone with an interest in ancestral DNA testing, from new test-takers to project administrators and professional genealogists who work with DNA testing, to the companies that sell the testing and analytics. They cover the two broad areas within the activity – eleven standards for the testing process and a further ten for the interpretation process – creating benchmarks to judge best practice.
Several standards cover the areas we have examined in this chapter, including rights to privacy, confidentiality, ownership and for informed consent. The right to anonymity is expressed, but warning is made that this can be curtailed or compromised once results are published, which should be an informed decision of the owner of the DNA test results alone. Genealogists offering testing or interpretation services are expected to be knowledgeable so that information and advice offered is accurate and up to date, and the limits of what can be learnt or proven from DNA testing are understood. There are also useful standards on storage of samples, appointing beneficiaries to guard against kits being orphaned by decease of the test-taker and the use of results in scholarship and educational activities.
Standards will continue to evolve as the tendency for rapid change of the technology creates new uses and situations – and the years 2018 and 2019 delivered a number of shocks to the perceived status quo of ideas within the field. But the twenty-one Standards make the clearest statement so far of an ethical approach to tracing ancestral relations with DNA, and all users are recommended to make use of them.