“Consult a professional” is advice you hear in all realms of life when your questions outweigh your ability to find answers. But I’m a professional genealogist, and I don’t have all the answers — I could build a fortress with all the brick walls in my research. My dad’s been waiting decades for me to figure out his family tree, but patient research hasn’t gotten me very far. He’s 84 now and he keeps asking for more.
Every time I flip through a magazine, read the newspaper or turn on the television, I’m confronted with genealogy’s hottest new problem solver: DNA. The stuff that solves crimes on “CSI” and proves paternity on “Maury” also can link you to ancestral roots and present-day cousins. And it looks easy — after all, talk show host Oprah Winfrey did it, and so did comedian Chris Tucker for the PBS series “African American Lives” <www.pbs.org/wnet/aalives>. But I’ve never been much of a scientist, so I’m a bit intimidated by the multitudes of tests, the jargon-littered literature and the ongoing debate over what DNA really reveals about genealogy. Can DNA help this frustrated family historian find answers?
To find out, I consulted with several genetic genealogy companies and ordered a variety of DNA tests (provided gratis for this article). Don’t worry — if you take the DNA plunge, you’ll only need to shell out for a test or two. But my experiences (and mistakes) will help you maneuver through the multitude of options. So join me on an exciting excursion into the world of genetics and family history.
Preparing for the journey
It’s not like I’ve been living in the Stone Age — I’m fascinated by any new technique or tool that promises to help me trace family. At a conference in 2001, I gave a blood sample for a Brigham Young University genetic genealogy study (the Sorenson Molecular Genealogy Foundation <www.smgf.org> is now running that project). I was intrigued, but unsure where this DNA stuff was going. Later, I became more curious when I heard Bryan Sykes talk about his book The Seven Daughters of Eve (W.W. Norton & Co.), and the science that inspired it. Sykes started UK-based Oxford Ancestors (<www.oxfordancestors.com>, +44 186 5847594) in 2000 to help people learn about their ancient ancestry through mitochondrial DNA (mtDNA), which is passed down from mother to child. That means mtDNA tests tell you only about the part of your ancestry that comes from your mother’s mother and her mother, and so on.
I found genetic genealogy hard to sort out even after Sykes’ talk. For one thing, the lingo is confusing. And what about all those tests? So I started reading articles such as “Demystifying DNA” by Megan Smolenyak Smolenyak in the February 2005 Family Tree Magazine. Smolenyak, who has an informational Web site called Genetealogy <www.genetealogy.com>, is pretty good at translating what can look like scientific mumbo jumbo into layman’s language. I also read the book she co-authored with Ann Turner, Trace Your Roots With DNA (Rodale).
I learned that defining a problem to solve is the best route to genetic genealogy success. So I spell out two things I want to know: First, do I have American Indian connections? They’re rumored on both sides of my family, but my mom’s relatives talk about it most. This is a common oral tradition among French Canadians, and one that can be hard to verify in records. One of my mother’s brothers supposedly found the link by contacting a tribe in Canada, but he died soon after; now the documentation is lost.
A biogeographical test such as DNA-Print Genomics’ (<dnaprint.com>, 941-366-3400) AncestryByDNA can estimate my percentage of Native American heritage as well as the other major anthropological groups: European, Sub-Saharan African and East Asian. Biogeographical tests evaluate autosomal DNA, which makes up all your genetic material except for what’s on the XX and XY chromosomes, says DNA-Print Genomics scientist Lou Charlton.
Second, I’m hoping genetic genealogy can tell me more about my paternal line by identifying relatives who might have new ancestral information. That’s where Y-DNA, which is passed from father to son, comes in. Tests assign values to DNA markers, and the more markers a test evaluates, the more accurately it predicts when you and a match share an ancestor. Since Y-DNA mutates over time, the more markers you match, the closer the probable relationship. The most markers you can test as of this writing is 67, available for $349 at Family Tree DNA. Relative Genetics has a 43-marker test for $195; DNA Heritage has one for $199. (DNA Heritage also lets you choose which markers you want to test, so you can pick the same ones, say, that a suspected cousin had tested.) Most DNA labs facilitate — and offer discounts for participants of — surname studies: groups of people with the same or similar last name who hope to uncover family ties.
Since women don’t have Y-DNA, a couple of problems arise. One, I’ll need to have a close male relative, namely my dad or brother, contribute DNA samples. (Anyone related through paternal lines, such as a father’s brother’s son, would work, too.) Also, Y-DNA tests will help find people related to me only through those paternal lines. Since my dad’s mother didn’t contribute to his stash of Y-DNA, his test won’t tell me anything about her branches of our pedigree chart.
Now that I have my goals and a better grasp of genetic genealogy, it’s time to order some tests. Companies offer various types, and their Y-DNA tests use different DNA markers, so I study the Web sites of labs mentioned in Family Tree Magazine and Trace Your Roots With DNA. To investigate the American Indian story, I request an AncestryByDNA test from DNA Print Genomics for myself and my brother. Charlton says the results of this test often differ between siblings (except for identical twins) because genetic markers from mothers and fathers combine randomly.
I’ve already ordered a 26-marker Y-chromosome test from Relative Genetics (<relativegenetics.com>, 800-956-9362) as my first baby step into DNA testing. On the lookout for companies with Taylor surname studies, I find Family Tree DNA (<familytreedna.com>, 713-868-1438) and request a 12-marker Y-DNA test. Again, you’ll likely need just one test, but for this article, I also decide on a 17-marker test from Chromosomal Laboratories (<chromosomallabs.com>, 877-434-0292). I ask for haplogroup analysis with all the tests (see glossary, page 25).
I also checked into DNA Heritage (<dnaheritage.com>, 866-736-2362), whose founder, Alastair Greenshields, recommended getting samples from two male cousins — he encourages Y-DNA testing within a group so you can compare results, rather than hoping you’ll find a match in a database. Since I’m not acquainted with any cousins on my dad’s side, though, I pass.
Each kit arrives with the same tools — a swab or two to painlessly scrape cells from inside the cheek, a return envelope and a consent form. My father good-naturedly scrapes the inside of his mouth for Family Tree DNA and Relative Genetics; my brother does it for Chromosomal Laboratories and DNA Print Genomics. Dad wants to know what will happen to his DNA samples after the test — do they languish in a lab where mixups could happen? Could an unqualified person get his hands on them? I consult a Y-DNA lab comparison chart on the International Society of Genetic Genealogy Web site <isogg.org> (click Resources). Most labs save samples in case you order more tests, but they’ll destroy them upon request.
When I interviewed Bryan Sykes about his book, he recommended I take the Oxford Ancestors MatriLine test, which can use mtDNA to tell me which “daughter of Eve” (akin to a maternal branch of the world family tree) I belong to. Since I have English ancestors on my paternal side, Sykes also suggested my father try the Y-Clan and Tribes of Britain tests — the former correlates Y-DNA with a haplogroup; the latter determines whether your ancestors were from the Celtic, Anglo-Saxon or another group that settled Britain. When Dad hears I need another sample, he just opens his mouth and doesn’t say a word. I think he’s starting to feel like a lab rat.
The envelope, please
Waiting is the hardest part of this whole process: Tests can take up to eight long weeks. Finally, results begin dribbling in, accompanied by a variety of explanatory literature. Since science isn’t my natural gift (did I mention that already?), I go over this technical material several times in an attempt to understand it. To show what you might expect to get, here’s a test-by-test breakdown:
• Relative Genetics: I get an e-mail with easy instructions for accessing my dad’s Y-marker results and projected haplogroup online. A click and a password, and I’m in. A few days later, a report arrives with a sheet that explains his results. Several people in Relative Genetics’ database match some — but not all — of my dad’s markers. I surf to the online learning center’s refreshingly helpful Guide to Understanding Your Matches, where I find a low “relationship rating” for anything less than a 23-out-of-26 match. That means these individuals may be related, but not within a genealogically helpful time frame. I’m discouraged. When I call my Dad to report that his markers appear in a few Scottish clan projects, he’s not excited. As I bide my time for more test results, I plug the values for my dad’s 26 markers into a series of public DNA databases (see the list above). No exact matches in any of them (sigh).
• Family Tree DNA: An e-mail announces that 26 (!) people in this company’s private database match all 12 markers in my dad’s test. I’m excited — until I read on: “If you share the same surname or variant, this means that there is a 99.9 percent likelihood that you share a common ancestor in a genealogical time frame. If you match another person without the same surname or variant, you still probably share a common ancestor, but this ancestor most likely lived in the time before surnames were adopted.” I click on the link in the e-mail message and discover none of the matches has the surname Taylor. But I wonder about nonpaternity events — suppose the person is adopted or his ancestors changed their last name when they immigrated? I check with Family Tree DNA vice president Max Blankfeld, who says nonpaternity might cause you to match a person whose last name is completely different from yours.
I study Family Tree DNA’s online database of pre-1900 family trees (for privacy, it doesn’t include information on people who might be alive) that some matches have provided along with their e-mail addresses, but I don’t find anyone to research. The fewer markers you test, the more likely you are to get “false positives.” Had I understood that better, I would’ve had more markers tested to start with. But if I think I’m related to a 12-marker match and can’t prove it through traditional research, I can buy an upgrade to a higher-marker test and see if the match holds up.
My dad’s test results are on a password-protected Web page. I click a tab for Recent Ethnic Origins and notice all but one matching person lives in England, Ireland or Wales. Most are from Ireland. That confirms family stories of Irish roots, which census records back up. I also can use the results page to automatically fill in a search form for Ysearch, Family Tree DNA’s free public Web database.
• Chromosomal Laboratories: Once only a forensic and paternity testing lab, this company began offering genetic genealogy tests within the past year or so. The results contain certificates listing my brother’s haplotype and haplogroup, easy-to-understand reports on interpreting the results, and a “Country of the Closest Matching Ancestor” chart. My brother’s haplogroup, R1b, matches my Dad’s. According to the accompanying literature, this haplogroup is common in Western Europeans — it includes 70 percent of people in southern England, 90 percent of Ireland’s residents and 90 percent of Spaniards. As Smolenyak explains, R1b is “the genetic equivalent of being a Smith.” R1bs descend from Cro-Magnons, characterized by broad faces and tall stature. Interesting, but is my haplogroup relevant to my family tree? Not really — it doesn’t help me make a connection to living cousins.
More valuable are the genetic markers I can use to search databases. Chromosomal Laboratories also searched three public Y-DNA databases — Ysearch, the Sorenson Molecular Genealogy Foundation and YHRD (this site is geared toward scientists rather than laypeople) — and included a chart showing the matches. Several people from Ireland, England and Tennessee matched my brother on multiple markers, but not on all 17, and none of them are named Taylor. Chromosomal Laboratories CEO Vladimir Bolin tells me his customers can get “a flavor of what can be accomplished” with his company’s Y-DNA test, but other labs test more markers. The moral of the story: A low number of markers is plenty to see if I’m related to someone I think is a cousin, but a higher-marker test would be more useful for searching online databases.
• DNAPrint Genomics: A package arrives with two pieces — one is a fairly technical, hard-to-understand “user manual” for interpreting the AncestryByDNA test results; the other is my actual results report. I skip to the latter because it’s easier to read. My brother and I have different percentages of three different groups — European, Native American and Sub-Saharan African. I expected the first two, but that last one was a big surprise. I call Lou Charlton, who says, “No one knows how far back these tests go.” In other words, there’s no telling when African heritage entered our family tree — it could be eons ago, when the Romans and their African slaves were wandering all over Europe.
Confidence ranges are another consideration. The ethnic percentages on my results certificate are most likely estimates (called MLEs). Charts in my AncestryByDNA results kit show other possible combinations: My African and Native American confidence ranges go up to 10 and 16 percent, respectively, and down to zero, so maybe I don’t actually have either ancestry. Smolenyak says it’s common for tests to confuse Native American and Asian heritage.
Since I showed mostly European ancestry, I order a EuroDNA test, which further breaks that heritage into Northern European, Middle Eastern, Southeastern European and South Asian. My European ancestors are mostly from Northern Europe, with Middle Eastern coming in second. Genetics can tell me what broad region a small section of my tree comes from, but because DNA predates national borders, that’s as specific as it gets.
• Oxford Ancestors: Since this was the last DNA test I sent in, I’m still waiting for results. I’m planning to read more about my “daughter of Eve” in Sykes’ book and tell my Dad about his tribal roots in Britain. I hope that will keep him busy for a while. Though I like the idea of knowing this information, again, it’s outside a genealogical time frame. But Oxford Ancestors has introduced customers-only mtDNA and Y-DNA databases, so I’ll be able to use my MatriLine results to search for people who match my mtDNA sequences.
Though it’s been thrilling to discover my genetic past, I still don’t have all the answers. I do, however, have some direction: For example, I’ll research my American Indian heritage knowing I’m probably not on a wild goose chase.
Of course, Y-DNA testing hasn’t led me to any previously unknown relatives — yet. I can plug my dad’s and brother’s genetic markers into DNA results databases and keep searching for matches. After all, people are adding to these databases every day. I’m planning to have my 26-marker Relative Genetics test upgraded to 43 markers for more-accurate Y-database matches. And I’m looking into Taylor surname studies.
Meanwhile, I’m still working on extending my family tree through traditional genealogical research. That way, I’ll have the documentation to prove a relationship when a “DNA cousin” contacts me, or vice versa. Sykes agrees with this dual approach. “Genetics is best used to back up well-researched genealogy,” he tells me.
He predicts this will be even easier for researchers in the future, as costs come down and companies start testing more DNA markers. Researchers themselves are propelling such advances through their intense interest in the field. “Genealogists should congratulate themselves for catching on so quickly,” Sykes says.
I think one day genealogists at conferences will wear badges bearing their genetic markers in addition to lists of family surnames. I still don’t consider myself a scientific genius, but after all the homework I’ve done, I’m ready for that day. My dad will be, too, as soon as he grows a new supply of cheek cells.
In a Nutshell
Before you speed dial the nearest genetic genealogy lab, review the lessons I learned during my DNA adventure:
• Y-DNA and mtDNA tests tell you about a small section of your family tree, not all of your ancestors.
• Err on the side of more markers for Y-DNA tests to reduce the number of false positive matches.
• Biogeographical tests provide the broad region — not the specific country — your ancient ancestors came from.
• Knowing which haplogroup and “maternal clan” you belong to is interesting, but it doesn’t tell you about your traceable ancestors.
• Understand the confidence ranges for your test results before making assumptions about your family tree.
Each DNA testing company told me whether my dad or brother matched any of its other customers. Some also compared results to public databases — including those listed here. Although labs often test different markers, these free databases help you “convert” your test results so you can compare them to others in the database.
• mitosearch <www.mitosearch.org>: This Family Tree DNA site has the only public database of mtDNA test results as of this writing.
• Sorenson Molecular Genealogy Foundation <www.smgf.org>: You can contribute a sample and pedigree chart to this study — but you don’t get results. Search the database of study participants by marker values or a surname.
• Ybase <www.ybase.org>: Though DNA Heritage runs this site, you can enter up to 49 Y-DNA markers from any lab’s results.
• YHRD <www.yhrd.org>: This site’s technical language makes it primarily for the scientific community, but if you’re game, click Search Database.
• Y-Match <www.relativegenetics.com> (click Search): After registering, search this Relative Genetics database by haplotype or surname.
Except for the charts showing DNA markers and their values — or, for my AncestryByDNA test, my ethnic ancestry breakdown — the reports I got back from testing labs were vastly different (and for the most part, available online). Each company also sent me varying amounts of background information with assorted illustrations. Much of this material explains DNA analysis, the lab’s qualifications and the uncertainty that’s a given in genetic testing — not as interesting as your marker values, but worth a read, especially if you get surprising results.
Without a little help, some of the charts in my DNA test results reports made my head spin. Here are two basic charts and what they mean:
Family Tree DNA Most Recent Common Ancestor Estimates
This graph represents the chances that a genetic match and I share an ancestor within a given number of generations. For example, if I match someone on 12 alleles out of 12 tested (the black line), there’s an 80 percent chance we share an ancestor within the past 15 generations.
AncestryByDNA Genetic Ancestry Breakdown
I’m a genealogist, not a lab scientist, but learning these terms helped me understand the people in the white coats:
• allele result: also called a marker value, the numeric value assigned to a genetic marker
• autosomal DNA: all your DNA except what’s on the X and Y chromosomes
• biogeographical test: an autosomal DNA test that estimates percentages of your geographic origins
• haplogroup: an identification of the genetic group your ancient ancestors (10,000 to 60,000 years ago) belonged to
• haplotype: collectively, the marker values on your Y-DNA test results
• mitochondrial DNA (mtDNA): genetic material that mothers pass on to their children