Thursday, April 14, 2011

Worth The Cost?


I've been arguing with myself for a couple of years now over doing a DNA test for ancestral research purposes. I just got another ad for DNA testing -- this one was from the National Geographic Society. Ancestry.com and others have been pushing it for a long time now.

But I keep hesitating.

I'll admit that part of my hesitation is that I just don't understand that much about it. I've read all the ads they've sent me, researched it on several different websites, read the Wikipedia article, ... and either there's something I'm missing or I just don't think I'll get enough out of it. Either that, or they are all really bad at explaining the benefits of DNA testing.

Don't get me wrong, they are superb at hyping the virtues of DNA testing -- how you'll learn where your ancestors came from, find long-lost ancestors and relatives, etc. -- but when you read the details it sure doesn't sound like that is very likely. But, like I said, maybe there's something I'm fundamentally missing about genealogical DNA testing, so I'm going to describe what I currently understand and hope that, if I'm wrong about it, someone more knowledgeable will come along and correct me.

There appear to be two types of DNA test: the Y-chromosome test (Y-DNA) and the mitochondrial DNA test (MtDNA). Or, of course, you can buy a package deal that includes both. It looks like anyone can have a MtDNA test done, but only males can do a Y-DNA test since females do not receive a Y-chromosome. Y-DNA tests can be done in a wide range of levels of precision. The most basic Y-DNA tests look at only about 10 or 12 markers, more thorough tests examine 33, 46, or 67 markers, and there are a few vendors offering a 111-marker test. The more markers the test includes, the more accurately it will identify others who share your family line. MtDNA tests fall into three categories: those that test only the HVR1 sequence, those that test the HVR1 and HVR2 sequences, and those that examine the the entire MtDNA sequence including the "coding sequence". The MtDNA test identifies only the direct matrilineal descent -- your mother's mother's mother's mother's mother's ... and so on ... mother.

So the first thing that jumps off the page at me is that I'm plunking down all this money just to take a peek at two of my family lines: the exclusively patrilineal descent of my father, and the exclusively matrilineal descent of my mother. In other words, the Y-DNA test will display only the genetic information for my male Pellman ancestors and whatever they may have been called before taking the name Pellman, and the MtDNA test will look at the zig-zagging path through all the woman of whatever surname are in my mother's direct ancestry. So it will look at the sequence from Mom to Winnie Kerr to Maria Groves to Martha Baxter to whoever her mother was, and so on.

Now, at first blush both of those seem promising avenues of pursuit. So far I've only been able to trace my direct paternal lineage back to my great-great-grandfather, William Henry Pellman and my mother's matrilineal descent to my great-great-grandmother, Martha Baxter. So DNA testing offers the promise of a huge breakthrough on both fronts!!! But how?! And that's where I begin to question the benefits of the testing.

DNA cannot identify specific ancestors. It can only assign a probability -- based on the number of identical DNA markers -- that two or more people share a common ancestor within the last several generations. It cannot even identify the specific MRCA (most recent common ancestor) with certainty; only with a stated degree of probability. Worse yet, the only way it can do that is if both individuals are in the DNA database and both individuals have consented to the release of their information. This raises my big question: how likely is that that someone would have done genealogical DNA testing and consented to the release of the test results, but not bothered to post their ancestral information on one of the dozens of searchable genealogical databases or the World Family Tree? If they are interested enough to pay a sizable sum of money for DNA testing and willing to share those results publicly wouldn't they have been interested enough to share what they already knew about their own ancestry? If so, I would have already found them on some search engine and made the connection!

Unless I'm missing something crucial about how DNA test results are compared and shared, the odds of my learning something new about specific ancestors are extremely low.

So, absent a one-in-a-million discovery that leads to identifying one of two possible great-great-great-grandparents, I stand to gain little or no information that would identify a specific ancestor I haven't already found. And the DNA tests do nothing to identify members of other lines in my family tree -- of which I have already discovered more than 500 different surnames. So, unless those other lines somehow don't matter as much as the two that I can trace through DNA, it's usefulness is extremely limited.

The other thing the DNA testing companies tout is that they identify haplotypes and haplogroups. Whoopee! A haplotype is just a group of people who share a common set of genetic similarities. A haplogroup is a larger group of people who share a broader set of genetic commonalities. Often these haplogroups seem to have lived in the same geographic area for a long time, so it means that you might be able to determine that your ancient ancestors were from a particular geographnic region ... Well, not all of your ancient ancestors, but the ancient ancestors who are from just two of your thousands of ancestral lines. So, if my maternal grandfather was of African descent, for example, neither the Y-DNA nor MtDNA test would reveal it! Am I missing something, or is that not really much help in the greater scheme of things?

The high-precision DNA tests run $500 to $1,000 for both Y-DNA and MtDNA. There are low-precision tests available for $150 to $300, which examine only a dozen or so Y-DNA markers and just the HVR1 sequence, but I can't justify either expense for what little either is likely to reveal with any certainty. So, for the moment, I'm waiting until the price comes down significantly.

If I've overlooked anything important in my analysis, I hope my readers will set me straight, so please feel free to comment if you think I'm missing something.