Michael Cooley's Genetic Genealogy Blog GEN • GEN
11 September 2018

Summary of Recent Akins Big Y Testing

While maintaining the results pages at a dozen DNA projects at FTDNA, including Akins, I've been off getting a BA in history (2013 at age 63) and an MA in English and Writing (2017 at 67). After graduating last December, I've slowly played catchup on the projects, writing eighteen articles for this blog. It's finally Akins' turn to get some attention.

In fact, the project has done just fine in my absence as there's been considerable Big Y activity over the last five months. I had planned to go through each group in turn starting with recent results from Group AF04. Instead, I've churned out SNP trees for six groups. Because I'm not an Akins genealogist (I'm descended from Group AF05, by the way), I'm presently looking at these strictly from a genetic viewpoint. Below, I describe important key concepts while using the trees as examples. I hope it proves to be helpful.

Genetic Inheritance

We learn by the time we're in high school, and often earlier, that children inherit half of their genetic makeup from each parent. In the simplest terms, the biological mechanics is straight forward. Each of the 23 chromosomes found in the nucleus of every cell comes in pairs. We have a #8 chromosome, for example, that came from pa, and a copy that came from ma. They're mostly identical to one another but some variance is found. It's these small differences that makes us individual rather than clones.

Likewise, half of our parents' genetic code is made up of half their parents' genetics — the half we get from each parent is itself half-and-half. We end up with roughly one-quarter from each grandparent, an eighth from each great-grandparent, and so forth. Indeed, most Americans grow up being familiar with something of their mixed national heritage. One person might be half Italian, a quarter French, and a quarter "I dunno." Some of us can parse it down into eighths and sixteenths. For most of us, I imagine, this concept is a no-brainer.

The Sex Chromosomes

But 3rd grade arithmetic works for only the first 22 pairs of chromosomes, called autosomes. FTDNA's Family Finder product examines autosomes, and it's now the only test provided by Ancestry.com. Such tests are great for finding cousins (generally to about the 3rd or 4th degree) and for determining ethnic ancestry. Companies such as 23andme can also provide health reports. The TYR gene, for example, which helps determine eye color (it's not the whole story), is found on both the ma and pa versions of chromosome #11. The readings from each is factored into predictions for eye color. And so it is with roughly the twenty thousand genes we possess.

The 23rd pair, however, is rather roguish. The SRY gene, which determines maleness, is found only on the pa version. This fact probably has a lot to do with the separate evolutionary paths the 23rd pair, the XY chromosomes, took. Through the eons, the Y has been whittled down to a mere 58 million bases of the four genetic letters (A,C,T, and G), about a third of what the X has, and a tiny fraction of the 3 billion pairs of bases found in the human genome.

It all comes down to which of the father's sex chromosomes we were handed. If his Y, we're male; if his X, we're female. (Note that although the Y is male, the X is not female. Both sexes have it. It's the lack of the SRY that determines femaleness.)

Y Chromosomal Inheritance

It might well be true that if we go back six or seven thousand years, the ancestor of any one of us is the ancestor of all of us. But within this massive ancestral maze of tribal movements and familial intermarriage, we each have a single, direct paternal lineage: our father's father's father's father, going straight back as we can possibly go. Thanks to the unique inheritance pattern of the Y chromosome we can track the lineage — what the father has, the sons get. And thanks to the manner in which our society passes surnames, we can often give names and faces to several generations of paternal ancestry. Because of these biological and social oddities, I not only know the name of my 7th-great grandfather, I know he had the mutations YP4491, YP4492, and YP4493 in his Y chromosomes. All seven, eight, maybe nine sons had them, and they passed down to all their sons up to the present day.

Most Recent Common Ancestor (MRCA)

This is a genealogical term as well as a genetic one. The Most Recent Common Ancestors (MRCAs or TMRCAs) for two siblings are their parents; for first cousins, their grand-parents; and for second cousins, their great-grandparents. Relatives share not only an MRCA, but share all generations that came before it. We see an Akens MRCA represented in AF02 by the mutation BY9400. Not only is this marker ancestral to the testers, but so are the "grandparent" markers. Tester #N115713 in AF08 was born not with only the two Y-SNP mutations shown above him, but all those listed above — and more. Mutation A14574 represents an MRCA for both himself and the Edgar tester. The group of mutations listed just above it are shared with AF08 Akins, Edgar, and McLeods, and the next three upstream mutations (A10097, etc) could conceivably be found in the remains of a man (if ever dug up) who lived about 1100 years ago — right in the middle of the Viking invasions of the Britain Isles. It's no surprise, then, that those earlier mutations are still found among Scandinavians.

Single Nucleotide Polymorphisms (SNPs)

Those of you familiar with the Akins DNA Project are somewhat familiar with the results page. The numbers there represent Short Tandem Repeats (STRs) — the number of times a string of genetic letters (ATT at DYS388, for example) repeats. For AF08, there are 14 repeats at that position; for AF01, 13 repeats; and for most of the others there are 12 repeats. The problem with STRs, however, is that the number of repeats can go either up or down in succeeding generations. This obscures the probable genetic configuration of an MRCA. We can trust STRs to an extent, but they should never be given the keys to our cars. They're fickle and are apt to change course.

Not so with SNPs. They're single point mutations, an alteration from one of the four genetic letters to another. Tester #B185439 of AF04, for example, was born with an A rather then the usual C at position 19905301 of the Y chromosome. We don't know at what generation it emerged, but we do know that tester #B11295 doesn't have it. In other words, BY172259 defines a separate lineage. (Note that new results are pouring in for AF04 as I edit this. Updates will soon be needed.)

SNPs are far more stable than STRs and can persist from one generation to the next for tens of thousands of years. YFull.com, for example, estimates that BY2707, at the top of AF03, is about 4000 years old. My own L448, nicknamed "Young Scandinavian," is estimated be about 3000 years old. Make no mistake about it. There was a man born in Scandinavia, perhaps near present-day Bergen, Norway, who was the first to be born with that mutation on his Y chromosome. He passed it on to his sons and so forth. YFull lists 166 testers from Norway, Sweden, Ireland, Great Britain, and the United States who are in possession of that mutation (not including testers with other companies). In each case,they received this mutation down one of the multiple branches of the paternal line from a man (we can all him Mr. L448) who lived on the eastern shore of the North Sea a thousand years before Christ was born.

Placing SNPs on a Timeline

By looking at the trees and understanding MRCA, we can see how and why these SNPs are arranged hierarchically. For Group AF02, Akens, Kidd, Kelly, and Starke all share the Z36750 group of SNPs, but only the Akens pair share BY9400. Z36750 has a larger population than BY9400 and must, therefore, be older. We can take it further by looking at M269, a defining marker for an ancient R1b subclade. The SNP is found among 110 million male Europeans. About 92% of the male population in Wales has it. We can be sure, then, it's of great antiquity — YFull estimates it's between six and thirteen thousand years old. In contrast, my R1a-YP4491 is probably shared by only several dozen living men and may be only about 400 years old. The story will turn out similarly with AF02's BY9400.

By counting and averaging the number of SNPs in each lineage, YFull has learned that SNP mutations occur on average about one every 144 years (plus other factors). Let's take the simplified approach. The two Akens testers in AF02 each have three SNPs in their lineage from the MRCA, the aforementioned BY9400. If we count all SNPs downstream of it (six), divide by the number of testers (2) and multiply by 144, we can estimate that a fellow, probably of the name Akens, was born about 432 years ago. But this uses only the two present samples. Some MRCAs have dozens, hundreds, and even thousands of living representatives. The more testers, the more accurate the estimate will become.

Group AF04 has three testers descended from BY47217 and a total of 18 downstream SNPs. Dividing by three and multiplying it out, we have an estimate of 864 years for the birth of the man first in possession of BY47217. That might well be true, but only one has SNP emerged in tester #B185439's line in the same time it took #B11295's lineage to acquire seven. Obviously, 144 years is merely an average and says nothing about any single lineage. We can play with these few AF04 numbers and come up with something that sounds a little more reasonable, but AF04 has five Big Y's in the pipe. We should have better estimates by the end of the year. (Two have come in since this last edit, which already alters these figures.)


Haplogroups are a collection of SNPs that are known to exist in the same block, or at the same level, as other SNPs. Older haplogroups can have hundreds of SNPs in a single block simply because each tested descendant is shown to have all of them. For example, the major haplogroup I1, which AF08 belongs to, has 309 SNPs (according to YFull's data — including the defining SNP M253 and 308 more). According to Eupedia.com, I1 split from haplogroup I about 27,000 years ago. Haplogroup I population geneticists might have an explanation for the lack of genetic divergence, but this split happened a very long time ago. We might imagine there's a large population that simply hasn't yet tested, but perhaps a sizable lineage died in the icy-cold tundra of the far north of Scandinavia. Or perhaps they never got there in the first place. I don't know. But it would take only a single test, say from someone in a remote Saami village, to cause a split.

Population geneticists and their more populous and less-educated spinoffs, genetic genealogists, are looking for these splits — these previously unidentified branches of the Y-DNA tree. New tests bring them to light. Each time a haplogroup splits, we have a new haplogroup or subclade. For example, an early Big Y tester from one of the Cooley groups was found to have 24 novel SNPs, mutations that had not yet been seen in other testers. Subsequent testing isolated just two of those SNPs as belonging to a man who immigrated to New England in the 1630s. Any male who rightfully claims paternal descent from Benjamin Cooley will have those markers and be of haplogroup R1b-A12022.

Likewise, brand new test results for #631138 (AF06) also reveals 24 novel SNPs. Four of them, it turns out, are found in volatile regions of the Y and should probably not be seriously considered, but noted. (We want SNPs that consistently show up in descendant lineages.) Unlike AF04, however, we the present results say little about the Akins AF06 MRCA. (In order for an MRCA or a haplogroup to emerge, we need at least two matching testers.) For now, this old man of yore is buried inside those SNPs. Additional advanced testing, however, will reveal a significant percentage that belongs to an upstream haplogroup, one that consists of several families and of several surnames, as seen on the right side of some of these trees. When that happens, the small number of leftover SNPs will prove to have originated among the tester's more immediate ancestors.

Geography and SNPs

So, we can take any tester and arrange their SNPs in a vertical lineage, from the oldest at the top to the most recent. Each shift of a single genetic letter was born along with a single man who, in the vast majority of cases, will never be identified. But an ID can occasionally be made, just as happened with the above Benjamin Cooley. For example, by comparing the SNP results for the Strothers of Group 01, we know that Francis Strother was born in Virginia about 1709 with the SNP mutation A20343. We have not only a name, but era and geography — Virginia. Because I1's presence in Britain is best explained by the Vikings (AF08), as is true with my own R1a (there were also R1b Vikings and smatterings of other haplogroups), we can find a range of SNPs in the lineage that likely crossed the North Sea. For example, in earlier articles, I took my SNP lineage back from the US to England, to Scotland, to Norway, and beyond. At each stopover, new SNPs emerged in the lineage. We can observe a similar pattern in AF08. In time, given enough data, #B11295 will be able to track each of his SNPs through both time and place. In this same way, each of us stand to learn something that is substantially real about our deep heritage and can observe our place in the human pageant across the face of the globe.

Last Points

I mention YFull frequently. Anyone can submit their BAM file — a huge binary file that records all the raw data — and have a complete analysis done for $49. Their database is far more detailed than FTDNA's, every discovered SNPs and kit ID is listed on their public tree. Of course, not all FTDNA customers contribute, so there are some holes in the tree, but this is also true for FTDNA's tree. (Indeed, any company database is going to reflect only their customer data.) I would highly suggest that every NextGen Y chromosome tester — at FTDNA, Yseq.net, NatGeo, and elsewhere — open an account with YFull. It's well-worth the payment.

But one need not pay the several hundred dollars for a Big Y to determine where he fits in the SNP tree. Individual SNP testing can be done at Yseq.net for $18 each. For example, do you think you're an Akins of Group AF09? The two novel SNPs found for tester #N10367 and their parent SNP can be tested for under $60. The company is expertly ran by Thomas and Astrid Kahn, well-known genetic genealogists of sterling reputation and character.

If you're relatively new at genetic genealogy, this can be a lot to comprehend. If you must, throw out much of what you know about genetics and simply envision the Y as having a binary property. It's either there (a male) or not. Understand that it follows a unique and unambiguous patrilineal thread that weaves its way through the tapestry of human history. As I've said before, "SNPs are people too." Each of these nondescript alphanumeric labels are distinct markers that live in every cell of its host — and each of them originated with a single man. As we learn more about these markers, we learn something of the men who possessed them. Some of these guys were certainly bad asses, some clerics, but most were simple farmers and fishermen. With every newly discovered SNP mutation, we etch another previously unknown feature into the living memorials of our predecessors.

I'm always available for questions.