This isn't a tutorial about the Y chromosome. It's a wide interpretation of the latest test results from the Cooley DNA Project, their genealogical consequences, and hints for additional testing. A more detailed article in respect to CF02, CF03, and CF09 will come in a few weeks.
The project has grown considerably since Greg Parker founded it in 2005. I first tested in 2006 and came on board as co-admin in 2012. But in recent years, memberships have flattened in virtually every project I admin. I attribute that, at least in part, to the growing popularity of autosomal testing. (Ancestry.com can now market to 100% of the population rather than to only 49%.) But make no mistake about it, Y-DNA is the most powerful tool in single-surname studies. This is a discussion about the several helpful tests and upgrades we've had over the last couple of years. From an admin's point of view, this is a straight-forward, simple procedure. But it can take years to fully develop.
Tracking lineages through the nineteenth and early twentieth centuries is relatively easy. For the most part, the "targets" will have the same surname. The records — federal and state census schedules, tax lists, wills, marriage records, and much more — are readily available. (Libraries are a much-neglected resource these days.) Even the last seventy years can be traversed through obituaries, birth records, death records, family letters, and numerous other sources. To help me in this Cooley quest, I keep a record for the project's lineages at my public page, Early American Cooley Patrilineages. The maintenance of public web pages (as opposed to those that are accessed only through subscription) is important to genealogy: Search engines can find the data.
The toughest part in this whole enterprise is convincing people to test. Having project funds available helps encourage new testing. Anyone — project members or the general public — can contribute to the project's General Fund. The debited amount will sit in FTDNA's coffers until a group administrator places an order. The contributor can specify that the moneys be used for a particular tester and/or test, or she can leave it up to the administrators' discretion. Group contributions have helped project managers purchase kits that are otherwise unaffordable for the testers. The results not only aid the researcher but promotes the advancement of the surname's study, the building of the world-wide SNP tree, and genealogy as whole.
Before ordering a test, however, it's best to check with an administrator. A hasty testing decision can lead to redundant results. For example, if you've ordered the Big Y, don't do a backbone test — and I doubt I've ever advised anyone to do the Y-111. That money is best used elsewhere. Big Y tests are, of course, extremely useful, but if two or three Big Y's have tested in your lineage, the results will largely be academic. If it's a well-developed group, new testers might need only order a 12-marker test for general placement followed up by the group's defining SNPs. Cooley Group CF02, for example, is getting close to that point. However, FTDNA has pulled critical SNPs used by the project from their product menu and they've put a halt, if temporarily, on adding new SNPs to the lineup. Still, most SNPs can be tested at YSEQ and at better than half the cost ($18). (A separate kit is needed.) Such ongoing difficulties with FTDNA has prompted me to begin recommending these YSEQ products to Y-DNA newbies:
Still, FTDNA has lowered the prices for the Big Y considerably.
YSEQ-Alpha $58 16 Y-STR markers YSEQ-Gamma $95 36 Y-STR markers
Existing customers $339 Upgrades from Y-500 to Y-700 $209 New customers $449
We're talking trees here. SNP trees, family trees, indeed any kind of phylogenetic tree, and even mathematical trees, such as binary trees, have a common structure.1 The position at which a parent or couple sits is the parent node. The nodes above it (upstream) are ancestral nodes. The nodes below a parental node (downstream) are descendant nodes. The node common to descendants is called the Most Recent Common Ancestor (MRCA — some scientists call it the Last Common Ancestor, or LCA.) What makes a Y tree different from others is that only males and their sons are listed. Therefore, all descendants of a Y tree node will have not only the same ancestors, they will have the same Y chromosome. And if a small mutation creeps into the lineage at a birth, the marker will show up in all his descendants. (Don't confuse this with recombination and autosomal DNA, which has a very different inheritance pattern.)
Wikicommons: binary tree
I've put together a short glossary, appropriately entitled Glossary, rather than (once again) defining the terms referenced here. At present, it's pretty much limited to the words and phrases I use in the following. More entries will be coming.
The tested Y-SNP markers found among the CF01 group are consistent. Everyone in the group is positive for YP4491. The SNP is positioned on the tree several degrees downstream of L448, nicknamed "Young Scandinavian" for a straight-forward reason. Following that nicknaming scheme, I've separated the group into Old CF01 and New CF01. The New CF01 group has two surnames, Cooley and Whitfield. Two of each family has tested the Big with virtually identical results. There's no way around it: John Cooley (c1737-1811) and William Whitfield (1751-c1835) were closely related, whether they knew so or not. It has taken about fifteen years, but it seems that this phase for New CF01 is nearly played out — with one probably exception. There's a surprise in the branch of Edmond Cooley (c1773-1851, married Charlotte Speace). Edmond and some of his extended family moved to Spartanburg, South Carolina by 1805. Two of Edmond's children married Cantrells there and produced several generations of cousins. Somewhere along the line, at least three or four generations ago, an NPE was introduced, just where cannot yet be determined. But at least one Spartanburg Cooley is of a Cantrell haplogroup. I find this worth persuing. On the other hand, no Old CF01 members — descendants of probable brothers William Henry Cooley (1797-1877) and James Cooley (1808-c1872) of Pennsylvania, Ohio, and Illinois — have had advanced testing. We know it's fertile ground because there are two significant Y-STR differences (DYS464b and DYS570) between New CF01 and Old CF01. These differences might be reflected in an undiscovered SNP, which can be found only through Big Y testing. The test stands to give us a better understanding about the degree of relationship between these two subgroups and can help to point to the origins of CF01.
These are the Tring (Hertfordshire) Cooleys. All fully-tested members share the A12020 umbrella haplogroup, which appears to have had origins in France. It's not only a large group but includes several unique markers, the presence of which permits us to parse it into at least four subgroups: Benjamin Cooley (1615-1684) of Springfield, Massachusetts; Samuell Coley (1614-1684) of Massachusetts and Connecticut; Nehemiah Cooley (c1685-1759) of Hertfordshire, England; and a group having an unknown mutual ancestor who, for now, are collectively known as the Goshen Cooleys. This group is dubbed "Goshen" because the following progenitors all lived in or near Goshen, New York: Daniel Cooley (1688-1762), Abraham Cooley (c1740-1820), Thadeus Cooley (-1814), and Jabez Cooley (1730-1808). The connection, however, goes deeper than shared surnames and geography. The Goshen descendants share the same markers, as shown at Advanced SNPs for Cooley Group CF02. Likewise, Benjamin's descendants can confirm descent through testing individual SNPs. And we've just verified SNP FT122534 in the Samuell Coley lineage. It belonged to his descendant Peleg O Cooley, the MRCA for both testers. The question in front of us now is, How much further up the tree can the SNP found? In the meantime, I recommend YSEQ's R1b-DF27 panel, as described at the end of CF09.
This is an oddity. The single tester in this group (a descendant of Hezekiah Cooley, 1802-1870+) has closely-matching STRs with CF02, a genetic distance of only two from the 67-marker modal. He matches with twenty-eight Cooley testers. The surprise is that he tested positive for the R1b-U152 SNP, which tells us that the Most Recent Common Ancestor (MRCA) between himself and CF02 lived about 5,000 years ago. (CF02 is descended from U152's "sibling" haplogroup, R1b-DF27, which is also about 5,000 years old.) That there are Cooleys of different haplogroups is obvious. But that two groups share the same surname and the same STRs while belonging to a haplogroup that diverged thousands of years ago simply doesn't fully compute. Certainly, the laws of probability make it possible, but this case is so extreme that I believe the presence of the U152 mutation might have been a one-off; that is, it simply occurred out of context to the other markers. That happens. For example, SK411, which FTDNA uses to define CF09 (not my choice), was first discovered in 2014 within the B2b1a haplogroup — virtually light years away from the R1b major haplogroup most Western Europeans belong to. But there are inexpensive ways to sort this out: submit a new kit to YSEQ and test for the SNP A12020 ($18 + $6), the umbrella SNP for the whole of CF02. The R1b-DF27 panel, described for CF02 and, especially, CF09, includes the three known CF02 subclades. And there's an interesting genealogical aspect to this: Hezekiah was born in Canada, the place unknown so far as I can tell. Another Hezekiah went from Massachusetts to Canada and died on 29 May 1804 at Bromont, Quebec. There is speculation, of course, that they were father and son, but the Canadian birth first needs to be found before we can take the speculation seriously.
There's a 5,000 year difference between DF27 and U152
Testers in this group are from the US, Australia and England. I call it the Pennsylvania Cluster because the US testers, who comprise the majority of the group, have origins in Pennsylvania. The most distantly-removed STR results for CF04 are those for the testers who descended from Daniell Cooley (-1720) of Maryland and his grandson, Richard Cooley (1756-1840) of Fayette County, Pennsylvania. The Maryland group has yet to do advanced testing. But we do have three Big Y results from testers descended from probable brothers Robert Cooley (1754-1794) and Francis Cooley (1767-1842). This subgroup is well defined and positive for FT157756, one step downstream from the A12227 haplogroup. (FTDNA changed the name to FT156738 in one of their many confounding moves.) And recent Y-700 results, which includes testing for 50% more sample than the previous iterations of the Big Y, has expanded the number of SNPs in FT156738 to fifteen, as illustrated below. From past experience, I expect many of those will be moved upstream as more testers come on line. In other words, we don't have the final word yet for this group, and there's much to be learned from advanced testing from the England and Maryland lines, particularly a Big Y. But for those who would like to do individual SNP testing (and I'd advise it), I recommend testing those in blue at YSEQ.
No advanced testing has been completed for this group. The two testers are said to be descendants of Peter Cooley of Fredericksburg, Virginia who arrived with his family from England in 1774 as indentured servants. They are of the major haplogroup E-M35, rare in Europe. Most Cooley researchers have run into this family while researching for their own Southern Cooley origins. For this reason, CF06 rates relatively high on the Cooley "fame" index. It would be nice to know more about their Y-DNA.
Like CF06, new researchers often stumble onto the North Carolina Cooleys. It is often wondered, Are they related to one another and how? Y-DNA has sorted much of it out. There were at least two Cooley clans in Halifax County, one of which shares markers with a group from Franklin County. The Franklin County progenitor has long been mistakenly said to have been a man named John Cooley. However, the guardianship records discussed in article 61 prove that the Franklin County forebear was named Edward, not John. The Y-DNA results, in conjunction with the genealogical clues (not proofs), suggest that Edward (-c1787) might have been a son of John C Cooley (-c1767) of Halifax, if not a brother. Two of John's descendants have Big Y results proving the R1b-BY38665 haplogroup. Advanced testing for Edward's line will give us a better idea about the relationship between the two men and put us a step closer to the other end of the Atlantic. Speaking of which, CF07 has markers suggesting they are of Clan Colla of Ireland.
The genealogy for this family is interesting. They descend from John Cooley (1809-1888), an illegitimate son of Catherine Burner (as stated in public record). John, whom she later married, a son of Peter Peter (CF06). In other words, a scion from the CF08 lineage was adopted into the CF06 lineage. The genetics and the public record concurs on this point. But the tester has thirteen novel SNPs, which means that more testing needs to be done in order to narrow down toward a terminal haplogroup. I last wrote about this a couple of years ago in an article titled A Cooley By Any Other Name. Not an exacting title but kind of catchy.
I've written quite a bit about this group in the last year or so, most recently in Brief Cooley CF09 Redux, which includes a graphic detailing the latest SNP results. This is a well-populated, largely centralized group (geography-wise) but has little known genealogical connective tissue, meaning the MRCA (who likely resided in Virginia) isn't close to being identified. Four group members (including a Roach) have Big Y results with an average variance of five SNPs downstream from haplogroup R1b-SK411 (which, as I described above, is a SNP name that poorly represents the group. I will resume referring to it as A21492.) This suggests that the Cooley/Roach MRCA lived at least four hundred years ago. And that means we can safely assume that most of the participants in the group will be positive for A21492. Anyone wishing to confirm their placement can order A21492 at YSEQ for $18 (it's not available at FTDNA). Or I'd recommend YSEQ's R1b-DF27 Panel. The relevance of this panel for both CF02 and CF09 is illustrated in the following table. But I'd specially recommend a Big Y to a descendant of Abraham Cowley (1691-1779) of Charlotte County, Virginia. The STRs for him suggest that he was the most distantly-related of the group and could show slightly more variance than his counterparts. Additionally, I'd recommend the following. Instead of a second Big Y, descendants of Benjamin Coley (1794-1841) can test SNPs A22783, A22784, and A22785 at YSEQ for $18 each. Likewise for descendants of Washington Coley, SNPs A22607 and A22608, and for Seabourn Coley, SNPs A21490, A21493, and A21498. Any matches could mean they will be added to the said panel for future testers. Otherwise, the only subgroup defined so far for CF09 is for James Cooley (1758-1834) and Penelope Gargus. Their haplogroup (R1b-A21494) can be verified through the YSEQ R1b-DF27 Panel, which has more than 300 nodes represented. It's $88, an excellent price for anyone wanting to confirm their CF02 or CF09 placement. This illustrates only the "local" nodes for the two groups.
Y15926 ..Y24591 ....A12020 ......A12022 ......A20351 ......FT122534 ....A21492 ......A21494 Possibly France
CF02, the Tring Cooleys
the Benjamin Clan
the "Goshen" Cooleys
Samuell —> Peleg O Cooley
CF09, the Virginia cluster
James Cooley of VA and NC
The initial tests in this haplogroup were critical in revealing the falseness of Lura Hamil's Dutch Cooley assertions. We now know (and there's plenty of documentary evidence for it) that Elizabeth Firmin married William Cooley, not the John Cooley purported to have been the John Cooley of Stokes County, North Carolina (CF01). A descendant of William's has tested the Big Y. The presently known terminal haplogroup is R-S984, which includes 31 SNPs, putting it in the neighborhood of three thousand years old. The many STR matches have names that suggest Scots-Irish origins and Big Y testers of this group include Scully and Keenan. The Cooleys and Scullys have only two or three SNPs differences, which puts their mutual ancestor well within the genealogical timeframe. I can't help but wonder whether the two names (Cooley and Scully) were independently corrupted from the same name. In any event, the Cooley tester has three novel SNPs (FT43454, FT43285, and A24205) and can be easily verified through additional testing at YSEQ for $18 each. So, that's William. John Cooley (1749-1813) of New Jersey, married to Abigail Lippincott, is a near exact STR match to William Cooley. More testing will help home in on their relation, whether if be close or distant. (I would imagine that it would be closer than that of the Scullys.) Autosomal testing on both lines might also provide useful information.
The two current testers in this group are a genetic distance of 0 out of 37 markers from one another. Both lineages appear to have originated in Montgomery County, Maryland but their MRCA is unknown. Kit #344328 is a descendant of John Garrett Cooley (1804-1883), possibly the son of William Harrison Cooley (1770-1840) and Mary Garrett. Kit #MK65081 is a descendant of Thomas T Cooley (1811-1854), possibly the son of Edward Cooley and Elizabeth Talbott. It's easy to imagine that Edward and William were brothers. But it's also possible that one man has been split into two identities by researchers, a fairly common malady in genealogy. Interestingly, the 37-marker results for both men have more than 100 matches, the closest of which have surnames commonly found in North Sea and Baltic Sea countries (Sweden, Netherlands, Finland, Norway, etc.). At least four of these men have taken the Big Y. This fits with the testers' major I1 haplogroup. However, most of the matching names fall off the radar with the higher resolution 67-marker test by #344328, leaving only thirteen matches having mostly English names having genetic distances of at least 6 of 67, which is typically considered rather distant. This tells me that, although matching SNPs and a new downstream haplogroup will undoubtedly be found in Big Y testing, there will likely be a large "void," perhaps two dozen or more unbranched SNPs over a period of a couple thousand years or so. Needless to say, I'd like to see more work done on this group.
Despite having four members, this group isn't well defined. They might have been descended from James Cooley of Pittsylvania County, Virginia but the evidence is minimal. The group has a large number of low-level STR matches with men having Scottish and Irish names, at least five of whom have had Big Y testing. Those tests can provide a benchmark for advanced CF12 testing. Considering the Scots-Irish surname matches and that the group is of the major haplogroup I2b, these Cooleys might have had Viking origins.
The STR results alone were surprising for group CF01. As Lura Hamil had done with John Cooley (c1738-1811), she lumped Daniel Cooley (1765-1826, married Mildred Ball and Elizabeth Grundy) into her faux Dutch Cooley genealogy. Mildred Cooley Tallant, who had corresponded with Lura in the 1930s, doubted the claim. (Much of Mildred's correspondence is archived by Dennis Young at Mildred Cooley Tallant Papers and at the University of California, Santa Barbara.) But it might have been Dale Walker, a CF01 descendant, who first lent credence to the idea that Daniel was of the Stokes County Cooleys. Because he lived in many of the same locations as John's sons (Virginia, Kentucky, and Missouri), it seemed believable enough, and few of the researchers I corresponded with during the 1990s and 2000s expressed doubt of his placement as one of John's sons. But the Y-DNA is clear: CF01 and CF13 are genetically separated by about 25,000 years! Still, beyond the STR results, we know little about this group's Y-DNA. A SNP test for one of the descendants is positive for Z16372, a haplogroup of twenty-eight SNPs. It's estimated by Yfull to have emerged between 2100 and 3500 years ago. That doesn't tell us much, but it does reserve a spot for CF13 on the Cooley Y-tree. Of the twenty-eight 67-marker matches, thirteen have Big Y results. The terminal SNPs are highly divergent, but it provides a good benchmark for advanced testing on this group. A Big Y for CF13 could be very telling.
Z16372 < BY16 < S6365 < L513 < DF13 < L21 < Z290 < P312 < L151 < P310 < L51 < L23 < M269 (R1b1a1a2)
The Big Y tester in this group, a descendant of William Cooley (1802-1876) of South Carolina, has 26 novel SNPs downstream of R1b-BY154158. That's far too many to test individually. He has no Big Y matches and only five Y-37 STR matches, which suggests that his part of the world-wide Y-tree hasn't been well-tested. Advanced testing from at least one of the other group members is needed to narrow the results down to a haplogroup that formed within the genealogical timeframe.
This Kewley lineage is of haplogroup A421 and has distant Big Y matches to a Lewis in A421 and two Morrisons of haplogroup A420. He matches one each to collateral groups FGC32537 (Lewis) and FT87728 (Hamilton). In other words, each tester belongs to the upstream A18 haplogroup, as shown below. Kewley has six novel SNPs, two of which belong in the notoriously unstable DYZ19 region. His stable SNPs are FGC33438, FGC33441, FGC33442, and FGC33444. They will remain novel only until matching testers come along. Some or all of those SNPs belonged to his ancestor, Thomas Kewley (c1759-) of Kirk Marown, Isle of Man. Targeted testing of others, even testing only for those SNPs at YSEQ, will provide more information.
L1403 (3-4 kya) ..L1402 ....A18 ......A421 ........A420 ........FGC32537 ......FT87728
There has been no advanced testing in this group. It presently has two members, a descendant of Philander Coley (c1798-1869) of Albany, New York, and a descendant of James Cooley (1801-) of Yates County, New York. Interestingly, the first tester had significant matches only to Bixbys, but the match to Philander presents us with the likelihood that the lineage was Cooley (or Coley) at least since the 1790s. Their ancestors lived near CF02 territory, but the Y-DNA tell a CF06 story. A Big Y would be very helpful to further studies of this group.
Kit #400507 is the only unmatched Cooley in the project to have done advanced testing. His Earliest Known Ancestor (EKA), William Coley, was born in South Carolina in 1822. He's of the haplogroup R-BY166389 and FTDNA reports no novel SNPs for him. That means he's closely related to his matches, at least within the last 150 years or so. The tester has three Big Y matches to McClendons, and two to McDills, and stands to learn just when the name Coley originated in his lineage. (At this juncture it might be worth stressing again that our surnames and families are ours. It doesn't matter whether a name came into usage 800 or 200 years ago.)Closing Thoughts
The first take away here is that the Big Y matters. Apart from the far more expensive Whole Genome Sequencing, it's the best deal out there. It examines up to fifteen million positions on the Y chromosome while looking for previously undiscovered mutations. It's a SNP hunter. If you're positive for a known SNP, the results will report it. If the test finds a stable SNP that has never before been found, it will be reported as a novel SNP. Until someone else comes along and matches, these are considered personal SNPs. In this way, the Big Y finds SNPs that are unique to your surname, your ancestor's geographical region and, often, unique to your immigrant ancestors. If you have forty novel SNPs, there's vast territory — geographical and time-wise — that is yet to be uncovered. That can be fixed by hunting down an eligible tester or, most often, just waiting until he shows up. Pockets and huge swaths of the population have yet to take advanced Y-DNA. After all, only a small percentage of men in the world have tested.
The second take away is that you should consult your admins before taking a new test. You want to avoid wasting money as well as redundant results. For example, unless explicitly advised otherwise, a backbone test and SNP panel will place you in a haplogroup several thousand years old, and one that might already be identified for your lineage. For some of you, upgrading from the Y-500 to Y-700 might be useful. But consult us; that territory might also be covered. In fact, an admin might suggest, considering what has been collected and what your personal goals are, that your next best test might be autosomal.
And, finally, FTDNA need not be considered your only real Y-DNA testing option. For the Big Y, yes, and for its surname projects, yes. But I've included recommendations above for testing specific SNPs, STRs, and haplogroup panels at YSEQ. I plan more about that in a future article. In the meantime, please join us at the Worldwide Cooley Y-DNA Project. And I'm always happy to answer questions.
1. A family's pedigree chart is merely an inverted binary tree.