PDA

View Full Version : X-DNA testing & MayflowerDNA.org



Wing Genealogist
07-17-2016, 09:48 PM
As I have stated earlier, most folks will not have any autosomal DNA segments they inherited from their Mayflower Ancestor. The unique nature of X-Chromosome inheritance (where a son only inherits an X-Chromosome from his mother while a daughter inherits an exact copy of her father's X-Chromosome and a recombined copy of her mother's two X-Chromosomes) does greatly increase the odds, if their Mayflower ancestor was from one of the ancestral lines which passed along the X-DNA from generation to generation.

All of the lines (Y DNA, mtDNA, X DNA, and autosomal DNA) are somewhat akin to a lottery, where the chance of "winning" are fairly slim, but some have much better odds than others.

Baltimore1937
07-17-2016, 11:33 PM
Regarding my U5b2b2/G228A, it's not slam dunk. Firstly, My HVR1 is unique enough to classify it as U5b2. When you could access SMGF, there were only two matches to my HVR1 & HVR2. One had the earliest ancestress in Cincinnati, Ohio in 1812, with a later census giving her an origin in Pennsylvania. The other match was in Canada that I traced back to 1600s Massachusetts. I built a research tree on that match, which is called Ann Johnson (born 1640) tree (or etc), and is public at Ancestry. As for X-matches, I am totally confused and someone else would have to sort that out for me.

loisrp
07-18-2016, 06:00 AM
I think this sounds like a good idea, at least for experimental purposes. Not sure if it can prove anything, because the sister or cousin or similar relative of the Mayflower passenger could also be the original donor.

C5B1
11-03-2016, 05:05 PM
I think this sounds like a good idea, at least for experimental purposes. Not sure if it can prove anything, because the sister or cousin or similar relative of the Mayflower passenger could also be the original donor.

I stumbled upon some records last night that had Resolved White (passenger on Mayflower) as my 9th ggrandfather. That was interesting to find.

geebee
11-04-2016, 06:02 AM
As I have stated earlier, most folks will not have any autosomal DNA segments they inherited from their Mayflower Ancestor. The unique nature of X-Chromosome inheritance (where a son only inherits an X-Chromosome from his mother while a daughter inherits an exact copy of her father's X-Chromosome and a recombined copy of her mother's two X-Chromosomes) does greatly increase the odds, if their Mayflower ancestor was from one of the ancestral lines which passed along the X-DNA from generation to generation.

All of the lines (Y DNA, mtDNA, X DNA, and autosomal DNA) are somewhat akin to a lottery, where the chance of "winning" are fairly slim, but some have much better odds than others.

I think the odds of having inherited autosomal DNA from a Mayflower ancestor are not as bad as you seem to think. The real issue would be actually be how you would identify such a segment -- but more on that a little later.

In the blog we're introduced to an equation, ((22+33*(k-1)), where k represents the number of generations back the ancestor is from a particular parent.

So, for example, my wife's nearest Mayflower ancestor would be Love Brewster, who is my wife's 9th great grandfather. That means he's 11 generations back from my wife, but k = 10 because that's the number of generations back he is from my wife's mother.

So, substituting 10 for k gives us (22+33*(10-1)), or 319. Essentially, this represents the maximum number of distinct maternal ancestors my wife could have in this generation. But her maximum number of distinct maternal genealogical ancestors in this generation is 2^9, or 512. So only 319/512 of these genealogical ancestors may also be genetic ancestors -- or a bit over 62%.

However, this is true when looking at just one ancestor. In fact, my wife has at least five Mayflower ancestors, though two of these are the parents of one of the others; and four would have a k of 11, rather than 10. At k = 11, the estimated number of "chunks" is 352, but now distributed over 1024 possible genetic ancestors. That means that each of the four has only about 34% chance of being the source of any segment. But again, there are four such ancestors.

In addition, keep in mind that the early European population of Plymouth Colony would have remained relatively small for some time. Chances are good that many of those with a Mayflower ancestor (or ancestors) may descend from that ancestor by more than one path. Essentially, the odds given are not really just the "per ancestor" odds, but the "per path" odds. At any rate, in the early years, Mayflower DNA wouldn't just "drop out" -- to some extent it would be "passed back and forth".

Still ... that leaves the question of how to identify a Mayflower segment. It wouldn't be enough just for it to exist, it would have to be shared with other Mayflower descendants, who would need to test and be able to have their DNA compared to your own. Yet the more descendants who test, and the more they're willing to share their results -- at least to the point of being able to be compared to others to whom they are or might be related -- the better the odds will be.

The blog I mentioned is here: https://gcbias.org/2013/11/04/how-much-of-your-genome-do-you-inherit-from-a-particular-ancestor/. I think it's well worth taking a look at.

geebee
11-04-2016, 06:08 AM
In addition, it really shouldn't be necessary to get back to a Mayflower ancestor. It should only be necessary to connect with a proven descendant. If you are descended from B, and B is descended from A, then it follows that you are also descended from A -- regardless of whether you share DNA with A.

(Clearly, all of my ancestors -- both genetic and genealogical -- are also my daughter's ancestors. But, she only inherited half my DNA. Therefore, it is inevitable that some ancestral DNA still found in me, will not be found in her. Or, to put it in terms of this discussion, she is one more generation distant from the Mayflower ancestors. My wife's mother would have been a generation closer; and her mother a generation closer yet.)

EDIT: When I say "connect with", obviously I mean "be descended from". Mere collateral relatives won't cut it, if they are cousins; but they would be useful if they are blood aunts/uncles in some degree, since they would share common ancestry with which of their siblings would be ancestral to you. Again, you'd still have to prove the relationship.

Even sharing a DNA segment with a Mayflower ancestor would not be proof, since a shared segment might merely mean you are descended from a close relative of that person, not necessarily that you are their descendant. So, for example, my wife could share a DNA segment with, say, William White but not be his descendant. She could be a descendant of one of his siblings (assuming he had any).

At best I can only see DNA evidence being used in conjunction with documentary evidence, not as a complete substitute. It might be reasonable to use it to lower the amount of additional documentation required.

Wing Genealogist
11-04-2016, 10:55 AM
geebee,

You hit the proverbial nail on the head. In order for the MayflowerDNA.org project to link autosomal DNA segments to Mayflower passengers, they would need to identify common segments among descendants (which really is the "easy" part), then somehow try to show these individuals did not inherit these segments from some other ancestry they have in common.

Proving a negative is almost always impossible. However, as the number of individuals who match increases (and all are known to descend from a single Mayflower passenger), the odds of the match being from someone else other than this Mayflower line become more and more unlikely. In the end, we may simply have to accept that there is XX% chance such and such a segment came from (using your example) William White (or his wife).

While you make a valid point about not needing to get back to an actual Mayflower ancestor, one of the purposes of the MayflowerDNA.org project is to see if it may be possible to actually identify some of the autosomal DNA segments for some of the Mayflower passengers. Whether or not this can even be accomplished is still in question.

geebee
11-04-2016, 07:51 PM
wing,

I would think it should be possible to accomplish the task, if you can build a large enough database. There are without a doubt millions of Mayflower descendants. I've seen figures as high as 35 million, though of course no one can really know how many there are.

Obviously, the vast majority of these are unproven or even unknown. At the same time, many of the claims to Mayflower ancestry are incorrect. But my point is simply that the number is certain to include a pretty substantial portion of the U.S. population. It is certainly possible, at least, that most of the DNA of a number of Mayflower passengers still exists, spread through that population. The trick, of course, is to find it -- of more precisely, to recognize it once it's been found.

One huge problem, of course, is that only some of those who had taken an autosomal DNA test have really good trees. Most trees are not sufficiently extensive; and many of those that are, are not sufficiently documented -- or just plain inaccurate.

EDIT:

I've always wanted to find a Mayflower ancestor, myself; but, alas, they don't seem to have "mingled" sufficiently with my ancestors -- although I do have some gaps that some unknown Mayflower forebear is certainly welcome to fill.:beerchug:

(I thought at one point that I'd uncovered a connection to Rebecca Rolfe, otherwise known as Matoaka, Amonute, or Pocahontas. Unfortunately, it was through a Bowling ancestor and I seem to be descended from the wrong sort of Bowlings: "blue", not "red". I hadn't know any of my ancestors were color-coded, but apparently they were.)

Meanwhile, I content myself with having ancestors who were in French Canada by mid-17th century, ancestors in New Amsterdam by about that time, and a few ancestors who met the other ones at the boats. Most of my roots in America, however, only go back to the mid-18th century.

Wherever they came from, the important thing for me is that they all somehow manage to get together -- or at least their descendants did.

Anyway, I remain optimistic that Mayflower passenger autosomal DNA can be found. You may be right that most descendants will no longer have such DNA, but I would estimate that for a descendant of my wife's age -- which I won't reveal, but I just turned 60 and she isn't far behind me -- there may be at least a 1-in-3 chance for any one such ancestor to have passed down at least one segment.

There are certainly enough Mayflower descendants to have a decent chance of overlap.

Wing Genealogist
11-05-2016, 10:44 AM
I have been fortunate enough to have ancestors who came over during the "Great Migration" period to New England (1620-1640). I have been able to trace most of my lines to the immigrant ancestor. To date I have documented (to my satisfaction) 17 Mayflower lines from 10 different passengers.

I am in my early fifties, so it is good to hear that statistically speaking, there is a high likelihood at least some segments of the DNA in my body came from someone aboard the famous ship.

The question remains, will DNA technology (and databases) ever get to the point where such segments can be mapped?

C J Wyatt III
11-05-2016, 01:57 PM
The question remains, will DNA technology (and databases) ever get to the point where such segments can be mapped?

I think it is possible right now, but my view is that the current matching methodology is flawed.

One of my contrarian views is that small autosomal segments are very rarely false. If you are able triangulate them with someone who is not close, that eliminates any remaining worries about whether they are false. Also I believe many of us have a common ancestor within the past 300 years (I do not believe he comes from a Mayflower line). I think we have the capability to determine his autosomal genome. We need to be able to edit out his results in kits that we are comparing, so that we know that a match comes from somewhere else.

Myself, I have been able to make some connections as far back as the 17th Century.

Jack Wyatt

Saetro
11-05-2016, 11:36 PM
I think it is possible right now, but my view is that the current matching methodology is flawed.

One of my contrarian views is that small autosomal segments are very rarely false. If you are able triangulate them with someone who is not close, that eliminates any remaining worries about whether they are false. Also I believe many of us have a common ancestor within the past 300 years (I do not believe he comes from a Mayflower line). I think we have the capability to determine his autosomal genome. We need to be able to edit out his results in kits that we are comparing, so that we know that a match comes from somewhere else.

Myself, I have been able to make some connections as far back as the 17th Century.

Jack Wyatt

Segments can be gold or mullock: they need to be panned out.
But if you are picking out massive nuggets why work with gold dust?

DNA Genealogy provides evidence.
Just like other genealogical forms - oral, family records, vital and archival records, pedigree books and so on.
The more evidence you have on a particular issue pointing to a particular outcome, the more evidence you have for that outcome.
And the other problem is that small segments tend to come from a common ancestor from long ago.
This is a statistical situation not disproved by just one counter argument.

If you have lots of DNA examples that a couple of particular 3cM segments are definitive for a particular family line, (say by triangulation), then if you find them in the DNA of a total stranger, you have a very strong lead towards establishing a relationship, and it is worth devoting more resources to this task.

But if you are starting out with only your own genome, it is a total waste of time to put effort into tiny scraps.

I am in the position of having many close mtDNA matches: there is no question that there is a link somewhere.
With a couple, I share some small segments of 4cM and 5cM. Links to a common ancestor are well documented from paper sources and he was born around 1480 AD.
The segments provided a spur to look for the paper proof, but without the additional evidence of the mtDNA, I would have spent my time on something more promising. Good pedigrees on both sides were a bonus.

Dave Hamm's blog provides an example of using small segments as additional evidence to a well constructed prior framework of a Y DNA / surname study.

C J Wyatt III
11-06-2016, 12:03 AM
Segments can be gold or mullock: they need to be panned out.
But if you are picking out massive nuggets why work with gold dust? .....



With all respect, you are advocating for the flawed methodology
which I was speaking of.

The question was whether someday, and I was saying that I already was making connections that far back. I'd be happy to start another thread to discuss my approach and the shaky foundations of the current methodology if anyone really is interested.

Jack Wyatt

curiousII
09-21-2017, 12:39 AM
I have an X Match with a Hopkins who lists a Doty in Ancestral Names. Haven't tried email yet, I'll wait for my GSMD membership number before I start bothering anyone. I have other FF matches that claim Mayflower roots, but Hopkins is the only X Match I've found right off.