Thursday, February 4, 2016

Triangulation made me do it

This is a story that began with triangulation, and so relies on the assumption that genetic matches at the same chromosome location share a common ancestor, and that looking for other matches to the same segment might help the genetic genealogist discover that ancestor.

I have a brick wall at my GGG Grandmother, wife of John Addair (born around 1780), and mother of my GG Grandfather, also John Addair, born 1814.  That is the line I would most like to figure out, and I have been working on it since way before I got into genetic genealogy in 2011.  Many trips to courthouses have not turned up clues.

Several of 1780 John's descendants have tested and match me, and there is a triangulated group between three of us - J (a 4th cousin), A (a 1/2 3rd cousin once removed), and me.  

            Chr.    Start              End                 cMs    SNPs
J to me    1    232,268,768   238,713,885   16.9   2768
A to me   1    232,419,522   238,842,845   17      1967
J to A      1     230,975,451   238,713,564  19.6    2350

J, A, and I share John Addair and his wife as ancestors, so we can guess that our shared DNA came from those ancestors.

I started looking at other genetic matches in that spot, hoping for clues, and found three descendants of Peter Muncy (born in 1782 in Lee County, Virginia, died 1856 in Clay County, Indiana) and Frances Owens Wood (born 1782 in Montgomery County, Virginia, died 1844 in Clay County, Indiana) in the same spot.   Eventually a fourth turned up.  They also match J and A.

            Chr.    Start            End               cMs    SNPs
D to me  1    232,344,068 239,154,331 18.3 2066
S to me   1    232,344,068 239,154,331 18.3 2065
M to me  1    233,142,768 238,967,960 15.8 1730
St to me   1   232,430,908 238,961,206 17.5 2788

D, S and M are descended from three different children of Peter and Frances Muncy.  St and D are 2nd cousins, and descended from the same granddaughter of Peter and Frances Muncy.

I had noticed Muncy in match surname lists before, but hadn't followed up because I do have another Muncy ancestor, an Elizabeth Muncy, born around 1775, who married Gabriel Riffe, probably in Montgomery County, VA.  Gabriel and Elizabeth had at least 50 grandchildren, so I had been attributing matches who mentioned Muncy to them. But my John Addair cousin matches, A and J, are not in this line, so Peter was very interesting.  The more I looked at all my matches - 23andme, FTDNA, Ancestry and gedmatch (yes, I am addicted) - the more Muncys turned up, many as fairly close matches. 

Sometimes you get lucky.  And sometimes you get VERY lucky, and find amazing research about the person you're tracing.  I started googling Muncys and found Roberta Estes's blog entry 

"Samuel Muncy (1761/1768-1839), Who’s Your Daddy, Your Mamma, and Your Kids?, 52 Ancestors #56."  

The Samuel Muncy Roberta researched and blogged about married Ann Workman, daughter of Joseph Workman and Phoebe McMahon, in Montgomery County in 1788.  He is either a brother or a first cousin of the Peter Muncy my matches descend from.  And, according to some researchers, this Samuel and Ann had children who lived in Lawrence County, Kentucky, which is where my John Addair and his unknown wife lived at one point. That made me alert to Workman, and I found many of them in my match lists, including descendants of Joseph and Phoebe Workman other than the Ann Workman who married Samuel Muncy, several of whom form triangulated groups also.

This just turned up around the middle of December, and I've been obsessively pouring over match lists, and censuses, and trees ever since.  Peter Muncy's dates and locations don't work out to be an ancestor - I think he would have to be a brother, uncle or cousin of my GGG Grandmother.  I can't pull it all together yet, but there are some clues.

John Addair, the GGG Grandfather with the missing wife, (born around 1780, Montgomery Co., VA died around 1832, probably Lawrence Co., KY or nearby) is one of those hard to trace ancestors - he moved a lot, and I don't think he ever owned land.  He was born in Montgomery County, Virginia, probably where his father, James Addair, Sr. settled around 1773 on the New River at the mouth of the East River (this is now Giles County, Virginia). The East River is about 8 miles down New River (west) from the mouth of Wolf Creek, which was where several Muncys settled. During that period James Addair served in Robertson's Company in Dunmore's War, and Holton Muncy, an uncle of Samuel and Peter, was in the same company.  Later, about 1785, James Addair Sr. sold that land and bought land a little further east on the New River, about where Radford is now, where he operated a ferry, and Frances and Skidmore Munsey, uncles of Peter and Samuel, signed the petition to establish the ferry in 1794.  This would not have been far from Back Creek, another place where Munseys settled.  

Son John Addair is found on some Montgomery County tax lists, but then appears in Cabell Co. VA in 1810, where his name shows up on a petition, I think next to a Samuel Muncy, although both names are very faint. 

I didn't find either of them on the 1810 census, which is incomplete.  John Addair is on the tax lists for Cabell Co. through to 1820.  He's on the 1820 census, with a wife and several children.  In 1823 he shows up across the Big Sandy River in Lawrence Co., KY, in the part that came from Floyd Co., and is there on the 1830 census (next to a William Munsey), and a few tax lists.  And then he is gone. The family story was that he was cutting wood for a steamboat on the Ohio, gashed his leg with the axe, and bled to death. 

Back across the Big Sandy River there are three Adairs on the tax list for Logan County, VA (Logan came from Cabell) in 1835 - Samuel, James and Phoebe.  Samuel and James are the two oldest sons of 1780 John Addair.  Phoebe is not his child.   John Addair's family is very well documented, with the aggravating exceptions of his mother's name and his wife's name, because his half-brother, James Addair, died around 1846 in Pulaski County, VA, with a very large estate and no will. There was a series of lawsuits to settle the estate between James's full siblings and half siblings, including extensive lists of those siblings' children.  No Phoebe. 

There was a marriage in Lawrence Co., KY between a Phoebe Adair and William Farley, February 25, 1836, which I also couldn't explain.  Once I started looking for Muncys everywhere, though, another marriage turned up across the river in Logan Co., VA, between Phoebe MUNCY and William Farley, on the same date - February 25, 1836.  The record in Lawrence County is a minister's return, and I have a copy of the actual return.  The Logan Co. record just shows up on online lists of Logan County marriages, so I really don't have a good explanation for it being in two places. But my theory now is that John Addair married a Phoebe Muncy, probably around 1806-1810, and after he died around 1832 she showed up on the tax list as a single woman in 1835, and then she married William Farley.  And the name Phoebe makes me wonder about connections to Samuel Muncy and Ann Workman, daughter of Joseph and Phoebe Workman.  John Addair and his unknown wife's eldest son was named Samuel.  And their son John, my GG Grandfather, named his second daughter Phoebe (his first daughter was named Susan, for his wife's mother).

I think the William Munsey next to John Addair in on the 1830 Lawrence Co. KY census is the one listed in "Some Branches of the Workman Family" (one of the many references Roberta Estes's blog pointed me to), who married Peggy Hensley.  John Addair's sons Samuel and James also married Hensleys in Lawrence County, and the Hensley family is listed on the same 1830 census page.  That William Munsey was later involved in at least one of the lawsuits over the James Addair estate back in Pulaski County, VA, although he stated no relationship to the family.  One of the people he obtained depositions from is a Dicey Munsey in Wayne Co., Virginia (Wayne was taken from Cabell Co.).  On the 1850 Wayne Co. VA census, Dicey is the wife of Samuel Munsey, b. about 1798.  In Roberta Estes's blog another researcher is quoted as stating Samuel Muncy and Ann Workman had a son Samuel, who married Dicey Spalding.  

The pieces don't all fit together perfectly, but there are getting to be a lot of pieces.   I have at least one other match at the same location who doesn't fit either my tree or the Muncy tree, but does have family from the same area.  

There is quite a bit of doubt about the usefulness and even the possibility of triangulation.  Possibly it's not a worthwhile or valid technique.  Possibly I am just lucky.

Saturday, July 4, 2015 has new tools for subscribers, for downloading AncestryDNA data.   Thanks to Cora Lou for this link, which describes the tools:

I tried it out.  My main problem was being a Mac person in a PC world (since 1984!), but I will try to overlook that.  

The match file has the correct number of matches - 5181.  The ancestors file took about 3 or 4 hours to download, and has 306,346 lines. For comparison, the last Snavely ancestors file I downloaded, two months ago, had 253,411 lines, so - in the right ballpark.  I haven't played with it, but it does include a column labelled "relid" which appear to be Ahnentafel numbers for the ancestor, and also includes birth places and death places.

The shared cM in the matches file is most interesting, since it reflects post-Timber measures.  Here are 26 matches who transferred to other services (mostly gedmatch):

AncestryDNA cM gedmatch cM FTDNA cM 23andme cM actual relationship
46.939 55 4th cousin
36.205 61.8 48.11
33.326 44.8 42.95 42
30.437 41.7
28.005 39.4
27.804 23.6
27.555 37.41 35.4
24.58 25.2
23.477 56 double 5th cousin+
23.33 39.4 5th cousin 
23.021 32.8 12.2 1/2 3rd cousin +
22.812 34.8
22.793 38
22.67 34.4
22.301 34 23.32 28
21.969 33.9
21.69 33.9
21.52 25.4
20.854 36.3 31.23
17.574 31.6
17.389 33.9
15.684 35.9 33.5 32 double 4th+
12.332 36.5
12.254 20.1 6th cousin
10.208 22.2

Either phasing or Timber or both might account for the differences.   In one case I'm pretty sure it's phasing, and Ancestry has it wrong:  a 5th cousin where AncestryDNA shows 23 cM and 23andme shows 56.  This is one long segment, and the individual's mother has tested at AncestryDNA, but not 23andme.  She shows 63 cM.  I suspect the son inherited the whole segment, or most of it, but Ancestry phased his wrong.  Yes, this is a match from a family with a lot of endogamy.  In fact, I think all of these are.

Late edit:  For gedmatch, FTDNA, and 23andme cMs I included only segments over 7 cM.  The relationship listed is the closest one, "+" indicates there is at least one more, more distant relationship.

Friday, December 12, 2014

TIMBER! Forestry practices at AncestryDNA

How is AncestryDNA's new TIMBER procedure selecting our matches?  This is the story of two matches I had which, before Autosomalgeddon, appeared to be identical.

I became aware of these two matches because they showed up on GEDMATCH, and both matched me 18.9 cM on the far end of Chr. 11, where I have my only SubSaharan African segment.  These new Ancestry matches - R. M. and B. D. - were both estimated as Distant, 5th-8th, Low confidence matches under Ancestry version 1.   At GEDMATCH:

me to R.M. Chr. 11 126390369 to 134436845 18.9 cM 2583 SNPs

me to B.D. Chr. 11 126357648 to 134436845 18.9 cM 2574 SNPs

They matched each other, and matched other matches I have at Gedmatch.

I have been interested in this segment for a few years, and am tracking several other matches to it from 23andme and FTDNA's Family Finder, but B. D. was particularly interesting since she is predominantly subSaharan African.  All the other matches I had previously on this segment were mostly European, although they all shared the SSA segment at the end of Chr. 11.   I have not tracked down a common ancestor, but several (including me) have Collins ancestors, two (including R. M.) have Sexton ancestors, one of the Collins appears to have Lumbee connections, and all  for whom I have any information have Appalachian ancestry.   B. D. had a great grandmother from Roanoke, Virginia, who was believed to have Native American ancestry, which fits the mixed race pattern.

Then came the new, revised match list.  Now R. M. is estimated as a Distant, 5th-8th cousin Good match.  But B. D. is no longer a match at all.

I'm puzzled by this - the segment appears to be the same.  

Since we were all AncestryDNA matches before the rollout of version 2, the lack of phasing at gedmatch isn't the problem, since the version 1 set of AncestryDNA matches were phased.   In addition, I have a sister who is also at gedmatch, and B. D. also has a sister there (both sisters tested at 23andme).  My sister matches B. D. in exactly the same place, B. D.'s sister matches me in exactly the same place, and the two sisters match each other.  If a lack of phasing produced a pseudo-segment, it produced exactly the same segment in all four of us.

So I think TIMBER is the most likely reason for the change, but am still puzzled.  What could make the algorithm treat what is apparently the same segment differently in different people?   

Is this significant? While R. M., B. D. and I all appear to have the same number of matches in our shared segment, over the whole genome R. M. matches 1638 segments and I match 1343 segments, while B. D. matches only 442 (this is at gedmatch).

Could having relatively few matches overall affect the way TIMBER treats a particular match?  After the rollout of version 2 I did note several African-Americans saying they had lost mostly European matches.

Thursday, April 11, 2013

The X Chromosome

The previous post described the inheritance of mitochondrial DNA, Y-DNA, and autosomal DNA (Chromosomes 1 through 22), but left out X inheritance, an interesting special case.

You probably know that men carry an X chromosome and a Y chromosome, and women carry two X's.   These are the chromosomes that determine whether you're a man or a woman.    

Women get an X from each parent, recombine them, and pass them on to sons and daughters.   Men get only one, from their mothers, and pass that one chromosome on intact to their daughters.   That causes an interesting pattern of inheritance, shown in charts here from Blaine T. Blettinger's blog, The Genetic Genealogist .

The chart below shows which ancestors can contribute to a male's X chromosome.

This chart shows which ancestors can contribute to a female's X chromosome.

X ancestors include the ancestors who contributed your mitochondrial line, but they also include many more, in the case of females even including some ancestors from the father's side.   One way of thinking of X inheritance is that it can't involve two males in a row in the inheritance line, since X's no man gets an X from his father - he gets a Y.

Because of the special case of fathers passing on the X they inherited from their mothers intact, females have one X that came directly from their paternal grandmothers.    So, in contrast to autosomal DNA, which is divided and recombined every generation, X DNA is divided and recombined only when inherited from a female.   This makes it hard to estimate how many generations back a given size match on the X is.

Tuesday, April 9, 2013

Haplogroups, Autosomal DNA and DNA Relatives

Autosomal DNA determines your DNA Relatives at 23andMe, and comes from the 22 pairs of autosomal chromosomes that everyone inherits:  22 from your mother, and a matching set of 22 from your father.   Your mother and father sliced and diced the 22 pairs they inherited from their mothers and fathers to come up with the chromosomes they gave you.

Autosomal DNA is divided and recombined in an unpredictable way every generation.  More than 99% of your 2nd cousins (the great grandchildren of your great grandparents) will share enough autosomal DNA with you to detect, but less than 5% of your 6th cousins, the descendants of your GGGGG grandparents, will share enough autosomal DNA with you to detect.   You have thousands of sixth cousins, though, so even 5% of them is quite a few.  Some of your DNA matches may share a common ancestor even further back, since some segments can, through random chance, escape the axe and survive intact for many generations.

I thought my haplogroups would identify my relatives.   What's with all these other haplogroups in my DNA Relatives list?

Although the most important part of our 23andme results are the autosomal results outlined above - those are the results used to identify our DNA Relatives - 23andme also includes our maternal line and paternal line haplogroups, found in our mitochondria, for the maternal line, and Y-DNA, for the paternal line.

Mitochondria are found in every cell, outside the nucleus, and are inherited along the maternal line. Your mitochondrial DNA is inherited with very little change from your mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's mother's (you get the picture) mother.

Only relatives in that direct maternal line share your mitochondrial haplogroup, unless they happened to inherit the same haplogroup by chance.  So most of your relatives don't share your mitochondrial maternal line haplogroup, since they don't have the same female-to-female line.

Y-DNA is found on the Y-Chromosome, which is the chromosome that makes a male a male, so females don't have it.

Men inherit Y-DNA from their father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's father's (etc) father.   Only the men in that direct line share that haplogroup, unless they happened to inherit the same haplogroup by chance.   A female can find her father's Y-DNA haplogroup by testing her father, brother, father's brother, father's brother's son, or other male relative in the direct paternal male-to-male line.  Most of your relatives don't share your paternal Y-DNA haplogroup, because they don't share your direct male-to-male line.

And that is why DNA Relatives is your gateway to finding relatives at 23andMe.   It can include cousins from every part of your family tree, not just the narrow lines that share your haplogroups.