This story is a continuation of my discssion on usng autosomal DNA test results to find living distance cousins. The story focuses on five “extended family” autosomal DNA (atDNA) matches from AncestryDNA. These are genetic matches between me and individuals who are living second or third cousins.
The autosomal (atDNA) match results from AncestryDNA indicated five potential matches that are defined as ‘extended family‘. All five genetic matches are cousins on my maternal side of the family. Three of the extended family matches were documented by information provided by the DNA testers through their family tree information. The traditional genealogical information helped with locating our common ancestors and substantiating the family relationship. I was able to document the common ancestors for the other two matches through documentation in my family tree. As stated, all of these matches are newly discovered living relatives on my mother’s side of the family.
My ‘Extended Family’: AncestryDNA Matches on My Mother’s Maternal Side
Photograph of Margaret, Pierce and Theresa Hart
Click for Larger View | Source: Family Collection, circa 1939 Starting at the top of the ladder: Margaret Hart, Pierce Hart and Theresa Schuh (Hart)
My mother was estranged from her mother, Theresa Schuh, before she was 10. Her older brother of three years, Pierce, was severely intellectually disabled. [1] When Theresa left the family, my grandfather was unable to take care of Pierce. Pierce was institutionalized for the remainder of his life. [2] My mother was shuttled between various ‘foster’ homes and finally stayed with my grandfather’s sister, Mae Whalen. My mother provided little information about her mother or the Schuh side of the family when I was growing up.
Information about my mother’s paternal side of the family was also limited. My grandfather, Frederick Hart, worked in a Ford assembly line plant in Troy, New York for much of his career. He along with his parents, five older brothers and a sister spent most of their lives in the Troy, New York area. [3]
I was, however, fortunate to grow up with and know my grandfather Hart, his sister and three of his five brothers. I have wonderful memories of them, suitable for future stories. However, little was said about the generations before them.
I have been able to piece together much of the history of the Hart family after the passing of Fred Hart’s generation. My mother’s story and stories of his family will be found in future family stories.
Subsequent to my mother’s death, I have learned a bit more about my maternal grandmother’s family. However, I still have many gaps in knowledge on that side of the family. [4]
Having atDNA matches that are from my maternal grandmother’s side of the family is noteworthy. Finding out about my maternal grandmother’s family is akin to what perhaps those who are adopted experience when discovering information about their biological family.
Illustration one lists the five ‘extended family’ matches in AncestryDNA. All matches are above my initial 90 cM filtering threshold for reviewing autosomal matches.
Illustration One: Extended Family AncestryDNA Match Results
Three of the five matches can be traced from my mother’s maternal side of the family. The other two matches have common ancestors from my mother’s paternal side of the family.
The common ancestors for all the matches on my mother’s paternal side of the family are Samuel and Honora (Hannah) Hart. They had two children, Francis Hart and Mary Ann Hart. Francis or Frank Hart was my great grandfather. Mary Ann Hart, Frank’s sister or my great grand-aunt, was married twice and has family descendants from each of the marriages. (See illustration two.)
Two of the five matches are traced through my mother’s paternal side of the family. One of the five matches can be traced through the Nash side of Mary Ann Hart’s descendants and another match can be traced through the Watson side of her descendants.
Illustration Two: Common Ancestors on the Paternal Side of Mother’s Family
The other three matches can be traced through my mother’s maternal side of the family: the Schuh family. The common atDNA ancestors are George and Dorothy Schuh and the common denominator for the atDNA matches are the two siblings Elizabeth and Teresa Schuh.
Illustration Three:Common Ancestors on the Maternal Side of Mother’s Family
(1) Kaja: 179 shared cMs – Second Cousin Once Removed
An individual named Kaja was listed as an extended family match with unweighted shared DNA of 179 cM across 10 segments. The longest cM segment we share is 31 cMs. [5]
If I were to compare this result with the results in the Shared cM Project database, the furthest back I might need to go to find common ancestors for a match of 179 cM is my 4th-Great-Grandparent level or the seventh generation on my pedigree chart. (See illustration four.)
Illustration Four: Estimated Relationships for cM Matches of 179 cM
The results from AncestryDNA indicated that Kaja provided limited information on her family tree. She only had listed her grandparents, Willam and Elizabeth Mittlested, on her mother’s side. I had the Mittlested’s documented in my mother’s family tree.
AncestryDNA deduced that we share common ancestors based on information found in each of our respective family trees. Combining our ‘Ancestry Public Member’s family tree’ information, it was confirmed that our common ancestors are my maternal great-grandparents and her maternal great great-grandparents: George Schuh (1867 – 1929) and Dorothy (Dora) Schutz (1869-1840). Illustration five depicts our family relationship . Kaja’s mother is my second cousin. I have not listed Kaja’s mother name since she is living. (See illustration five).
Illustration Five: Common Ancestors Between Second Cousins Once Removed – Maternal Grandmother’s Line
As indicated in illustration one above, AncestryDNA pegged William as an half second cousin. William did not provide much in terms of a family tree or other information about his family surnames. His publicly linked AncestryDNA family tree lists three private individuals. Obviously this does not help with researching our possible common ancestors.
In order to determine what our relationship is, based on the cM value of the DNA test, I needed traditional genealogical evidence to narrow the possible genealogical relationships that I may share with William. What gave me an initial clue to determine our common ancestor was his AncestryDNA ‘handle” or name he used for his AncestryDNA account. His AncestryDNA account name was his first name and last name!
Knowing William’s last name gave me a sufficient lead to find additional information to corroborate genetic clues with traditional genealogical sources of fact and evidence. A direct line of genetic descent through my maternal side started with our mutual great grandparents: George and Dorothy Schuh. (See illustration six.) His grandmother Elizabeth Schuh was my maternal grandmother’s sister, Theresa Schuh. His mother Elizabeth Mittelsted, my mother’s first cousin, married Felix Durivage Jr.. I do not know if Elizabeth Mittelsted and Felix Durivarge had more than one child but I am fairly confident that William Durivage is their son.
Illustration Six: My Connection with William (141 Shared cMs)
The third promising ‘extended family’ match was associated with “David D”. We share 122 cMs across 8 segments. The longest shared segment is 22 cMs.
Similar to William, David did not provide much in terms of a family tree. His publicly linked AncestryDNA family tree lists three private individuals, one of which is probably himself and the two other family members are probably his wife and child. Similar to the challenges faced with the ability to confirm William’s relationship, the lack of information provided by David did not help in isolating a common ancestor.
Applying 122 cMs in the Shared cM Project tool indicates that the furthest back I might need to go to find common ancestors for a genetic match with David is the 4th-Great-Grandparent level or generation seventh generation on my pedigree chart. (See illustration seven.).
Illustration Seven: Possible relationships with a 122 cM Match
As reflected in illustration seven, based on 122 cMs, David conceivably could be one of many relationships to me. Depending on who our common ancestors are, we could be one of the following relations, based on relationship probabilities. [6] The probability that David is any of these relations to me is presented in table one. Based on estimated probabilities of our relationship found in table one, I initially believed there was a 29 percent chance that he was a third cousin or a 12 percent chance he was a second cousin.
Table One: Relationship Probabilities Based on 122 Shared cM
Source: Inputting the cM value of 112 into the Shared cM Project Tool found at Bettinger, Blaine,, The Shared cM Project 4.0 Tool v4, Mar 2020, DNA Painter, https://dnapainter.com/tools/sharedcmv4
In order to determine what our relationship is based on the cM value of the DNA test, I needed traditional genealogical evidence to narrow the possible genealogical relationships to one that I may possibly share with David. Similar to William, what gave me an initial clue to determine our common ancestor was his AncestryDNA ‘handle” he used for his AncestryDNA account. His AncestryDNA account name was, like his first cousin, his first name and last name! Knowing David’s last name gave me enough information to find information to corroborate genetic clues with traditional genealogical sources of fact and evidence.
A review of my current information and facts associated with the maternal side of my mother’s family led me to an obituary for Marie (Mittlested) Decker. Marie was one of my mother’s first cousins. Marie Decker’s obituary provided information on her children, her grandchildren and her siblings. David is one of Marie’s four children. [5]
David D is my second cousin, as depicted in illustration eight. Our common ancestors are our great-grand parents.
The connections between me and my two maternal second cousins and my second cousin once removed can be viewed in the following family tree in illustration nine.
Illustration Nine: The Mittelsted Family Connection in Autosomal DNA Testing
(4) James: 113 Shared cMs – Third Cousin on the Mother’s ‘Paternal side’
The fourth promising ‘extended family’ match was associated with “James E”. We share 113 cMs across 7 segments. The unweighted shared DNA is 113 cM and longest segment is 29 cM. AncestryDNA infers that James is possibly my 3rd cousin.
James’ information on his AncestryDNA public family tree is limited. It only goes back to his grandparents. However, based on the names of his grandparents, I was able to trace his ancestral line back to our common ancestor. I had developed an extensive family tree that established a connection with his maternal grandfather’s family line with my mother’s paternal side of the family tree.
His line of descent goes through the Nash side of Mary Ann Hart’s lineage.
According to the information found in AncestryDNA member trees, James and I have two common ancestors that make us third cousins. We share great great-grandparents: Samuel Justus Hart and Honora Rouen. (See illustration ten.)
Illustration Ten: Discovered Third Cousin– Maternal Grandfather’s Line
‘Sharburl’ and I share 113 cMs across 7 segments of DNA. Our unweighted shared DNA is also 113 cM and the longest segment is 29 cM. As reflected in illustration ten,
Similar to James, the fifth listed atDNA ‘extended family match is also a third cousin that is traced through our common great great-grandparents Samuel and Honora Hart. However, our line of descent goes through Marry Ann Hart’s Watson family line.
Illustration Ten: Discovered Third Cousin –Maternal Grandfather’s Line
For my personal experience, the use of atDNA tests have provided a promising and fruitful approach to identify information and facts for relatives that are both living and deceased. The test results have definitely helped in my efforts to obtain knowledge of a part of my mother’s family.
My initial filtering criteria involved a two step process. The first step involved the use of a cut off value for cMs shared at 90. The second criteria was focusing on matches that provided family tree information.
My next phase in researching the atDNA results from AncestryDNA will involve looking at the 760 matches that are between 16 and 89 cMs. That is a lot of matches! In the second phase, I will focus on matches that have family tree information associated with their DNA results and where common ancestors are identified in their family tree information.
Sources
Feature Image: An amalgam of three different images: (1) The network image of individuals is from Woodbury, Paul, “Who Is This?” 6 Steps to Determine Genetic Relationships of Your DNA Matches, April 2019, LegacyTree Genealogists, https://www.legacytree.com/blog/determine-genetic-relationships . (2) The Colorized chart from Johny Perl, Introducting the updated shared cM tool, 27 Mar 2020, DNA Painter Blog, https://dnapainter.com/blog/introducing-the-updated-shared-cm-tool/ and (3) A Diagram depicting the relationship of a third cousin from 23andMe.
[1] There are the four main levels of intellectual disability (ID), categorized by IQ ranges and functional abilities. Contemporary diagnosis of intellectual disability does not rely solely on IQ scores but also considers adaptive functioning across three domains: conceptual, social, and practical life skills. The level of support needed in these areas is now considered as important as IQ in determining the severity of intellectual disability.
In the 1930’s and 1940’s, individuals who could not rely on extensive support from family members were often placed in state institutions. Many families were forced to institutionalize their children due to: financial inability to provide care, lack of community services, and limited educational opportunities.
Committee to Evaluate the Supplemental Security Income Disability Program for Children with Mental Disorders; Board on the Health of Select Populations; Board on Children, Youth, and Families; Institute of Medicine; Division of Behavioral and Social Sciences and Education; The National Academies of Sciences, Engineering, and Medicine; Boat TF, Wu JT, editors. Mental Disorders and Disabilities Among Low-Income Children. Washington (DC): National Academies Press (US); 2015 Oct 28. 9, Clinical Characteristics of Intellectual Disabilities. Available from: https://www.ncbi.nlm.nih.gov/books/NBK332877/
Intellectual disability, Wikipedia, This page was last edited on 11 November 2024, Intellectual_disability
[2] Rome State School, 1940 United States Federal Census, New York, Oneida County, Rome, Page 13B, Line 67
Click for Larger View | Source: Rome State School, 1940 United States Federal Census, New York, Oneida County, Rome, Page 13B, Line 67
[3] The Ford plant on Green Island was an assembly plant that primarily manufactured radiators and heater cores for Ford vehicles. The factory was located on the northern end of Green Island near Troy, New York, along the Hudson River.
The plant opened in 1923 and included both a manufacturing facility and a hydroelectric station to power its operations1. While Ford initially considered using the facility to manufacture tractors and other farming implements, by 1923 the plant was producing radiators, ring gears, and springs.
The plant was located on the north end of Green Island, on 154 acres of former Tibbits family property. The main building of the plant was nearly 1,500 feet long, parallel to the Hudson River. Initially the plant employed 1,000 workers, reaching approximately 1,500 by 1973. The factory closed in 1988 and was demolished in 2004. The hydroelectric plant, which was separate from the manufacturing facility, still remains active today under the Green Island Power Authority.
[4] In the course of conducting genealogical research on the Hart family in the last six years, I met a third cousin, Janes Scribner McCrary. Our common ancestor is Samuel and Honera Hart. We have shared facts and evidence on Samuel and Honora Hart. Jane has also assisted my efforts to get a cleared picture of Francis Hart. She had compiled and impressive blog of stories of her family at https://selectedstoriesofmyancestors.blogspot.com
See Jane’s informative stories on the Hart family at:
[5] The unweighted DNA value represents the raw amount of shared genetic material measured in centimorgans (cM) before any algorithmic adjustments by AncestryDNA.
Timber is AncestryDNA’s proprietary algorithm designed to filter DNA matches for more accurate genealogical connections. Timber analyzes DNA segments across AncestryDNA’s entire database to identify and filter out segments that are likely shared due to distant population-level ancestry rather than recent common ancestors. It essentially “chops down” DNA segments that don’t help identify recent family connections.
I completed a variety of autosomal (atDNA) tests from various genealogy companies to determine if I could find unknown living genetic relatives and to provide additional information on our common ancestors.
Autosomal DNA testing can accurately trace ancestry back approximately 5-6 generations and provide information on living relatives. These tests can provide added information and facts on completing family trees for generations that contain both living and deceased relatives and the younger generations that virtually contain all living relatives.
There are different strategies for comparing DNA results from different companies. [1] I simply took atDNA tests from each of the sampled companies to compare differences in estimating ethnicity descent results and to determine if here were any matches with unknown living genetic relatives in each of their respective DNA databases.
I have completed atDNA tests with the following DNA companies: ancestry.com (AncestryDNA), 23andMe, LivingDNA, FamilyTreeDNA, and CriGenetics. [2] The most promising results were obtained from AncestryDNA and 23andMe.
As indicated in the first part of this story, the DNA companies analyze hundreds of thousands of single nucleotide polymorphisms (SNPs) across the 22 autosomal chromosomes. These SNPs are specific locations in the genome where genetic variation commonly occurs between individuals. The companies use different proprietary algorithms and imputation methods to analyze the DNA. The DNA match results can vary between companies based on the software and micro chip technology utilized, the selection and number of SNPs tested, and the size of company’s database .
The atDNA tests may result in hundreds of ‘matches’. Each of the companies compare the SNP data between two individuals to identify segments of DNA that appear to be identical. These matching segments indicate DNA that are likely to be inherited from a common ancestor. [3]
The length of matching SNP segments is measured in centimorgans (cM). (See illustration two below.) [4] The amount of shared cMs is compared to expected ranges for different relationships to predict how two people may be related. Close relationships like parent/child or full siblings have very distinct amounts of shared DNA, while more distant relationships have overlapping ranges or shared DNA or cMs.
Where the individual being tested, such as myself, has a number of shared SNPs and a number of consecutive SNPs in common with another tested individual in the company’s database, it can be inferred that we share a segment of DNA at that part of our respective genomes. If the segment is longer than a threshold amount set by the testing company, then we are considered to be a match. [5]
Clarifying Relationships: The Number of Shared cMs and Length of cMs
Two central or key statistics for interpreting atDNA matches are the number of shared cM segments and the length of the shared cM segments(or the percentage of the genome that is shared). When looking at atDNA comparisons, the “shared cM” is the total length of the DNA one shares with another person. The “shared segments” are how many blocks the matching atDNA is broken into. Longer block segments generally indicate a closer relationship. Companies set minimum thresholds for segment length and total shared DNA to be considered a match. [6]
To put the concept of cMs in perspective, an individual typically receives approximately 3,400-3,700 cMs from each biological parent. Each person has about 6,800 cMs in total. [7]
As indicated in the first part of this story, the random nature of genetic inheritance leads to variability in how much DNA is shared through successive generations. This is known as variable expressivity. [8] When moving through successive generations, the amount of DNA inherited from each ancestor roughly halves with each generation, meaning you inherit approximately 50% from your parents, 25% from each grandparent, 12.5% from each great-grandparent, and so on. As you move further back in generations, the percentage of DNA inherited from each ancestor decreases exponentially.
The process of genetic recombination during meiosis ensures that the exact DNA segments inherited from each parent are randomized, leading to variation even between siblings in the next generation. [9] For example, full siblings may share anywhere from about 35% to 65% of their DNA or 1613 cMs to 3488 cMs. First cousins typically share around 12.5% of their DNA or about 866 cMs but the actual range can vary significantly between 396 to 1397 cMs. This variability increases with more distant relationships, making it harder to precisely determine the degree of relatedness based solely on shared atDNA percentages.
“(S)ome genuine genealogical third cousins will not show up as a match because, although both cousins will have inherited DNA from their common ancestors they do not share any overlapping DNA segments. Beyond about five or six generations we will have some genealogical ancestors with whom we share no DNA. They are our genealogical ancestors but not our genetic ancestors. If we go back 450 years we have over 32,000 genealogical ancestors but we will have DNA from only around 1000 of these ancestors.” [10]
The diminishing amount of cMs shared with relatives can be depicted in charts that display information using centimorgans. A notable “shared cM chart,” which is used to visualize the likelihood of a relationship based on the amount of shared DNA measured, is called the “The Shared cM Project 4.0 tool chart”. It can be accessed through the DNA Painter platform. [11] (See illustration three.)
Illustration Three: Chart of Matching Relationships and Their Respective Average cM and Ranges of cMsto an atDNA Tester
The Shared cM Project 4.0 tool chart graphically illustrates the range of cMs and overlap that are associated with specific genetic relationships. The chart provides information on the average cMs and the range of cM values from low to high for each genetic relationship. The values are derived from a collaborative data collection and analysis project created to understand the ranges of shared cMs associated with various known relationships. Version four of this project is based on shared cM data for nearly 60,000 known relationships. [12]
The Shared cM Project is a collaborative data collection and analysis project that helps genealogists understand potential relationships between DNA matches based on shared genetic material. The project was created by Blaine Bettinger to understand the ranges of shared centimorgans (cM) associated with various known family relationships. The table one below indicates the range of cMs that are associated with ten groupings of relationships that were part of the Shared cM Project. [13] The table provides the number of samples in each group and the average value, min and max values, the standard deviation (which can be used to determine the frequency curve of the values), and the expected cM value for each group.
Table One: Centimorgan Groupings by Relationship in the Shared cM Project
The following example (illustration four) depicts the variability of inheriting autosomal DNA and the utility of using the number of shared cM segments and the length of the shared cM segments to determine possible relationships. Illustration four below shows 22 lines that symbolize the 22 autosomal chromosomes that were inherited from one parent for a fictitious person. The darker color in each of the lines represents the segments of cMs that were inherited from the parent. In this example, the person has inherited 3524 cMs and 22 segments from one parent. The individual has inherited 22 segments in 22 chromosomes. They are really long segments. In fact, 16 of the 22 segments reflect entire chromosomes. In addition, the child inherited 3524 cMs, well within the cM range of 2400 – 3700 cMs for a parent – child relationship, as depicted in table one.
Illustration Four: Example of Inherited cMs from One Parent
Illustration five below is an example of a grandparent – grandchild match for the same person (the grandchild). In the example, the number of shared segments is very similar to the number shared segments found in illustration three with the parent. In fact there is one more segment than that found in the parent relationship. However, the length of the segments are all shorter than the segments shared between parent and child. In addition, the total number of cMs are significantly less than that found with the parent.
Illustration Five: Example of Inherited cMs from One Grandparent
Click for Larger View | Source: Modified version of an illustration found in Nguyen, Tiffany, What’s the difference between shared centimorgans and shared segments?, 11 Nov 2019, The Tech Interactive, https://www.thetech.org/ask-a-geneticist/articles/2019/centimorgans-vs-shared-segments/
The total number of centimorgans that an atDNA tester shares with a match is typically a good indicator of how closely related they are to someone else. However, shared cMs cannot always indicate exactly how that tester is related to a match. Different types of relationships may share similar amounts of DNA.
In the grandparent – grandchild example above, the furthest back you might need to go to find common ancestors for a match of 1518cM is the grandparent level. If you are uncertain of the actual relationship of the match and limited to just reading the cM results of a test, the connection may be closer. Depending on the tester’s family, this match could be a close younger generation relative, such as the descendant of the tester’s sibling if he or she had one or more siblings.
Based on the ranges of cMs associated with specific family relationships, 1518 cMs could point to a grandparent, aunt or uncle, a half sibling, a niece or nephew or a grandchild. This is illustrated by plugging in the 1518 cM value in the Shared cM Project 4.0 tool. (See illustration six).
Illustration Six: Expected Relationships Based on Shared cMs
Click for Larger View | Source: Johnny Perl, Introducing the updated shared cM tool, 27 Mar 2020, DNA Painter Blog, https://dnapainter.com/blog/introducing-the-updated-shared-cm-tool/
Comparison of Results from Different DNA Companies
As indicated in the first part of this story, my experience with atDNA tests have largely resulted in the initial discovery of many living second and third generational cousins. This fact alone is amazing. However, many of the matches fail to document their respective lines of descent. The lack of this additional genealogical information makes it difficult to document where our common distant family connections are located.
The ability to analyze and comprehend the sheer number of autosomal DNA matches from the various DNA tests is daunting. As reflected in table two, there are thousands of ‘possible’ matches based on my atDNA results provided by four DNA companies..
Table Two: Number of atDNA Matches for My Test Results by DNA Company
DNA Company
Number of Genetic Matches with Shared DNA
Maximin cM Match Value
Minimum cM Match Value
Ancestry DNA
17,118
3479
8
23andMe
1,500
1586 [14]
16
FamilyTreeDNA
3,639
49
7
LivingDNA
302
43
11
Testing companies compare your DNA profile to their database of test results from other customers. They search for other individuals who share significant DNA segments with you. Each company has specific threshold values associated with cMs and cM length. (See table four in the first part of this story.) [15] The size and pattern of shared DNA is used to predict the likely relationship. Relationships are typically detected confidently for third cousins or closer, with decreasing probability for more distant relations.
Depending on the DNA company, as reflected in table three, the predicted relationship is depicted by different measures: the total percentage of shared DNA, the number of shared segments, the length of the shared segments, the longest block of cMs. Different companies may also provide slightly different relationship estimates due to variations in their testing algorithms and reference databases. Despite the different presentation of match results, comparing results between different compamies is possible.
Table Three: Type of Matching Results Provided by DNA Company
DNA Company
Total cMs
Percent of cMs Shared
Number of Shared Segments
Longest cM Block
AncestryDNA
X
X
23andMe
X
X
FamilyTreeDNA
X
X
LivingDNA
X
X
X
Separating the Wheat from the Chaff
DNA MatchFiltering Criteria
It would be very tedious to comb through all of the matches identified by the four DNA companies. I also believe the majority of match results will not yield many true matches. There is a need to utilize filtering criteria to reduce the large number of potential matches to a manageable number and focus on matches that may lead to legitimate family relationships.
Based on my review of research strategies associated with autosomal testing, I have employed two criteria in recursive order. The first involves establishing a cut off value for cMs shared. The second criteria is focusing on matches that provide family tree information.
Many researches have documented that it was less likely of discovering a kinship relationship with match cases that shared less than 15 cMs. [16]
“While it may seem at first that all shared segments of DNA could constitute genealogical evidence, unfortunately some small segments are … creating “false positive” matches for reasons other than recent ancestry. These segments sometimes match because of lack of phasing, phasing errors, or a variety of other reasons. One thing, however, is clear: there is no debate in the genetic genealogy community that many small segments are false positive matches. There IS debate, however, regarding the rate of false positive matches, and what that means for the use of small segments as genealogical evidence.“ [17]
Most people are unable to trace all of their ancestral lines back ten generations or beyond. The common ancestral couple cannot therefore be identified through atDNA testing. Even if a shared ancestral couple can be identified, without tracing all the other ancestral lines in your family tree you cannot eliminate the possibility of shared ancestry on other as yet undocumented lines. [18]
Many matches under 15 cMs will share ancestry more than five or more generations ago and will be mostly beyond the reach of genealogical records. Therefore, I have summarily eliminated matches that are 15 cMs or less. Based on estimates derived from the Shared cM Project Tool, a match of 15cM has a wide range of remote, possible relationships. (See illustration sseven.) The connection may be within 4th-Great-Grandparent level, but the common ancestors could also be more generations back. [19]
illustration Seven: Expected Relationships Based on 15 Shared cMsor Less
Click for Larger View | Source: Johnny Perl, Introducing the updated shared cM tool, 27 Mar 2020, DNA Painter Blog, https://dnapainter.com/blog/introducing-the-updated-shared-cm-tool/
There are other professional genealogists, such as Blaine Bettinger, that lower the bar to above 5 cMs.
“(T)he very first thing I do with the spreadsheet is sort it by size and delete every segment smaller than 5 cM. For me, there is too much uncertainty surrounding small segments to base any conclusion on them.” [20]
“I call small segments (which I usually classify as 5 cM or less) as POISON because it is currently impossible to decipher between which are real segments and which are not.” [21]
According to population genetics theory all individuals have common ancestry in the distant past, and we all have what are called short, old IBD segments in common. [22] Identity By Descent (IBD) segments are stretches of DNA that two or more individuals have inherited from a common ancestor without any recombination occurring within that segment. The probability of an IBD segment persisting decreases exponentially over generations due to recombination. [23]
In general, the larger the shared segments the more likely that the match is valid. Valid segments under 7 cM cannot be reliably detected with the currently available genetic genealogy tests. [24]
For the purposes of genetic genealogy the focus is on detecting large IBD segments within a genealogical timeframe (effectively within the last ten generations) where there is a possibility of identifying the common ancestor through documentary records. In general terms, the larger the segment the closer the relationship, but the frequency of the segment also needs to be taken into account. High-frequency IBD segments are more likely to be a signal of distant sharing at the population level whereas a segment that is only observed in two independently sampled individuals is more likely to be IBD. [25]
“Although there is only a low chance of sharing enough DNA with a specific distant cousin for the relationship to be detected, we have a large number of distant cousins and so many of these more distant cousins will appear in our match lists.” [26]
The following table four shows the expected number of cousins at different degrees of relationship, the expected amount of cMs for each degree of cousin and the percent of detecting the relationship. [27]. The trend indicates an high probability or even chance of discovering first through fourth cousins. This would imply matches above 60 cM are less likely to be influenced by false positive matches and matches between 15 and 60 have a ‘fifty-fifty’ chance of being real or reflect the back ground noise of distance common ancestors.
Table Four: Chance of Detecting Cousins
Degree of Cousin
Expected Amount of IBD (cM)
Percent Chance of Detecting
First
900
100%
Second
225
100%
Third
56
88.7
Fourth
14
45.9
Fifth
3.5
14.9
Sixth
0.88
4.1
Secventh
0.22
1.1
Eighth
0.055
0
Nineth
0.014
0
Tenth
0.0034
0
Source: Henn, Brenna M, Hon L, Macpherson JM, Eriksson N, Saxonov S, Pe’er I, et al. (Apr 3 2012) Cryptic Distant Relatives Are Common in Both Isolated and Cosmopolitan Genetic Samples. PLoS ONE 7(4): e34267. https://doi.org/10.1371/journal.pone.0034267
As reflected in table five, my match results from the four DNA companies have isolated 3,047 matches above the 15 cM level. This number of potential matches is still unreasonable to comb through and assess their fruitfulness of establishing a genetic relationship. I considered initially tightening the cM limit to 90 cMs.
Table Five: Number of Genetic Matches Based on cMs
Company
10 cM & Less
15-11 cm
15-89
Greater than 90
AncestryDNA
9850
6515
760
9
23andMe
0
0
1496
4
FamilyTreeDNA
1185
1939
515
0
LivingDNA
4
35
263
0
Total
11039
8489
3034
13
The furthest back you might need to go to find common ancestors for a match of 90cM is 4th-Great-Grandparent level or generation 7 on your pedigree chart. [28]
Once I have assessed the fruitfulness of establishing a kinship relation with matches above 90 cMs, I figure I can look at the other matches at successive lower cM levels.
The second filtering criteria that I have used is the availability of family tree information found with specific matches over 90 cMs. As indicated in table six below, three of the four DNA companies that I have completed atDNA tests provide the ability or option to either add family tree information or, at the minimum, family surnames.
Table Six: The Option to Add Family Tree or Surname Information with atDNA Results
Company
Family Tree Infor- mation
Family Sur- names Infor- mation
Parents’s Birth- place
Grand- parents’ Birth- place
Other ancestors’ birth- place
Identify Common Ancestor
AncestryDNA
Yes
Yes
*
*
*
Yes
23andMe
Yes
Yes
Yes
Yes
Yes
FamilyTree DNA
Yes
Yes
*
*
*
Yes
LivingDNA
No
No
No
No
No
No
* This information can be found if a tester provides information on their family tree. It is not a separate category of information .
My AncestryDNA and 23and Me Matches: Close Family
The largest group of match results and the largest number of confirmed matches is from AncestryDNA. As of October 2024, the atDNA results in the AncestryDNA data base identified 854 possible matches on both sides (paternal and maternal) of my family. Of those 854, there are four DNA testers that are identified as ”Close Family”. An additional five are identified as “Extended Family”. The remaining 845 matches are categorized as “Distant Family”. [29]
In the early 2000’s, based on my suggestions, my father and paternal aunt completed autosomal tests. More recently, my aunt’s grandson, also completed an AncestryDNA test. As depicted in illustration seven, their results confirm their respective genetic ties to me.
The DNA results for my father confirmed the accuracy of the DNA testing process. The range of matching cMs associated with my match with my father corresponds with the cM range of parent/child. [30] The match results for my paternal aunt also confirm the accuracy of the atDNA match for aunt/uncle. (See illustration nine.)
The results for James were also spot on as my first cousin once removed (the son of one of my first cousins).
Illustration Ten: Average cMs for a First Cousin Once Removed
Source: Inputting the cM value of 359 into the Shared cM Project Tool found at Bettinger, Blaine,, The Shared cM Project 4.0 Tool v4, Mar 2020, DNA Painter, https://dnapainter.com/tools/sharedcmv4
The filtering match displays for AncestryDNA and 23andMe are a bit different. While AncestryDNA provides total shared cMs and the percent of shared DNA, 23andMe provides percent of DNA shared and number of shared segments. There are two ways to convert 23andMe matches to centimorgans (cMs). You can use the Shared cM Project tool at DNA Painter and enter the percentage of shared DNA in the percentage box on the online tool. The tool will show you the number of cMs. [31]
If I use the 90 cM are an initial cut off for viewing matches, I need to convert percent shared cMs to an estimate of the amount of cMs shared.
There were two 23andMe matches that were above the 90 cM filtering criteria. The first was Christopher G who shared an estimated 641 cMs or 8.62 percent of shared cMs with me. As indicated in illustration eleven, 23andMe predicted that Christopher is my first cousin once removed. The second match was with Alexandra G. who shared an estimated 494 cMs or 6.64 percent of shared cMs with me. 23andMe predicted that Alexandra is my second cousin.
Utilizing the Shared cM Project tool for Christopher’s 8.62 percent shared cMs and Alexandra’s 6.64 percent indicates that the furthest back I might need to go to find common ancestors for each match is my 2nd-Great-Grandparent level or generation 5 on my pedigree chart. For Christopher.
As indicated in table seven, the inputted cM values in the Shared cM Project Tool also predict that there is a57 percent chance that Chritopher is my first cousin once removed. The inputted cM values in the Shared cM Project Tool predict that there i a 90 percent chance that Alexandra is my first cousin once removed, contrary to 23andMe’s prediction.
Table Seven: Probability of Expected cM Relationship based on Percentage of Shared DNA
Source: The Shared cM Project Tool found at Bettinger, Blaine,, The Shared cM Project 4.0 Tool v4, Mar 2020, DNA Painter, https://dnapainter.com/tools/sharedcmv4
Both Christopher and Alexandra are son and daughter respectively of one of my paternal cousins. As such, they are both first cousins once removed.
Continuation
A continuation of my discussion of atDNA matches continues in the next part of “Ongoing Discoveries from Autosomal DNA Tests”. The continuation will discuss AncestryDNA “extended family” matches and 23andMe match results.
Sources
Feature Image: An amalgam of three different images: (1) The network image of individuals is from Woodbury, Paul, “Who Is This?” 6 Steps to Determine Genetic Relationships of Your DNA Matches, April 2019, LegacyTree Genealogists, https://www.legacytree.com/blog/determine-genetic-relationships . (2) The Colorized chart from Johny Perl, Introducting the updated shared cM tool, 27 Mar 2020, DNA Painter Blog, https://dnapainter.com/blog/introducing-the-updated-shared-cm-tool/ and (3) A Diagram depicting the relationship of a third cousin from 23andMe..
[2] With the exception of CriGenetics, each of the companies provide atDNA matches that correspond to family relationships within six geneations. I did not purchase the ‘premium’ level of testing with CriGenetics which included possible DNA matches from their database.
Pereira, Rita, Pietro Biroli, Stephanie Von Hinke, Hans Van Kippersluis, Titus Galama, Niels Rietveld, and Kevin Thom. 2022. “Gene-environment Interplay in the Social Sciences.” OSF Preprints. March 4. doi:10.31219/osf.io/d96z3
[14] 23andMe provides their match results in two forms: percentage of DNA shared and the number of cM segments. There are two ways to convert 23andMe matches to centimorgans (cM), you can use the Shared cM Project tool at DNA Painter:
Go to the Shared cM Project tool at DNA Painter
Enter the percentage of shared DNA in the percentage box
[17] One source indicates that any match with 15 cM has a 60 percent chance of being a sixth cousin, sixth cousin once removed, fifth cousin, sixth cousin twice removed, fourth cousin once removed, fifth cousin once removed, seventh cousin, half third cousin twice removed, fourth cousin twice removed, fifth cousin twice removed, seventh cousin once removed, third cousin three times removed, fourth cousin three times removed, fifth cousin three times removed, eighth cousin, and more distance relationships.
Another source indicates that there is a remote change of finding a kinship tie with a match at 14 cM or less.
See: Henn, Brenna M, Hon L, Macpherson JM, Eriksson N, Saxonov S, Pe’er I, et al. (Apr 3 2012) Cryptic Distant Relatives Are Common in Both Isolated and Cosmopolitan Genetic Samples. PLoS ONE 7(4): e34267. https://doi.org/10.1371/journal.pone.0034267
Browning SR, Browning BL. Identity by descent between distant relatives: detection and applications. Annu Rev Genet. 2012;46:617-33. doi: 10.1146/annurev-genet-110711-155534. Epub 2012 Sep 17. PMID: 22994355. https://pubmed.ncbi.nlm.nih.gov/22994355/
Bettinger, Blaine, Small Matching Segments – Friend or Foe?, The Genetic Genealogist, https://thegeneticgenealogist.com/2014/12/02/small-matching-segments-friend-foe/
[20] Bettinger, Blaine, Small Matching Segments – Friend or Foe?, The Genetic Genealogist, https://thegeneticgenealogist.com/2014/12/02/small-matching-segments-friend-foe/
[22] “In genetic genealogy, the identical ancestors point (IAP), or all common ancestors (ACA) point, or genetic isopoint, is the most recent point in a given population’s past such that each individual alive at that point either has no living descendants, or is the ancestor of every individual alive in the present. This point lies further in the past than the population’s most recent common ancestor(MRCA).“
“We find that a pair of modern Europeans living in neighboring populations share around 2 12 genetic common ancestors from the last 1,500 years, and upwards of 100 genetic ancestors from the previous 1,000 years. These numbers drop off exponentially with geographic distance, but since these genetic ancestors are a tiny fraction of common genealogical ancestors, individuals from opposite ends of Europe are still expected to share millions of common genealogical ancestors over the last 1,000 years. There is also substantial regional variation in the number of shared genetic ancestors.“
Browning SR, Browning BL. Probabilistic Estimation of Identity by Descent Segment Endpoints and Detection of Recent Selection. Am J Hum Genet. 2020 Nov 5;107(5):895-910. doi: 10.1016/j.ajhg.2020.09.010. Epub 2020 Oct 13. PMID: 33053335; PMCID: PMC7553009. https://pmc.ncbi.nlm.nih.gov/articles/PMC7553009/
Ringbauer H, Huang Y, Akbari A, Mallick S, Patterson N, Reich D. ancIBD – Screening for identity by descent segments in human ancient DNA. bioRxiv [Preprint]. 2023 Mar 9:2023.03.08.531671. doi: 10.1101/2023.03.08.531671. Update in: Nat Genet. 2024 Jan;56(1):143-151. doi: 10.1038/s41588-023-01582-w. PMID: 36945531; PMCID: PMC10028887. https://pmc.ncbi.nlm.nih.gov/articles/PMC10028887
Nait Saada, J., Kalantzis, G., Shyr, D. et al. Identity-by-descent detection across 487,409 British samples reveals fine scale population structure and ultra-rare variant associations. Nat Commun11, 6130 (2020). https://doi.org/10.1038/s41467-020-19588-x
Ringbauer, H., Huang, Y., Akbari, A. et al. Accurate detection of identity-by-descent segments in human ancient DNA. Nat Genet56, 143–151 (2024). https://doi.org/10.1038/s41588-023-01582-w
[24] Doug Speed and David J. Balding, Relatedness in the post-genomic era:
Durand EY, Eriksson N, McLean CY. Reducing pervasive false positive identical-by-descent segments detected by large-scale pedigree analysis. Molecular Biology and Evolution advance access publication online 30 April 2014. https://watermark.silverchair.com/msu151.pdf
[25] Browning SR, Browning BL. Identity by descent between distant relatives: detection and applications. Annu Rev Genet. 2012;46:617-33. doi: 10.1146/annurev-genet-110711-155534. Epub 2012 Sep 17. PMID: 22994355. https://pubmed.ncbi.nlm.nih.gov/22994355/
[26] Cousin Statistics, International Society of Genetic Genealogy Wiki, This page was last edited on 27 February 2022, https://isogg.org/wiki/Cousin_statistics
[27] Henn, Brenna M, Hon L, Macpherson JM, Eriksson N, Saxonov S, Pe’er I, et al. (Apr 3 2012) Cryptic Distant Relatives Are Common in Both Isolated and Cosmopolitan Genetic Samples. PLoS ONE 7(4): e34267. https://doi.org/10.1371/journal.pone.0034267
[28] This is the result of inputting 90 cMs into the The Shared cM Project 4.0 tool v4. 90 cMs is approximately equal to 1.21 percent of the total % shared with a match.
[29] AncestryDNA has 7 different groups for predicted relationships: Parent/Child, Immediate Family (full siblings), Close Family (half siblings, grandparent/grandchild, aunt/uncle/niece/nephew), 1st cousin, 2nd cousin, 3rd cousin, 4th cousin, Distant Cousin.
