Finding and Sorting DNA Matches

Check Blood Type, and compare to Documents

In the last few weeks, I’ve seen several questions and discussions about how to identify parents when what is on a birth certificate doesn’t seem to be the actual case, or one has been adopted, or you have an orphaned or adopted grandparent, etc. We have had several situations causing ‘brick walls’ in both my husband’s and my families.

For the parents and children, first blood type is a very good clue. Several decades ago it was thought that there were either four or eight genes inherited by individuals, half from each parent, that cause blood type to be one type or another. We have now known for the past 15 years or so, that each person only inherits and has two genes that make a blood type, one gene from each parent. There are four major blood types: A, B, AB and O. O is recessive and a person with an O blood type has two O genes, one inherited from each parent. This also tells you that the parents had a blood type with an O in its two genes. The parents were either AO, BO, or OO. A and B blood types can be A or B with two combinations of genes: either AA or AO, and BB or BO. Any person who inherits an A and a B gene from his or her parents will always be AB. Most people know not only their own blood types, but their parents’ and sometimes their siblings as well. This is a big clue as to whether a father on a birth certificate is the actual father for those who are raised by their biological mothers. There are witnesses to births and so the mother of a child is always known. Unless a woman or girl lied on a birth certificate and the hospital, doctor, midwife or county didn’t ask for corroborating identification documents, or the person had false identification papers, then the birth certificate is most likely to be accurate for the mother.

Next, you need a good, well documented tree for what you DO know, for at least 6 generations, ideally 7 or 8, because there could be some cousin marriages skewing results. Generations are an average of 20-25 years, but can be greater in range depending upon whether or not a child was a younger child in a large family, meaning his parents were older than the average. This will skew generation numbers on one line or another. My parents were both the youngest in their respective families and my father was an unexpected “surprise.” His father was 48 years old when he was born and his mother was 36. Then my father didn’t marry until he was 23 and my mother was 25 and my mother was 27 when she had me. My father’s oldest brother was 18 years older than he was and I was four when I attended my first cousin’s wedding and he was in his 20’s.. This creates a generational offset for cousin relationships. My second cousins were contemporary with me in age, not my first cousins from my father’s brother’s line. Be sure you know which generation you are comparing for matches as that affects centiMorgan numbers. Many people in the DNA databases thoughtfully provide at least a year of birth and a general location (city or county and state) where they are. Both of those help in sorting out relationships and families. A recent study has shown that most people–some 60 to 70% of the U.S. population, as adults, always live within 100 miles of where they were born, or mostly raised, or else return to that area later in life. Most people also know something about their parents and where they claimed to be born and raised. Hopefully they have verified this by obtaining copies of birth certificates. Again this helps a lot.

So how do you use DNA testing to identify your father, and his family, and find cousins and more when you are not sure of the father, or have figured out your paper stated father isn’t.

First identify as many of your mother’s relations as possible, adults, that is, you don’t need to always know the names of minor children as they are less likely to have had a DNA test and be in a database and you don’t want to risk their privacy. This means having a good tree of your mother’s family, with up to eight degrees (at least six) of relationships. Here is why you need a minimum of six. You inherit 50% of your DNA from each parent, as you have two parents. From your four grandparents, you inherit 25% from each. Here is where it gets a little more difficult. From your great-grandparents, 8 persons, you inherit about 12.5% of your DNA each. However after two generations, the recombination that occurs within your mother to create you, can vary and slightly more or less can be inherited from individuals as there may be similarities in one line or another already from unknown distant cousin relationships that may have occurred in a distant past that skews what DNA, and the amount, are being recombined

It’s important to remember that until the Industrial revolution that only started about 180 to 190 years ago for even the best developed countries, and until there was inexpensive and more extensive mass transportation, like steamboats and railroads, most families lived in the same areas for CENTURIES and MARRIED COUSINS, unless displaced by war or disasters. Even with the latter, they would move only as far as needed to an area that was somewhat familiar and might still have cousins, and spoke the same dialect of the same language and had the same religion. So until the mid 1800’s, most families were inbred, cousin marriage after cousin marriage after cousin marriage for centuries. This DNA is going to be strong and will have a greater impact in the 20th and 21st century when parents marry or have relationships with others from the same ethnic groups who may have even come from the same small areas in Europe their own ancestors did.

If you were born before 2000, six generations back takes you to the beginning of the industrial revolution. At four generations back, you inherit 6.25% of your DNA–more or slightly less, from each of your 16, great-grandparents. At 5 generations back you will inherit only 3.125% of your DNA, again slightly more or less–can be as little as about 2%, from each of 32 great-great grandparents. At 6 generations, you inherit 1.625%, or less–as little as a very difficult to measure and identify trace amount from 64 great-great-great grandparents. Most DNA databases will tell you that admixture amounts, based on estimates from what they have in their databases that your DNA appears to match, of less than about 2% should be considered unreliable, not guaranteed accurate. In terms of numbers of centiMorgans, matches of less than about 7 or 8 centiMorgans should be considered carefully, especially if of all the matching segments, no single matching segment is more than about 5 centiMorgans. One exception to this, is if you have two segments on a chromosome of 3 cM or greater and which are separated by only a very thin band of non-matching. This could easily be a modern mutation within you, from a change in environment of your parents, exposure to chemicals or radiation, etc..

For the purposes of finding families and cousins, though, most easily, and identifying missing families, you want to be looking at matches of about 90 centiMorgans or greater, about four degrees out in any direction. You want third cousins or closer, or second cousins once or twice removed. However, because of cousins marrying other cousins, you may have results that include fourth cousins or third cousins once or twice removed whose double relationships make them appear closer to you than they are. So the general rule of thumb is you want to know everything you can about all the relationships five or six degrees out, or within a range of 150 to 180 years. Based on the average of a range of numbers of centiMorgans,, you are looking for 3rd cousins, 2nd cousins once removed and 1st cousins twice removed or closer. The average for matches between two third cousins, for instance, is 73 cM. A third cousin match means that you share a 2nd great-grandparent. Your common ancestor is four generations back from both of you. This website has a good chart for readers to reference when they are looking at a match of a certain number of centiMorgans and trying to determine the degree of the relationship and where in a tree a common ancestor might be. https://thegeneticgenealogist.com.

You will notice that this chart, and others like it, gives a range of centiMorgans for relationships. Not many people realize or remember that males have less DNA than females do. Both males and females have 24 chromosomes BUT females have two X chromosomes and males have an X and Y. The Y chromosome is shorter and has less DNA than the X. This means that male to male relationships will have fewer centiMorgans than male to female and female to female. Female first cousins and sisters will have more shared centiMorgans than brothers and male first cousins where both cousins are males. Cousins who are one male and one female will usually be in the middle of the range, and two female cousins will usually be in the top of the range and two male cousins will usually be in the bottom of the range. These ranges were established by examining thousands of actual relationships. If someone you think is a sibling does not fall in full sibling range, that person is a half sibling. DNA is a scientific test with repeatable results. It doesn’t lie, but people do. If someone you think is first cousin does not fall in a first cousin range and falls in a “half first cousin” range, one of you does not have the parent you all think you do and you need to look at other matches and the amount of centiMorgans for those for each person.

Make sure all the stated trees that people give you, or or published or online, are well documented and people are who they are claimed to be on paper. Good documentation involves copies of as many primary and valid secondary sources of information you can get your hands on, or uploaded into your computer. Look for consistencies among the sources.

Then take the trees you have and compare the surnames in them with the DNA matches on the databases. Create a simple database of your own, or somehow mark the names or surnames in the matches, to identify mother’s family, mother’s mother’s family, mother’s father’s family, etc. to see what you have. What is left that doesn’t fit in any of the other categories are your “brick wall lines” or relatives you didn’t know you had.. Then look at the matches of your “unknowns” to see if their MATCHES include any of your known family lines. Whatever is left after these two steps is your unknown lines: a father’s line, a paternal grandfather’s line, etc.. Start writing to those “unknown matches” to see if any of them have family trees, or have information that might lead you to a missing father, or grandfather, or whatever.

It’s not easy and it takes patience, but it can be done. Ce Ce Moore and others are making a business of this. Most genetic genealogists now in these businesses did not start out with anthropology degrees but learned how to read and analyze the same as most other people. It does not require a four year college degree to become a fairly good genealogist and be able to actually use DNA tests and analysis to identify all your ancestors and cousins. It does take a bit of luck though. It’s a lot easier to identify missing ancestors if you have DNA matches from all parts of your family with all lines represented, each, with several persons with 70cM or more in matching DNA, ideally 150 cM or more.

It’s also easier to identify ancestors and relatives whose ethnic background is from some parts of the world and difficult for others. The current data bases of all the big testing and analysis companies have data bases that are heavily dominated by samples from the British isles and some parts of West Europe. Other than that, samples from a few countries in west Africa are next The samples are almost all countries east of Vienna and south of the Alps and Pyrenees are almost non-existent and you may know and have well documented East European, Spanish, Italian or Balkan or Greek ancestors and they will be chronically mis-identified by the companies’ algorithms used to ESTIMATE ethnic ancestry. Also bear in mind the matches and reference matches’ locations are mapped based on where they live today–not where their ancestors lived. Two world wars and the Soviet-Russian invasion and occupation of a large part of east, east-central and southeast Europe caused a lot of survivors to move west. Most persons with Jewiish ancestry will find that in the 20th century their parents and grandparents moved very far from places their families had lived for centuries and that many of their relatives were killed between 1939 and 1945. The revolutions in these areas that failed caused many hundreds of thousands of refugees to flee west also. There were refugees from the 1848 revolutions that swept most of Europe, 1867 in Austria Hungary, and in Ireland, from the genocides in Armenia, Romania, Czechoslovakia and Yugoslavia that took place just after World War I, and revolutions in 1953, 1956 and 1968 after World War II. Irish fled famine and lack of economic opportunities. Many ended up in England where the industrial revolution, the railroads, etc. all needed unskilled labor. Some ended up in Glasgow, Scotland, for the same reason. Warfare in other parts of the world caused similar displacements and migrations. A little knowledge of the general history of wars, disasters and resulting migrations for at least the last two centuries is helpful to finding more ancient roots for families, as well as understanding why 2nd and 3rd cousins who ought to be of a particular ethnicity are in some other country today.

And don’t forget to get tested by all the DNA companies you can afford. The more databases you are in, the more relatives you will find, and the better your chances of breaking down your ‘brick walls’.