How do I use DNA in my research?
Traditional genealogy research uses probabilities. A birth certificate is considered very good evidence, but it’s not an absolute proof. Information is collected and given a relative value based upon how much it is believed the data can be trusted. DNA is also based on probabilities, but it’s not susceptible to forgery or alteration. It can’t be misplaced and there’s no chance of a transcription error.
Genealogical research is normally an attempt to prove a lineage. This is done by collecting documents or oral statements attesting to a person being the child of someone else.
– The odds the lineage is true is high if the evidence is good.
– If the evidence is poor, the odds the lineage is true is low. (Click here to read about genealogical proofs).
DNA research works in a similar manner — the evidence is weighed to give a probability of relationship. But it’s not like a birth certificate or Bible record. The reason is that many people might have the same DNA signature. So, DNA information must be compared to that of others to see if they match. If a match occurs, it is then necessary to include the traditional methods (paper trail, etc.) to prove the DNA signatures are descended from the same line.
– If a match is not found, no relationship exists.
– If a match is found, and there is supporting evidence the two people are from the same lineage, the DNA data can then provide a powerful tool to narrow the scope of the search to particular branches of the family and to specific time periods in which their common ancestor lived.
Traditional genealogy research is based on probabilities.
– The odds a person is descended from another person is high if a birth certificate exists.
– The odds a person is a child of someone else is high if the mother or a witness to the birth says it.
– The odds a person is a child of someone is low if there is no supporting evidence for it.
Genealogical DNA research is also based on probabilities.
– The odds two unrelated people will have the same DNA signature are low — but possible.
– The odds two people are related decrease as the differences in their DNA signature grow larger.
– The likelihood of two people being related are high if they share the same DNA signature.
– The likelihood two people are related are very high if they (1) share the same DNA signature and (2) share the same surname.
Data from DNA testing is an adjunct to genealogical data. Perhaps the most significant adjunct to DNA data is the surname. Having the same surname does not imply a relationship, but a close DNA match along with the same surname virtually ensures it.
DNA data, coupled with traditional genealogical evidences can effectively prove a relationship.
DNA proofs are based upon comparison of DNA samples of two or more people. Since it is unlikely a sample will be available of a distant ancestor, samples of living people are compared. The typical scenario is for people who have a DNA match to compare their ancestral genealogies to see if they might be related. If it appears they could be related, further investigation for genealogical proofs is undertaken. In this way the scope of the search is narrowed.
– The odds a relationship exists is low if the genetic distance is great (far apart).
– The odds a relationship exists is high if the genetic distance is small (close together).
– The odds a relationship exists is greatly increased if the same surname exists between the comparison subjects.
– The odds a relationship exists is increased even further with supporting genealogical evidence (e.g. the same surname, a supporting oral history, or other clues indicating similar ancestries).
What increases the probability of a relationship?
– A close genetic match in the DNA test (small genetic distance).
– Having the same surname.
– Any of the standard genealogical proofs (birth certificate, oral tradition, census, etc.)
How do you use the table of test results — and what does it mean?
Click on one of the kit numbers to have it become the standard of comparison to all the other kits. The fewer the differences between kits, the more likely those people are to be related through a fairly recent common ancestor.
To “refresh” the chart back to the default display (without any comparison showing), click on the word “reset” at the top of the left-most column.
What do the differences mean?
10×1-x.gifFirst, keep in mind that we are all related — it’s just a matter of how long ago our “Most Recent Common Ancestor” (MRCA) lived.
10×1-x.gifEverything in this help and from the genealogical DNA sites is directed to finding a MRCA within approximately the last 1,000 years. Prior to that there were no surnames, making documentation and proof much more difficult. Statements about not being related are predicated within this approximate 1,000 year time frame, unless specifically stated otherwise.
10×1-x.gifBasically — the fewer the number of differences between two people, the more closely related they are assumed to be. But DNA values can move toward one another, making it appear two people are related when they are not. The weight given the numeric difference shown in the far right column depends on (1) whether any other evidence exists of kinship (such as a common surname or documentation) and (2) the number of markers tested (the more markers, the more information to work with). DNA evidence supports traditional evidence, it does not replace it.
10×1-x.gifA flaw in the design of this table is that the difference value shown does not indicate how many markers are being compared, and that is very important, because a difference of 3 on 37 markers is not significant, but it is very significant if only for 12 markers.
Click here to see the team Liddell Interpreting Your Test Results.
On a test of the first 12 markers, differences of 3 or more are considered unlikely to be related unless there is a common surname or other evidence to indicate kinship. Visit the Family Tree DNA Genetic Distance FAQ links below for a better explanation.
FTDNA 12-marker FAQ
For the 25 marker test, a difference of 4 or more is considered most likely not related.
FTDNA 25-marker FAQ
For the 37 marker test, a difference of 6 or more is considered very unlikely to be related.
FTDNA 37-marker FAQ
– The Y-DNA 12-marker test examines deep ancestry (up to 2,000 years or more).
– The Y-DNA 13 to 25-marker test examines more recent relationships.
– The Y-DNA 26 to 37-marker uses faster moving markers to provide even more refinement of the results.
Regarding large genetic distances.
DNA evidence is a statistical probability: For example, your house is very unlikely to be struck by a tornado. It can happen, but the probability of it happening is very low. If two people have a large genetic distance between them, they are not likely related within a useful genealogical time frame (about 1,000 years). But if they share the same surname, the likelihood of kinship goes up dramatically. If they have a paper trail showing their ancestors were from the same places and times, the likelihood increases even further. If they can find a person they are both kin to, then they have found their “missing link” and can infer relationship with each other.
A large distance does not prove no kinship exists, only that the probability of it is low. The greater the distance, the less likely kinship exists. (Conversely, a small genetic distance does not prove a kinship does exist either, as it is possible for markers to move toward each other, but the probability of kinship is much higher.)
Kinship can exist even if the distance is large when the genetic material changes more quickly than usually happens and/or when the changes occur in the same direction and away from each other (see the theoretical family scenario below).
The probability a kinship exists is increased if a surname is shared, or if there is documentation or family history indicating kinship.
Larger-than-normal DNA genetic distances can still validate a family relationship if the intermediate distance movements can be located. This requires finding other branches that are related and which have closer genetic distances — finding the “in between” cousin. In other words, two people who are 5 distances apart might be related, but the probability is low. But, were each of those two people to find one other person they both were only 2 markers distant from, then they are probably related to that person and so are a little more likely to be related to each other. If their family traditions indicate they are both related to this person, the probability of kinship increases even further. If they can find written documentation, then the amount of evidence has raised the probability to the point of virtual certainty. (See Cloud’s Transitive Principle of Kinship.)
The markers in red move more quickly, so are given less weight. A distance of 2 red markers might approximate an actual distance of 1 of the others when analyzing the data (each marker has its own characteristics).
Some family lines experience rates of change greater than the norm. One of the Cloud branches has been proven to have a mutation rate much faster than the average.
As a family tree grows over many generations, some of the branches may drift apart in DNA genetic distance.
So far, all of our project members fall into haplogroup R1b. The haplogroup is an indicator of your ancient ethnic origins and R1b indicates origins in Western Europe many thousands of years ago. That particular group (R1b) has the least diverse DNA compared to any other group because those peoples came from only a few groups and remained isolated from others for centuries. This makes R1b the most difficult to interpret, since all the descendants come from closely related lines. Haplogroup R1b can expect to find matches with unrelated individuals more often than the other groups. It is therefore more important for participants in that group to be even more careful with their accompanying evidences.
One possible avenue of research are the collateral lines our families lived and traveled with over the years.
Click on the links below to see the CLOUD surname project results compared to other surname project results. (There are stories of the early Cloud family coming from the clan McCleod, and the Youngblood family was one collateral line to ours. The Black family is shown below because one of our Cloud lines changed their name to Black. We can probably learn more from how distant these branches are from ours than from anything else.)
Click on a kit number to have it become the standard of comparison (using one of the modal values is recommended). The differences will be color-coded. Haplogroups in green have been confirmed with SNP testing. Haplogroups in red have been estimated from the STR data (the allele values under the DYS columns).
U106+/S21 SNP The majority of our group belong to haplogroup R1b1b2g. This group is identified by a SNP named U106 (aka S21). It may have occurred 5,000 or more years ago. The “parent” haplogroup, R1b1, is believed to have originated about 20,000 years ago in Western Europe. Studies are underway to formulate the possible migration pattern of our sub-group and to determine where this group lived after the U106 mutation occurred.
McLeod Many people believe that the Cloud family comes from the MacLEOD clan. A study of the MacLeod project shows that surname contains many unrelated families, which is common for Scottish surnames. If is possible one of these family groups is related to one of ours but there doesn’t seem to be any evidence of it at this point.
Black This family group is shown primarily because one of our Cloud families changed their surname to Black. No relationship is indicated, but viewing it might help one better understand the use of DNA and to get a feel for how far apart DNA signatures can be.
Youngblood This family group is shown because it is collateral with some of our Cloud families. No relationship is indicated, but viewing it might help one better understand the use of DNA and to get a feel for how far apart DNA signatures can be.
(Keep in mind that DNA research can give us ideas about our more ancient ancestry too. It is helpful if test participants agree to allow their test results to be put into the large Y-Search database. The Cloud Project Administrator can help with doing that and can help you submit your paternal lineage gedcom so that matching lines can look for clues.)
Members of haplogroup R1b often need to take more definitive tests (i.e. 25 or 37-marker tests) and to have other members of their branch (2nd, 3rd or 4th cousins) to also submit samples in order to better define their base DNA signature and to show where any mutations have occurred.
A brief explanation of DNA testing as used for genealogical purposes.
The Y-DNA (passed from father to son) remains identical for many generations, with virtually no changes occurring for about 10 generations (250 years). This characteristic makes Y-DNA testing ideal for finding closely related lines. (Ed. note — based on my own observations and correspondence with other project coordinators, this is not true. Perhaps this refers to 12-marker “deep ancestry” markers only, for several projects are experiencing numerous allele changes in lines separated by 10 or less generations.) DNA was first used for genealogical research in 1997 and in 2000 the first information on the rates of change for the markers was identified. Though precise, the application / interpretation of the results is still being refined.
— Visit the National Geographic Genographic Project.
— Visit the Blair surname project – DNA 101.
Consider a theoretical family lineage:
Assume a man has three sons.
Assume two of these sons have a mutation (change) in their y-DNA, each in opposite directions (an unlikely, but possible, circumstance used to illustrate a point).
Now the father and one son have a perfect match, while the brothers are one or two mutations apart — called “genetic distance”.
during the next ten generations, at least one more mutation occurs, perhaps two.
The results of DNA testing the descendants of the tenth generation might then show:
one branch might be a perfect match with the first male ancestor.
other branches may have genetic distances of 1 to perhaps 3 or 4, depending on the direction and quantity of mutations.
It is also possible for two unrelated lines to experience mutations toward each other, causing confusion when they appear to be more closely related than they are.
How is DNA testing used in genealogical research?
It can prove that no relation exists.
It needs supporting evidence to infer relationship.
It can validate a paper trail — but it cannot replace it.
It can indicate with certainty where a relation does NOT exist. (This is very useful in preventing wasted time and money searching the wrong line. It can also prove very disappointing to people whose line is the result of a “false paternity”, an adoption, etc.)
It can provide valuable information about which lines are closest and which warrant further research.
It is very inexpensive — especially considering the time and money lost pursuing the wrong line or lines.
How is the testing done?
A sample of your DNA is obtained by scraping the inside of your cheek. It is sent to the lab, processed and the results are returned to you.
If you are part of a project, your results are also sent to the project coordinators.
Click here to see the DNA sample kit contents.
Should I be concerned about giving my DNA sample to the lab?
Perhaps. Some companies may not exercise care in protecting your identity or your sample results, though the information available from the tests is unlikely to compromise your privacy.
The company we have chosen (Family Tree DNA) has strictly enforced privacy policies. Click here to read them.
Read the Family Tree DNA Privacy FAQ here.
Why should I join a surname project?
You receive a substantial quantity discount on the cost of the test.
You will participate with others pursuing the same goal — finding ancestral brick walls.
How do I join a surname project?
Click here to join the Cloud Surname Project.
Click here to search for another surname project.
How do I print the Results Page? How do I print the colors on the pages?
To print the DNA Results page from your browser:
Print Settings — set to Landscape mode
Set browser to print background colors (see below):
Tools > Internet Options > Advanced > Printing > Print background colors & images)
File > Print Preview > Page Setup > Format & Options > Print Background (colors & images)
DNA testing helps find ancestral ties,
and provides validation for other evidence.