Is there any value in having siblings or parents and their children take DNA tests?

+3 votes
DNA testing is getting to be a fad. Is there anything to be gained by siblings taking the same DNA test? (I know siblings don't get the same mix of genes, but still, how much of importance can the average person learn from these variations?) Or even children of parents who have already been tested? Aside from paternity questions, I can't think of any. If anybody sees value where I see redundancy, could you enumerate what information of value could be found?

in Genealogy Help by Bob Scrivens G2G6 Mach 1 (19.4k points)
recategorized by Jillaine Smith
It probably depends on what you want to find out.

I have tested, both my parents have tested, and recently my mother’s sister has tested too (all with Ancestry).

We don’t know who my maternal grandfather’s father was, and my grandfather died some years ago. DNA tests are the only way we’ll find out who the mystery man was.

It’s been interesting to compare the matches of my mother and her sister - as you would expect, they don’t all match the same people and some of the numbers of cM are quite different. In some cases the difference is enough to make me reconsider whether it’s worth researching that shared match or not.

As my parents have both tested it probably wasn’t essential for me to test as well. The benefit for me is that Ancestry automatically marks my matches as “Mother’s side” or “Father’s side”; if that’s not there I can probably ignore the match as a false positive.
Yes. A parent or their siblings. Consider, if your father's sister tested autoDNA ALL of her matches would be good matches for you as well. Another example: I'm trying to prove a distant relation with Auto DNA to a paternal ancestor because I haven't found a living male descendant from that line.  I have a match with a guy but he doesn't match me on the chromosome I'm looking for. When he had his mother test however, there is that match I need. Can make all the difference when your looking at 6 generations past. Y-DNA testing for a dad or sibling is a waste I feel,(unless as mentioned elsewhere, you suspect a NPE). Better off finding a distant relative or something like that to fund or an upgrade to a promising match.

Actually, when the cMs are especially different for two siblings, that's when it tells you the MOST about what the relationship level is between that person and the sisters.

For example, if one sister matches 180cM, while the other matches 20cM, that tells you that the match is their 2C1R. If you only had the 180cM match, that would mean either an average 2C-level match, or a high 2C1R-level match. But since you have the OTHER match, at 20cM, that's too low to be a 2C, so it has to be a 2C1R-level match.

Of course, a "2C1R-level" match can also be a half-2C, a half-1C2R, a 1C3R, etc, depending on how many generations apart they are.

If they don't BOTH match, that also tells you that the match is at LEAST a 3C-level relation.

The same things apply when 1st cousins test.

If you have an example of your own, I'd be happy to see what I can make of it.

Thanks Frank. I'm in the middle of comparing the two sets of results and might try to write it up properly when I've finished. 

Just a few examples so far – none of them earth-shattering but I think they show it is worth testing siblings:

My aunt has a 49cM match but my mother doesn’t share any DNA with them (by Ancestry’s calculation). There are a lot of matches in the 20-30cM bracket like this.

Some others:

  • A 2C1R with whom my mother shares 222cM but her sister 155cM.
  • A 3C1R where my mother shares 105cM but her sister 66cM.
  • A 4C where my mother shares 70cM but her sister only 27cM.
Before I get into it, Julie & I have had a related discussion under her recent "twinship" question that you might want to check out.

* As to the 49cM case, this shows part the value of having more than one sibling test! For relatives that are 3C and beyond, there's a chance that you won 't get a match. So when one sibling doesn't get the match, the other might.

So what I can tell you right away - since one sibling matches, while the other does not - is that the relation is 3C or more distant. At 49cM it's probably too high a match to be as distant as 5C1R, so I'd say the relation is somewhere between 3C and 5C. Probably something in the middle 4C is the most likely.

* Almost ALL matches in the 20cM to 30cM bracket are at LEAST 3C, so it's not at all unusual to have one sibling match, while the other does not. The more distant the relation, the more likely it is to happen. Matches in this range can be anywhere from 2C1R to 6C1R, from what I've seen.

* 2C1R can go practically down to zero - in fact, there may be a very tiny chance (too small to really worry about) of NOT getting a match at this level of relation. About 1 in 30 of my 2C1R matches are below 20cM, the lowest being 8.5cM. The 2nd-highest 2C1R I've seen on AncestryDNA is 213cM, so your 222cM just sounds like an especially strong 2C1R match to me. My highest is 274cM, but I think I need to investigate that one. If there's intermarrying (which is to say that if the two test takers are related in more than one way) then it throws the numbers off - makes them higher. My 274cM match could EASILY have that going on. The average I'm seeing for 2C1R is about 94cM, so those two are both above average.

If you had just told me 155cM and 222cM, 2C would have been my guess, but I wouldn't have completely ruled out 2C1R. Again, the value is that the 155cM is too small to be a 1C1R while 222cM is too big to be a 3C, so using BOTH numbers you can tell the person is at the 2C or 2C1R level.

* I have a strange gap with the 5 newer tests I have where I have ZERO matches between 76cM and 90cM. My highest 3C1R is about 70 (the same as for 4C, and that makes no sense). So my own data says 105cM should be a 3C or closer, but I question my data on this a bit. On GEDmatch, I have 3C1Rs as high as about 90cM (which is again the same as 4C, oddly). On MyHeritage, I have a 3C1R as high as 96.6cM. Julie reports a 3C1R at 93cM.

So 105cM sounds a bit high to me for a 3C1R, but I'm not really sure. Anything much below that (your 66cM, for example) makes perfect sense. I would think 66cM could be anything from 2C1R to 4C1R.

There's something like a 70%-80% chance of matching a 3C1R, so it's not surprising that they both match.

* 70cM is about the max I've seen for a 4C, so any number between that and zero make perfect sense. The 70cM essentially tells you the match is between 2C1R and 4C. The chance of getting a 4C match is about 46%, so maybe it's a bit lucky for them both to get that match.

11 Answers

+4 votes
I have found matches to my mom's brother that were not matches to her, and vice versa. Who knows how the "average person" feels... all of these matches are precious to me.

If a parent has passed, then testing the child will give genetic information about the other parent. If both parents have passed and the child tests at a different company than the parents, that can be useful extra information. But otherwise, testing living parents is more useful than testing their child, and testing a child at the same company as the parent won't give much added value and the money is probably better spent elsewhere. It won't allow you to make additional cousin matches over having the parents' tests. (Although you can still learn from these redundant tests -- for instance, information about which sections of the child's genome came from which grandparent.)
by Barry Smith G2G6 Pilot (220k points)
+1 vote
Other than the family surprises of "oops, my brother and I don't have the same father" or "I didn't know I was adopted" or "my grandmother put horns on my grandfather" and such ilk -- and a solid medical history since the distribution of DNA / genes is diverse (random) or whatever -- and the proving or disproving of descent from royalty, nobility, or gentry (which is of importance to many) or the chance one's DNA will be scooped up and used to pinpoint one of three possible persons responsible possibly responsible for an egregious crime -- ?

If you are not considering one of those possible scenarios -- and granted there are other scenarios to consider lest anyone is moved to say I have ignored any such -- then no, there might not be any particular value except perhaps  thoroughness? Being thorough just in case.

If you don't feel it would be of value to yourself, it is most certainly not required at WT in the sense that you will be penalized for not testing.  Testing for DNA is not usually vital, as would be air, water, food, shelter and such.
by Susan Smith G2G6 Pilot (510k points)
+1 vote
Parents, siblings of parents, and grandparents if they are still living, definitely,.

Your siblings are not so important, but they may have inherited some genes that you dont have.
by Robynne Lozier G2G6 Pilot (944k points)
Except if you have a brother, because he carries the Y-DNA that women do not....

 But since I have no brothers I dont usually think about that.....
I actually had a bunch of "Lee" male cousins all do the Y-DNA test to assure that we really have a different Y-DNA than the famous Virginia Lees.
+9 votes

If you are serious about trying to get as much information as you can and trying to find all connections that you can and wish to verify as many as possible then you should test parents, uncles, aunts and siblings.

Why? You loose half the information each generation, so the oldest generation you can find will provide information you can not get anywhere else.  Siblings have different DNA and they will have different matches, not for 1st or 2nd cousins but once you get past that the differences can be significant.  I am working on matches for my full sister, no question, that I do not match.  If there are half-siblings in mix, even better as far as finding DNA connections.

If you only wish to look at close connections, no more than 3rd cousins or so - you should not spend the money - if you want more than than you should.

I must warn you when you start looking for cousins with only 10 or 15cm matches - you can take a very long time and may not find anything - but once in awhile you do.  Only you can decide it that is worthwhile.
by Philip Smith G2G6 Pilot (289k points)
I chose to make my cut off point at 20 cM.

So I dont look at anyone who shares less than 20 cM with me. Which of course is the vast majority of my matches.

And of those who share more than 20 cM - many of them do not have a tree so I cannot even begin to figure out how they might be related to me.
It's interesting how many people test, but don't post any kind of tree. Seems to me enabling exchange of family information is one of the prime reasons for testing in the first place.
+5 votes
Sometimes this "redundancy" is necessary because you need several people to test to prove connection to a line. Example, I have a lady ancestor who does not have much of a paper trail luckily, there are several descendants of her sisters who DO have solid paper trails, with whom I share genetic material, which proves the proposed match. She is not royalty, just a regular old midwest settler, but it answers a question that many of us who are related to her have had for twenty years.

The more people you get to match to this person, the better off you are.

So if you don't like answering questions, maybe it is just a "fad".

If you have siblings who are suspicious of DNA testing and don't want to do it, or other complications, that's something else entirely.
by Dina Grozev G2G6 Pilot (103k points)
+6 votes
Yes!!  Do it!  Test all your siblings and anyone still living in your parents' generation, or grandparents' if you're lucky enough to have any still living.

The only exception I would make is that any time you have results for two parents, testing their children will add nothing useful.

Any time you have three siblings' test results, you can do chromosome mapping.  That is extremely useful.

Siblings will have different matches from each other.  I learned a lot from my two maternal aunts' tests that I had not learned from my mother's.  Because my father died in 1984 and couldn't be tested, I also asked two of my paternal cousins to test and their matches have been very useful for me also.
by Julie Kelts G2G6 Pilot (451k points)
+12 votes
I see Julie Kelts has expressed much the same view as I will state here, but let me just bullet point the topics:

1) If both parents have tested, they will be most useful for research ancestry lines than any of the children's test separate or combined.

2) If one is in doubt of one parent or another or simply want to mark the relationship 'DNA Confirmed' then each relevant child must test.

3)  If one or both parents have not tested, half siblings and first cousins (as well as more distant identifiable cousins) can help determine the line and are important to test.

4) In general, the more the merrier but in some circumstances there will be fewer returns.  For example, if neither parent has been tested in a family with 14 children, then testing 3 to 5 of the children will give good coverage.

5)  Tests can also be taken at different providers and/or using different chip types that can both affect the possible matches even if testing the same person.

6) You did not mention the type of test and in addition to autosomal tests, there are yDNA and mitoDNA tests that can be taken.  Only one male per patrilineal line is needed so money should be devoted to testing that one person with a more extensive test rather than throwing it away on duplicate tests.  Only one test is needed for each matrilineal line.
by Thom Anderson G2G6 Mach 5 (55.4k points)
Thom, you are probably right about the number of children testing, and when I said test them all I wasn't thinking of 14.  But I only recently realized that in my own case, between my two living siblings and myself, there are large areas of several chromosomes where all three of us have Kelts DNA and none of us has Van Wye (my father's mother's line).  If I had more siblings, I'd test more until I could achieve 100% coverage of our Van Wye ancestry, and likewise for our other grandparents.
Absolutely, the tests are some darn inexpensive nowadays that testing everyone will maximize the benefit for those who have not had both parents tested.  I was merely stating that it is the law of diminishing returns.  Most of the coverage is with the first three tested siblings and each additional tests provides a more coverage but at a smaller and smaller rate.  

We did not discuss monozygotic multiple births or recipients of bone marrow transplants but I figured everyone understands those extraordinary cases implicitly..
+5 votes
My half sister and I both did tests and some of the people she was matched with on our shared (dad's) side did not match with me, though I had overlaps with their close relations -- some of whom did not overlap with her. I would definitely go for it, the more data you get, the better.
by Monica Edmunds G2G6 Mach 3 (30.5k points)
+2 votes
If you have tested parents, you don't need siblings and children, because they are a subset of their parents.

If you cannot test a parent, then siblings are very useful because they have segments that you won't have.
by Leandra Ford G2G6 Pilot (127k points)
+1 vote
I have a younger sister who has told me that she does not belong to my father.  She said that a friend of my Mother's told her that she belonged to her brother.  When I look at her I can see my Mother's friend in her so I suspect this claim is true.  

Also, I strongly suspect my youngest sister may not belong to my father. I have my suspicions but that is all they are where she is concerned.  She resembles my aunt (my Mother's sister) so closely she could have been her twin.

What information could I gain about them if the possible other families do not have trees on WikiTree, Ancestry, FTDNA, etc.?  And yes I have looked for trees of the ones I suspect hoping there would be a DNA test connected with them.
by Paula Franklin G2G6 Mach 4 (47.5k points)
0 votes
Definitely - because we each inherit a different combination of DNA from our parents and our parents from their parents.  The futher back you are researching the more help a robust and varied (YDNA, mtDNA, autosomal) DNA test results database will help  

For example, about 11 years ago I had my father take the first DNA test in our family.  I had him take a 37 marker YDNA test with Family Tree DNA. He passed in 2011 but I have continued to have additional tests on his sample.

For several years I kept running into the same researcher in my Angel/Angle line.  I asked if she had a male Angle in her line that would take the same test as my dad.  So she had her brother take the test and it turned out he matched my dad with a varient of 3.  This helped us to focus on who we determined to be our 5th ggrandfather, Johan Wendel Angle/Engel (JWA).

We learned JWA arrived from Germany in the mid 1760's and lived for a short time in Pennsylvania serving in the Continental Troops then the Rev War in Virginia.  After the war he headed south and west through Tennessee then ending up in Missouri Territory in the 1790's. Vital records are rare or non existant.

We found land documents in Missouri - one saying he had 9 children another saying he had 10 children.  To date we have only identified 5 with a couple of additional unproven potential matches.  This gives us a target either way.

I found another researcher that could possibly belong to our family and asked him to take the same YDNA test and he matched with a variation of 2.  We established he is descended from my JWA's son Francis' family through his son John, mine being the son Moses.  I had numerous researchers that claimed Francis' son John Angle was theirs but the DNA proved this researchers John was in my family.  

I have a one name study of this family of sorts and have identified 60 John Angles in the latter 1700's and early 1800's.  It is very difficult to determine which may be the same individual as there are few frames of reference, but I chip away as time allows.  There are limited JWA family researchers making it a challenge, but there are so many John Angles.

With the help of inventory sales in JWAs probate record we have researched many of the individuals that are listed and have found some family connections.  Aaron Quik shows up in multiple records with JWA over a period of time.  They are contemporaries.  I kept running into a very thorough Quik researcher who claimed a Nancy Angle as a wife to Aaron.  I reached out to her about the potential that she may belong to our family.

I had everyone I could recruit take an autosomal DNA test.  The Quik researcher and I did not share any DNA (very disappointing) but one of my siblings and a cousin matched.  That was all we needed to add Nancy to our family tree and a daughter of JWA. But is also helped us determine a more specific area to look for other Angles in Tennesse where Nancy and Aaron had lived.

It is a slow but very effective process.  I have worked with other YDNA projects, recruited individuals to test and confirm genealogy as well.  It has been very rewarding.  

Posting DNA to various family tree venues has also been helpful.  I have found more cousins in, My Heritage and wikitree, but not more children of JWA yet.  

If you are going to bother to take a DNA test please create and develop your family tree otherwise it is extremely difficult to determine relationships.  I have done so much research for other families trying to find connections to mine....

Related questions

+7 votes
2 answers
+14 votes
6 answers
+4 votes
1 answer
+7 votes
1 answer
182 views asked Jun 9, 2017 in The Tree House by Jourdi Cleghorn G2G6 Mach 2 (29.4k points)
+4 votes
2 answers
+3 votes
2 answers

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright