Tag Archives: GEDmatch

What Is a Sibling Anyway? Full, Half, Three-Quarters, Step, Adopted, Donor-Conceived & Twins

Posted on by
12

I’ve seen the term sibling used many different ways, sometimes incorrectly.

When referring to their own siblings, people usually use the term brother or sister, regardless of whether they are talking about a full, half or step-sibling. It’s a term of heart or description. It’s often genealogists who are focused on which type of sibling. As far as I’m concerned, my brother is my brother, regardless of which type of brother. But in terms of genetics, and genealogy, there’s a huge difference. How we feel about our sibling(s) and how we are biologically related are two different things.

Let’s cover the various types of siblingship and how to determine which type is which.

  • Full Siblings – Share both parents
  • Half-Siblings – Share only one parent
  • Three-Quarter Siblings – It’s complicated
  • Adopted Siblings
  • Donor-Conceived
  • Step-Siblings – Share no biological parent
  • Twins – Fraternal and Identical

Full Siblings

Full siblings share both parents and share approximately 50% of their DNA with each other.

You can tell if you are full siblings with a match in various ways.

  1. You share the same fairly close matches on both parents’ sides. For example, aunts or uncles or their descendants.

Why do I say close matches? You could share one parent and another more distant relative on the other parent’s side. Matching with close relatives like aunts, uncles or first cousins at the appropriate level is an excellent indicator unless your parents or grandparents are available for testing. If you are comparing to grandparents, be sure to confirm matches to BOTH grandparents on each side.

  1. Full siblings will share in the ballpark of 2600 cM, according to DNAPainter’s Shared cM Tool.

Keep in mind that you can share more or less DNA, hence the range. It’s also worth noting that some people who reported themselves as full siblings in the Shared cM project were probably half siblings and didn’t realize it.

  1. Full siblings will share a significant amount of fully identical regions (FIR) of DNA with each other, meaning they share DNA at the same DNA address from both parents, as illustrated above. Shared DNA with each other inherited from Mom and Dad are blocked in green. The fully identical regions, shared with both parents, are bracketed in purple. You can’t make this determination at FamilyTreeDNA, MyHeritage or Ancestry, but you can at both 23andMe and GEDmatch.

At GEDmatch, the large fully green areas in the chromosome browser “graphics and positions” display indicates full siblings, where DNA is shared from both parents at that location.

I wrote about the details of how to view fully identical regions (FIR) versus half identical regions (HIR) in the article, DNA: In Search of…Full and Half-Siblings.

  1. If your parents/grandparents have tested, you and your full sibling will both match both parents/grandparents. Yes, I know this sounds intuitive, but sometimes it’s easy to miss the obvious.

At FamilyTreeDNA, you can use the matrix tool to see who matches each other in a group of people that you can select. In this case, both siblings are compared to the father, but if the father isn’t available, a close paternal relative could substitute. Remember that all people who are 2nd cousins or closer will match.

  1. At Ancestry, full siblings will be identified as either “brother” or “sister,” while half-siblings do not indicate siblingship. Half-siblings are called “close family” and a range of possible relationships is given. Yes, Ancestry, is looking under the hood at FIR/HIR regions. I have never seen a full sibling misidentified as anything else at Ancestry. Unfortunately, Ancestry does not give customers access to their matching chromosome segment location data.
  2. Y-DNA of males who are full siblings will match but may have some slight differences. Y-DNA alone cannot prove a specific relationship, with very rare exceptions, but can easily disprove a relationship if two males do not match. Y-DNA should be used in conjunction with autosomal DNA for specific relationship prediction when Y-DNA matches.
  3. Y-DNA testing is available only through FamilyTreeDNA, but high-level haplogroup-only estimates are available through 23andMe. Widely divergent haplogroups, such as E versus R, can be considered a confirmed non-match. Different haplogroups within the same base haplogroup, such as R, but obtained from different vendors or different testing levels may still be a match if they test at the Big Y-700 level at FamilyTreeDNA.
  4. Mitochondrial DNA, inherited matrilineally from the mother, will match for full siblings (barring unusual mutations such as heteroplasmies) but cannot be used in relationship verification other than to confirm nonmatches. For both Y-DNA and mitochondrial DNA, it’s possible to have a lineage match that is not the result of a direct parental relationship.
  5. Mitochondrial DNA testing is available only through FamilyTreeDNA, but haplogroup-only estimates are included at 23andMe. Different base haplogroups such as H and J can be considered a non-match.
  6. A difference in ethnicity is NOT a reliable indicator of half versus full siblings.

Half-Siblings

Half-siblings share only one parent, but not both, and usually share about 25% of their DNA with each other.

You will share as much DNA with a half-sibling as you do some other close matches, so it’s not always possible for DNA testing companies to determine the exact relationship.

Referencing the MyHeritage cM Explainer tool, you can see that people who share 1700 cM of DNA could be related in several ways. I wrote about using the cM Explainer tool here.

Hints that you are only half-siblings include:

  1. At testing vendors, including Ancestry, a half-sibling will not be identified as a sibling but as another type of close match.
  2. If your parents or grandparents have tested, you will only match one parent or one set of grandparents or their descendants.
  3. You will not have shared matches on one parent’s side. If you know that specific, close relatives have tested on one parent’s side, and you don’t match them, but your other family members do, that’s a very big hint. Please note that you need more than one reference point, because it’s always possible that the other person has an unknown parentage situation.
  4. At 23andMe, you will not show fully identical regions (FIR).
  5. At GEDmatch, you will show only very minimal FIR.

Scattered, very small green FIR locations are normal based on random recombination. Long runs of green indicate that significant amounts of DNA was inherited from both parents. The example above is from half-siblings.

  1. At FamilyTreeDNA and 23andMe, most men who share a mother will also share an X chromosome match since men only inherit their X chromosome from their mother. However, it is possible for the mother to give one son her entire X chromosome from her father, and give the other son her entire X chromosome from her mother. Therefore, two men who do share a mother but don’t have an X chromosome match could still be siblings. The X is not an entirely reliable relationship predictor. However, if two men share an entire X chromosome match, it’s very likely that they are siblings on their mother’s side, or that their mothers are very close relatives.

Three-Quarter Siblings

This gets a little more complicated.

Three-quarter siblings occur when one parent is the same, and the other parents are siblings to each other.

Let’s use a real-life example.

A couple marries and has children. The mother dies, and the father marries the mother’s sister and has additional children. Those children are actually less than full siblings, but more than half-siblings.

Conversely, a woman has children by two brothers and those children are three-quarter siblings.

These were common situations in earlier times when a man needed a female companion to raise children and women needed a male companion to work on the farm. Neither one could perform both childcare and the chores necessary to earn a living in an agricultural society, and your deceased spouse’s family members were already people you knew. They already loved your children too.

Neither of these situations is historically unusual, but both are very difficult to determine using genetics alone, even in the current generation.

Neither X-DNA nor mitochondrial DNA will be helpful, and Y-DNA will generally not be either.

Unfortunately, three-quarter siblings’ autosomal DNA will fall in the range of both half and full siblings, although not at the bottom of the half-sibling range, nor at the top of the full sibling range – but that leaves a lot of middle ground.

I’ve found it almost impossible to prove this scenario without prior knowledge, and equally as impossible to determine which of multiple brothers is the father unless there is a very strong half-sibling match in addition.

The DNA-Sci blog discusses this phenomenon, but I can’t utilize comparison screenshots according to their terms of service.

Clearly, what we need are more known three-quarter siblings to submit data to be studied in order to (possibly) facilitate easier determination, probably based on the percentage frequency distribution of FIR/HIR segments. Regardless, it’s never going to be 100% without secondary genealogical information.

Three-quarter siblings aren’t very common today, but they do exist. If you suspect something of this nature, really need the answer, and have exhausted all other possibilities, I recommend engaging a very experienced genetic genealogist with experience in this type of situation. However, given the random nature of recombination in humans, we may never be able to confirm using any methodology, with one possible exception.

There’s one possibility using Y-DNA if the parents in question are two brothers. If one brother has a Y-DNA SNP mutation that the other does not have, and this can be verified by testing either the brothers who are father candidates or their other known sons via the Big Y-700 test – the father of the siblings could then be identified by this SNP mutation as well. Yes, it’s a long shot.

Three-quarter sibling situations are very challenging.

Step-siblings, on the other hand, are easy.

Step-Siblings

Step-siblings don’t share either parent, so their DNA will not match to each other unless their parents are somehow related to each other. Please note that this means either of their parents, not just the parents who marry each other.

One child’s parent marries the other child’s parent, resulting in a blended family. The children then become step-siblings to each other.

The terms step-sibling and half-sibling are often used interchangeably, and they are definitely NOT the same.

Adopted Siblings

Adopted siblings may not know they are adopted and believe, until DNA testing, that they are biological siblings.

Sometimes adopted siblings are either half-siblings or are otherwise related to each other but may not be related to either of their adoptive parents. Conversely, adopted siblings, one or both, may be related to one of their adoptive parents.

The same full and half-sibling relationship genetic clues apply to adopted siblings, as well as the tools and techniques in the In Search of Unknown Family series of articles.

Donor-Conceived Siblings

Donor-conceived siblings could be:

  • Half-siblings if the donor is the same father but a different mother.
  • Half-siblings if they share an egg donor but not a father.
  • Full siblings if they are full biological siblings to each other, meaning both donors are the same but not related to the woman into whom the fertilized egg was implanted, nor to her partner, their legal parents.
  • Not biologically related to each other or either legal parent.
  • Biologically related to one or both legal parents when a family member is either an egg or sperm donor.

Did I cover all of the possible scenarios? The essence is that we literally know nothing and should assume nothing.

I have known of situations where the brother (or brothers) of the father was the sperm donor, so the resulting child or children appear to be full or three-quarters siblings to each other. They are related to their legal father who is the mother’s partner. In other words, in this situation, the mother’s husband was infertile, and his brother(s) donated sperm resulting in multiple births. The children from this family who were conceived through different brothers and had very close (half-sibling) matches to their “uncles'” children were very confused until they spoke with their parents about their DNA results.

The same techniques to ascertain relationships would be used with donor-conceived situations. Additionally, if it appears that a biological relationship exists, but it’s not a full or half-sibling relationship, I recommend utilizing other techniques described in the In Search of Unknown Family series.

Twins or Multiple Birth Siblings

Two types of twin or multiple birth scenarios exist outside of assisted fertilization.

Fraternal twins – With fraternal or dizygotic twins, two eggs are fertilized independently by separate sperm. Just view this as one pregnancy with two siblings occupying the same space for the same 9 months of gestation. Fraternal twins can be male, female or one of each sex.

Fraternal twins are simply siblings that happen to gestate together and will match in the same way that full siblings match.

Please note that it’s possible for two of a woman’s eggs to be fertilized at different times during the same ovulation cycle, potentially by different men, resulting in twins who are actually half-siblings.

A difference in ethnicity is NOT a reliable indicator of fraternal or identical twins. Submitting your own DNA twice often results in slightly different ethnicity results.

Identical twins – Identical or monozygotic twins occur when one egg is fertilized by one sperm and then divides into multiple embryos that develop into different children. Those children are genetically identical since they were both developed from the same egg and sperm.

Two of the most famous identical twins are astronauts Mark and Scott Kelly.

Identical twins are the same sex and will look the same because they have the same DNA, except for epigenetic changes, but of course external factors such as haircuts, clothes and weight can make identical twins physically distinguishable from each other.

DNA testing companies will either identify identical twins as “self,” “identical twin” or “parent/child” due to the highest possible shared cM count plus fully matching FIR regions.

For identical twins, checking the FIR versus HIR is a positive identification as indicated above at GEDmatch with completely solid green FIR regions. Do not assume twins that look alike are identical twins.

Siblings

Whoever thought there would be so many kinds of siblings!

If you observe the need to educate about either sibling terminology or DNA identification methodologies, feel free to share this article. When identifying relationships, never assume anything, and verify everything through multiple avenues.

_____________________________________________________________

Follow DNAexplain on Facebook, here or follow me on Twitter, here.

Share the Love!

You’re always welcome to forward articles or links to friends and share on social media.

If you haven’t already subscribed (it’s free,) you can receive an email whenever I publish by clicking the “follow” button on the main blog page, here.

You Can Help Keep This Blog Free

I receive a small contribution when you click on some of the links to vendors in my articles. This does NOT increase the price you pay but helps me to keep the lights on and this informational blog free for everyone. Please click on the links in the articles or to the vendors below if you are purchasing products or DNA testing.

Thank you so much.

DNA Purchases and Free Uploads

Genealogy Products and Services

My Book

Genealogy Books

Genealogy Research

DNA Shows Peter Johnson and Mary Polly Philips Are My Relatives, But Are They My Ancestors? – 52 Ancestors #350

Posted on by
10

One of the requests by several people for 2022 article topics revolved in some way around solving challenges and showing my work.

In this case, I’m going to show both my work and the work of a newly-discovered cousin, Greg Simkins.

Let’s start by reminding you of something I said last week in Darcus Johnson (c1750-c1835) Chain Carrier – Say What??.

Darcus is reported in many trees to be the daughter of Peter Johnson (Johnston, Johnstone) and his wife Mary Polly Phillips. Peter reportedly lived in Pennsylvania and died in Allegheny County, PA. However, I am FAR from convinced that this couple was Darcus’s parents.

The distance from Shenandoah County, VA to Allegheny Co., PA is prohibitive for courting.

The Shenandoah County records need to be thoroughly researched with various Johnson families reconstructed. I’m hoping that perhaps someone has already done that and a Johnson family was living not terribly far from Jacob Dobkins father, John Dobkins. That would be the place to start.

Greg, Peter Johnson’s descendant through son James reached out to me.

Hi Roberta, I read your essay today on Dorcas Johnson. I wanted to write to you because I am a descendant of Dorcas’s brother James and have DNA matches to support our connection.

Clearly, I was very interested, but I learned long ago not to get too excited.

Then, Greg kindly shared his tree and DNA results with me. He was also generous enough to allow me to incorporate his information into this article. So yes, this article is possible entirely thanks to Greg.

I was guardedly excited about Greg’s communication, but I wasn’t prepared for the HUGE shock about to follow!

Whoa!!!

Greg has done his homework and stayed after school.

First, he tracked the descendants of Peter through all of his children, to present, where possible, and added them into his trees at the genealogy vendors. The vendors can do much better work for you with as much ammunition as you can provide.

Second, he has doggedly tracked matches at MyHeritage, FamilyTreeDNA, Ancestry and GEDmatch that descend through Peter Johnson and Mary Polly Phillips’s children. By doggedly, I mean he has spent hundreds to thousands of hours by his estimation – and based on what I see, I would certainly agree. In doing so, he pushed his own line back from his great-great-grandmother, Elizabeth Johnson, three generations to Peter Johnson and Mary Polly Phillips – and proved its accuracy using DNA.

Altogether, Greg has identified almost 250 matches that descend from Peter Johnson and Mary Polly Phillips, and mapped those segments across his chromosomes.

Greg made notes for each match by entering the number of matching cMs into their profile names as a suffix in his tree. For example, “David Johnson 10cM” instead of “David Johnson Jr.” or Sr.  That way, it’s easy to quickly see who is a match and by how much. Brilliant! I’m adopting that strategy. It won’t affect what other people see, because no living people are shown in trees.

Of course, DNA is on top of traditional genealogical research that we are all familiar with that connects people via deeds, wills, and other records.

Additionally, Greg records research information for individuals as a word document or pdf file and attaches them as documents to the person’s profile in his tree. His tree is searchable and shareable, so this means those resources are available to other people too. We want other researchers to find us and our records for EXACTLY this reason.

One thing to note is that if you are using Ancestry and use the Notes function on profiles, the notes don’t show to people with whom you share your tree, but links, sources and attached documents do.

Greg has included both “Other Sources” and “Web Links” below.

Click images to enlarge

For example, if I click on Greg’s link to Historic Pittsburg, I see the land grant location for Peter Johnson. Wow, this was unexpected.

Ok, I love maps and I’m hooked. Notice the names of the neighbors too. You’ll see Applegate again. Also, note that Thomas Applegate sold his patent to Richard Johnson. Remember the FAN club – friends and neighbors.

Ok, back to DNA for now.

The Children

Ancestors with large families are the best for finding present-day DNA matches. Of course, that’s because there are more candidates. More descendants and that means more people who might test someplace. This is also why you want to be sure to have your DNA in all 4 major DNA vendors, FamilyTreeDNA, MyHeritage, Ancestry, and 23andMe, plus GEDmatch.

This is a portion of Greg’s tree that includes the children of Peter Johnson and Mary Polly Phillips. Note that two Johnson females married Dobkins men. I’ve always suspected that Margaret Johnson and Dorcas Johnson were sisters, but unless we could use mitochondrial DNA, or figure out who the parents of either Peter or Mary are, there’s no good way to prove it.

We’re gathering some very valuable evidence.

At Ancestry, Greg has 85 matches on his ThruLines for Peter Johnson and Mary Polly Phillips, respectively.

  • Of course, Greg has the most matches for his own line through Peter’s son James Johnson (1752-1826) who married Elizabeth Lindsay and died in Lawrence County, IL: 35 matches.
  • Next is Margaret Johnson (1780-1833) who married Evan Dobkins in Dunmore County, VA, brother of my ancestor, Jacob Dobkins. She probably died in Cocke County, TN: 25 matches. Dorcas named one of her children Margaret and Margaret may have named one of her children Dorcas.
  • Solomon Johnson (1765-1843) married Frances Warne and stayed in Allegheny County, PA: 8 matches. Notice one of Peter’s neighbors was a Warner family. Dorcas named one of her children Solomon, a fairly unusual name.
  • Mary Johnson (1770-1833) married Garrett Wall Applegate and died in Harrison County, IN: 7 matches. The Applegates were Peter Johnson’s neighbors and Garrett served in the Revolutionary War in the 8th VA Regiment. Clearly, some of these settlers came from or spent time in Virginia.
  • Dorcas Johnson (c1750-c1835) married Jacob Dobkins in Dunmore County, VA and died in Claiborne County, TN: 5 matches.
  • Peter Johnson (1753-1840) married Eleanor “Nellie” Peter and died in Jefferson County, KY: 4 matches.
  • Richard D. Johnson (1752-1818) married Hannah Dungan and Elizabeth Nash: 2 matches.

Unfortunately, since most of those matches are between 7 and 20 cM, and Ancestry does not display shared matches under 20 cM, we can’t use Ancestry’s comparison tool to see if these people also match each other. That’s VERY unfortunate and extremely frustrating.

Greg matches more people from this line at MyHeritage, GEDmatch and FamilyTreeDNA, and thankfully, those vendors all three provide segment information AND shared match information.

Cousins Are Critical

While Greg, unfortunately, does not match me, he does match several of my cousins whose tests I manage.

Two of those cousins both descend from Darcus Johnson through her daughter Jenny Dobkins, through her daughter Elizabeth Campbell, through her daughter Rutha Dodson, through her sons John Y. Estes and Lazarus Estes, respectively.

Another descends through Jenny Dobkins son, William Newton Campbell for another 5 generations. These individuals all match on a 17 cM segment of Chromosome 20.

Other known cousins match Greg on different chromosomes.

Looking at their shared matches at FamilyTreeDNA, we find more Dobkins, Dodson and Campbell cousins, some that were previously unknown to me. One of those cousins also descends through William Newton Campbell’s daughter for another 4 generations and matches on the same segment of chromosome 20.

DNAPainter

Emails have been flying back and forth between me and Greg, each one with some piece of information that one of us has found that we want to be sure the other has too. Having research buddies is wonderful!

Then, Greg sent a screenshot of a portion of his chromosome 20 from DNAPainter that includes the DNA of the cousins mentioned above. I didn’t realize Greg was using DNAPainter. It’s an understatement to say I’m thrilled because DNAPainter does the cross-vendor triangulation work automatically for you.

Just look at all of those matches that carry this Johnson/Phillips segment of chromosome 20. Holy chimloda.

Greg also sent his DNAPainter sharing link, and it turns out that this is only a partial list, with one of my cousins highlighted, dead center in the list of Peter Johnson’s and Mary Polly Phillip’s descendants. Greg has even more not shown.

Trying Not to Jump to Conclusions

I’m trying so hard NOT to jump to conclusions, but this is just SOOOO EXCITING!

Little doubt remains that indeed, Peter Johnson and Mary Polly Phillips are the parents of Dorcas Johnson who married Jacob Dobkins and also of Margaret Johnson who married Evan Dobkins. I’ve eliminated the possibility of other common ancestors, as much as possible, and verified that the descent is through multiple children. This particular segment on chromosome 20 reaches across multiple children’s lines.

I say little doubt remains, because some doubt does remain. It’s possible that perhaps Dorcas and her sister weren’t actually daughters of Peter Johnson, but maybe children of his brother? Peter was reported to have a brother James, a sheriff in Cumberland County, PA. but again, we lack proof. If Dorcas is Peter Johnson’s niece, her descendants would still be expected to match some of the descendants of Peter and his wife.

Also complicating matters is the fact that Greg also has a Campbell brick wall with a James Campbell born about 1790 who lived in Fayette County, PA, in the far northwest corner of the state. Therefore, DNA matches through Dorcas Johnson Dobkins’s daughters Jenny and Elizabeth who married Campbell brothers need to be verified through her children’s lines that do NOT descend through her daughters who married Campbell men.

Nagging Questions

I know, I’m being a spoilsport, but I still have questions that need answers.

For example, I still need to account for how the Johnson girls managed to get to Shenandoah County, VA (Dunmore County at that time) to meet the Dobkins boys, spend enough time there to court, and then marry Evan and Jacob nine months apart in 1775. Surely they were living there. Young women simply did not travel, especially not great distances, and marriages occurred in the bride’s home county. Yet, they married in Shenandoah County, VA, not in PA.

What About the Records?

We are by no means done. In fact, I’ve just begun. I have some catching up to do. Greg has focused on Peter Johnson and Mary Polly Phillips in Pennsylvania. I need to focus on Virginia.

Of course, the next challenge is actual records.

What exists and what doesn’t? FamilySearch provides a list for Dunmore County, here, and Shenandoah, here.

Was Peter Johnson ever in Dunmore County that became Shenandoah County, VA, and if so when and where? If not, how the heck did his two daughters marry the Dobkins boys in 1775? Was there another Johnson man in Dunmore during that time? Was it James?

Where was Peter Johnson in 1775 when Dorcas and Margaret were marrying? Can we positively account for him in Pennsylvania or elsewhere?

Some information has been published about Peter Johnson, but those critical years are unaccounted for.

It appears that the Virginia Archives has a copy of the 1774-1776 rent rolls for Dunmore County, but they aren’t online. That’s the best place to start. Fingers crossed for one Peter Johnson living right beside John Dobkins, Jacob’s father. Now THAT would convince me.

Stay tuned!

Note – If you’d like to view Greg’s tree at Ancestry, its name is “MyHeritage Tree Simkins” and you can find it by searching for Maude Gertrude Wilson born in 1876 in Logan County, Illinois, died January 27, 1950 in Ramsey County, Minnesota, and married Harry A. Simkins. Elizabeth Ann Johnson (1830-1874) is Maude’s grandmother.

_____________________________________________________________

Follow DNAexplain on Facebook, here or follow me on Twitter, here.

Share the Love!

You’re always welcome to forward articles or links to friends and share on social media.

If you haven’t already subscribed (it’s free,) you can receive an email whenever I publish by clicking the “follow” button on the main blog page, here.

You Can Help Keep This Blog Free

I receive a small contribution when you click on some of the links to vendors in my articles. This does NOT increase the price you pay but helps me to keep the lights on and this informational blog free for everyone. Please click on the links in the articles or to the vendors below if you are purchasing products or DNA testing.

Thank you so much.

DNA Purchases and Free Uploads

Genealogy Products and Services

My Book

Genealogy Books

Genealogy Research