RootsChat.Com
General => Ancestral Family Tree DNA Testing => Topic started by: kateblogs on Friday 10 February 23 18:29 GMT (UK)
-
Hi :) I would love some input about the possible relationship between me and a person with whom I share the following:
'Shared DNA: 457 cM across 14 segments
Unweighted shared DNA: 457 cM
Longest segment: 114 cM'
This person has an extensive tree, but no matches in mine. They're not a half cousin (they're older than any of my aunts or uncles) and not a great, great grandparent, in fact, they're younger than my grandparents. Also not a half or full great niece/nephew, my niblings are far too young to have children of their own.
I suspect I know the relationship but would appreciate some fresh ideas in case I'm reading too much into this.
-
Are the ages such that half 1st cousin is possible? I.e 1 shared grandparent. More plausible if they have an illegitimate parent.
I have one such at 392 cM
-
No they're too old to be a half first cousin. They're younger than any of my grandparents, but not by a great deal.
I realise I'm being very mysterious here, so maybe it would help to give a bit more context. I didn't want to give too much detail because I'm worried I might be leaping to conclusions. Basically, I suspect this might be my grandfather's half sister.
-
87% probability 1/2 Great Aunt.
Go to DNA Painter’s Shared cM screen and key in 457
-
It fascinates me why people only give their cMs amounts with matches as there are 3 figures which work together...
cMs, % Shared & Segments
Most likely possibilities in the chart below.
Annie
-
Total cM and percentage shared are just two methods of expressing the same amount of shared DNA.
Dividing the total shared cM by 68 will give an approximate shared percentage. In this case 457/68 equals approximately 6.72% shared DNA. Either can be used in the DNA Painter Shared Project Tool. It's not necessary to know both.
Segment lengths and number of segments are also important at lower match lengths, where for instance a 30cM match over one segment is likely to be more significant that a 30cM match over say 4 segments, with a largest segment of 8cM. At the other extreme, with an overall match of 3385cM the longest segment and number of segments is pretty much immaterial to the conclusion.
-
Segment lengths only really come into play when you get to comparing matches.
Within Ancestry you currently do not have that that info, just cM & %age nor is there any comparison feature.
For Comparison you need other websites such as Gedmatch and then you hit the big But, in that the match you are interested in has to also upload there. So far I have had little luck with Gedmatch but it is still worth uploading Raw DNA data there.
The cM value via the relationship probabilities does give a Starter Place to look ie at the parents of a Great Aunt as being a common ancestor(s).
-
Segment lengths only really come into play when you get to comparing matches.
Within Ancestry you currently do not have that that info, just cM & %age nor is there any comparison feature.
Ancestry tell you the total shared cM and percentage after any Timber adjustment. They also tell you the number of segments that the match is across. Clicking on the link reveals the unweighted total shared cM (before any Timber adjustment) and the size of the largest segment.
You can view that information for any match and any shared matches between you and that match. Whilst you can't see the lengths of each and every shared segment, the length of the largest segment and number of segments allows a judgement between two or more matches of similar overall length and percentage, as to which is likely to have the closer relationship and be easier to resolve if other clues are missing or limited.
In many cases, you can also make a reasonable guestimate about the other segment sizes from that information. Deducting the largest segment size from the total shared cM gives you the remaining shared cM and number of segments. You know that they are all shorter than the largest segment, and that none are less than 7cM. If the match is fairly short, you should be able to estimate a reasonable approximation. If the number of shared segments is large, the estimation becomes more difficult, but a larger number of matching segments would suggest that the match is probably going to be closer/easier to resolve anyway.
So whilst nowhere near as helpful as a chromosome browser, there is enough information in many cases to make reasonably informed assumptions and selectively prioritise some matches over others.
-
Segment lengths only really come into play when you get to comparing matches.
Within Ancestry you currently do not have that that info, just cM & %age nor is there any comparison feature.
Ancestry tell you the total shared cM and percentage after any Timber adjustment. They also tell you the number of segments that the match is across. Clicking on the link reveals the unweighted total shared cM (before any Timber adjustment) and the size of the largest segment.
You can view that information for any match and any shared matches between you and that match. Whilst you can't see the lengths of each and every shared segment, the length of the largest segment and number of segments allows a judgement between two or more matches of similar overall length and percentage, as to which is likely to have the closer relationship and be easier to resolve if other clues are missing or limited.
In many cases, you can also make a reasonable guestimate about the other segment sizes from that information. Deducting the largest segment size from the total shared cM gives you the remaining shared cM and number of segments. You know that they are all shorter than the largest segment, and that none are less than 7cM. If the match is fairly short, you should be able to estimate a reasonable approximation. If the number of shared segments is large, the estimation becomes more difficult, but a larger number of matching segments would suggest that the match is probably going to be closer/easier to resolve anyway.
So whilst nowhere near as helpful as a chromosome browser, there is enough information in many cases to make reasonably informed assumptions and selectively prioritise some matches over others.
My highest Ancestry match has two children and both have been DNA tested.
Looking at and comparing the segment numbers and lengths there is no way that they appear to be related.
It is only though building a tree and getting the records found that they are found to be linked.
A Cousin of theirs also cannot be linked to them through the segment numbers and length.
Same with other DNA matches.
Going by the info Ancestry provides only allows you to go so far and making assumptions can lead to, well we all know where that can lead too.
-
No they're too old to be a half first cousin. They're younger than any of my grandparents, but not by a great deal.
I realise I'm being very mysterious here, so maybe it would help to give a bit more context. I didn't want to give too much detail because I'm worried I might be leaping to conclusions. Basically, I suspect this might be my grandfather's half sister.
Hi,
Just to rule out the other really high possibilities, which you may have done(?) : Does your grandfather have any older siblings than himself, that could have possibly fathered/ given birth
to a child of your match's age?
If say your grandfather has an older brother or sister who was a teenager at the time of your match's birth,( say 13 years old plus) . Your match maybe a first cousin Once Removed to you. ( and your parent's first cousin).
A first cousin Once Removed falls in with that high probability along with half great Aunt.
I share around your figure with my first cousin once removed,( my parent's older first cousin , ie my match's parent was my grandmother's older sister)
Otherwise, I would guess you are looking at the other very high probability and she could very possibly be your half great aunt.
My child shares just over 570 with a Half Great uncle ( where I share just over 1,160 with the uncle , half uncle to me)
Is your parent alive and are willing to take a DNA test? Or if not , one of your parent's siblings willing to take a test?
Kind regards
-
I have never understood why people don't look/compare % & segments with shared matches.
I'm no expert on DNA but by trying hard to understand what figures / % / segments matter when looking at shared matches as I've discovered some differing amounts may lead me to a possible NPE.
Too often I've seen people asking questions about relationships yet only providing cM amounts...
There are numerous charts available online which relate to cM relationships yet so many people omit any/all added info.
As with any query, regardless of it's nature, add all 'known' info. for best results.
Annie
-
My highest Ancestry match has two children and both have been DNA tested.
Looking at and comparing the segment numbers and lengths there is no way that they appear to be related.
It is only though building a tree and getting the records found that they are found to be linked.
A Cousin of theirs also cannot be linked to them through the segment numbers and length.
Same with other DNA matches.
Going by the info Ancestry provides only allows you to go so far and making assumptions can lead to, well we all know where that can lead too.
Biggles, you are totally missing the point, which surprises me.
It is an additional tool in the armoury to help guide your research.
You make assumptions every time that you use a tool such as DNA Painter. Do you automatically declare the relationship between matches based on the highest relationship probability proposed by DNA Painter? Of course you don't, but you do use the probabilities to assess the most likely relationships and use that information to guide your research. Could the match fall within the 1% or lowest probability range that DNA Painter proposes? Yes, of course it could, otherwise the probability wouldn't exist. But would you decide to initially concentrate your research on that possibility in preference to the higher probabilities that the tool proposes? I very much doubt it, and at that point you are making educated or guided assumptions, which you use to direct the research that will hopefully find the proof you are looking for.
Assessing the number of matching segments and segment lengths is no different. It is an additional tool in the box to help guide you, more with lower level matches than the higher ones. It can help filter unknown matches of similar overall length more effectively, and guide your direction of research towards those more likely to be helpful to you with less effort.
When you look at your matches on Ancestry or elsewhere, do you make assumptions as to which matches are likely to be closer based on overall match lengths. Of course you do, or should. Does that information by itself tell you how you are related? Of course not, but you are guiding your research using assumptions based on educated principles. Again, absolutely no different.
If you still don't get it, I recommend (again) Tracing Your Ancestors Using DNA, edited by Graham S. Holton, Pen & Sword Books. It is an excellent read from the basics to more complex areas of DNA investgation, with each chapter written by a different expert in that particular field, and my copy is now very well worn and tattered from use as reference material.
-
My highest Ancestry match has two children and both have been DNA tested.
Looking at and comparing the segment numbers and lengths there is no way that they appear to be related.
It is only though building a tree and getting the records found that they are found to be linked.
A Cousin of theirs also cannot be linked to them through the segment numbers and length.
Same with other DNA matches.
Going by the info Ancestry provides only allows you to go so far and making assumptions can lead to, well we all know where that can lead too.
Biggles, you are totally missing the point, which surprises me.
It is an additional tool in the armoury to help guide your research.
You make assumptions every time that you use a tool such as DNA Painter. Do you automatically declare the relationship between matches based on the highest relationship probability proposed by DNA Painter? Of course you don't, but you do use the probabilities to assess the most likely relationships and use that information to guide your research. Could the match fall within the 1% or lowest probability range that DNA Painter proposes? Yes, of course it could, otherwise the probability wouldn't exist. But would you decide to initially concentrate your research on that possibility in preference to the higher probabilities that the tool proposes? I very much doubt it, and at that point you are making educated or guided assumptions, which you use to direct the research that will hopefully find the proof you are looking for.
Assessing the number of matching segments and segment lengths is no different. It is an additional tool in the box to help guide you, more with lower level matches than the higher ones. It can help filter unknown matches of similar overall length more effectively, and guide your direction of research towards those more likely to be helpful to you with less effort.
When you look at your matches on Ancestry or elsewhere, do you make assumptions as to which matches are likely to be closer based on overall match lengths. Of course you do, or should. Does that information by itself tell you how you are related? Of course not, but you are guiding your research using assumptions based on educated principles. Again, absolutely no different.
If you still don't get it, I recommend (again) Tracing Your Ancestors Using DNA, edited by Graham S. Holton, Pen & Sword Books. It is an excellent read from the basics to more complex areas of DNA investgation, with each chapter written by a different expert in that particular field, and my copy is now very well worn and tattered from use as reference material.
Fully aware, but thanks for pointing it out.
We all have our own way of working.
Any assumptions made do require to be proved, discounted or to have limitations assigned to them.
In my example the Segment lengths are so different they would normally be assumed not to be direct relations but a removed cousin.
But as you effectively and frequently point out the passing on of DNA can be erratic.
Thanks also for the info on the book, I’ve not read that one, I’ll include a slide on it in my DNA Presentation that I am making next month.