I updated the tool to allow ENTRY of age rather than the slider, it was hot patched, just refresh the page and you should pick up the change, apologies to anyone who had it open at the time the hot fix went in.
@@Richard-zm6pt on the page, hold down shift key and hit F5 or click the page reload with shift key held down, either one should refresh the page for you and get the new page and overwrite what is stored locally in cache. if that doesnt work, let me know. That would mean your ISP is Caching the information and ill put code in the webpage to FORCE it to refresh
Maybe a future update (not nit-picking but thinking for the future) have the person put in the birth year of the tested person. Though there may have to be a whole 'nother thought when the possibility of both matches being deceased and over the probable age of someone who could have tested.
@@staceycoates1418 reason i didnt go with birth year was the fact it was more "extra" calculation as you would then convert year to age to get in same place (gens up and down based on age. plus 2 numbers versus 4 (has to be 4 due to century) and every key entered is a place of potential error aka 50% more
@@DNAFamilyTrees Hi Larry. My page still shows the age bar, yet I know that I did see the new version come up where you put the age, while I was in the site. If there is anything you can do to force the link on your site to refresh the page, that would be great. Thanks Teresa
RE: maternal vs. paternal DNA... I was always thinking of a road with houses on both sides of the street. I LOVE your zipper analogy. I'm using that one. It's neater, and the visual works better too.
Love this Larry. Something I would like to see is a version that could be used offline. Sometimes a person could be in a situation where they don't have access to internet but would still like to work on their genealogy mysteries. Also, age slider doesn't work for me.
Cooper mentioned that too, it does get "stuck" especially on tablets and touch screens. try refreshing the page but i will prioritize the change to a box for inputting the age!
Thanks, Larry! Still working to understand all of different passes and such. Would appreciate the probability chart to be bigger/easier to read. I had to increase the zoom and then can't see the chart. I'm 75 and my father was adopted. My highest match is 183 cM, so it is a struggle. Your videos help a lot!
183, whew that is one of the tough numbers, is that a low 212 or a multipath 106, that's hard to say. being that your in 70s and average genealogist mid to upper 50s i think we can assume they are a generation down from you. (assumption they may be your age looking too!) Since Dual most likely that leaves 2c1r s the most likely kinship. Now to estimate we have to "guess" are they a low 212 or a multi 106. Since the dual is not multi at 2c1r, my first guess would then be a low 212 (i normally would guess other way but one gen down there is only dual. if however they are in same generation or TWO generations down then multpath 106 is most likely!) the method would be the same either way but we will run with statistical probability. if dual then we double their cms of low 212 to be a 425 for their parent. this would mean the parent on same line as you. since not dual that means their parent likely to be a single path first cousin to you. Thus we infer that your grandparent, aka your adopted father's father. Now lets look at this from your fathers perspective. we double to move the ball up from you to him to 850 and again to 1700 to move up from the parent of the DNA match to their Grandparent. at 1700 and due to the age mapping which we assume consistent for generations (might vary, but we have to have our best guess on where to focus) this would put your father and their grandparent as half siblings. Find the 4 grandparents of that person and ONE of them is the most likely MRCA. since more likely a male, you can now whittle that to two. With DNA clustering you may be able to tell which side of theirs you have matches to and thus find which is the sibling and therefore the parents. That is where id look first. (and my first run through using the patterns. dividing down gives us kinship probability, BUT we want in all cases to really know the MRCA, so the kinship is just a step to that. SO, MULTIPLYing backwards takes us up each line, but NARROWS field and lesses margin for error tremendously thus honing in. Now we assumed a low 225, if this was a multipath, aka high, 106, then we do the same thing but it would require one more doubling to identify the place the MRCA would be. Hope this helps! Let me know if you get it with this!
@@DNAFamilyTrees Thanks so much for this detailed analysis. From his profile he is definitely a generation below while his father is also 12 years younger than I am. 2C1R is what I have been thinking. I believe the MRCA had an NPE put up for adoption also based on other matches, ages and circumstances. The plot thickens. Again, I love your videos, but this concept has been a little difficult to wrap my head around. I'll keep at it and will let you know if I find true answers. And I still have to find another paternal grandparent! Keeps me busy.
The probability chart (top right corner) needs to move left and up to give you room to make it much bigger. Also, I have known 2c1r from 95cM down to 35cM. Only the 95cM shows up as the correct circle of relationship. Love that you made an interactive tool!
Your example is perfect! the 95 fit spot on with 2c1r and while the 35 didnt, look at the CHART. for 35 the very first column of the chart is ... 2c1r :) if that is with ancestry, what is the "Unweighted" cms for that 35cm match? (just curious) They apply a WEIGHT to numbers under 90 that "shrink" so the 95 number is an UNWEIGHTED result while the 35 is a WEIGHTED result and in fact is NOT an apples to apples within their own list! (most dont realize this!) Thats why i included the chart as a way to cross reference! This is the best example of the day so far! Thank you!
@@DNAFamilyTrees 39cM went up to 50cM. 35cM went up to 39cM. This one still falls outside the predicted area. It is in the column before the blue balls. I totally forgot about TIMBER.
@@quiltingtangent459 What your saying is really leading edge on this, i had no idea people would be catching on so fast! OK, for those kinships that are KNOWN to be BEFORE the blue balls, this actually indicates a MULTI path! I have updated the tool display now, BUT i havent yet put the red in FRONT, still getting people on board with the "multi path" concept itself. I think people are getting it, so may have to bump that video up in plans :)
@@DNAFamilyTrees Time to explain "multi path" before blue balls! Need examples, now I'm lost again. You are telling me these cousins somehow are related to me in multiple ways. Need next video. I see that you made the other chart clickable. If you make the numbers a font size or 2 larger it might be readable on the page.
@@quiltingtangent459 still working on making it readable without click but I think the ONLY way will be to hand modify EVERY one of them with a different font/size rather than it auto generated. That will have to come later if i am going ot stay on schedule for full release this year of all parts including MRCA predictor.
I have an interesting match that doesn't seem to be following the ball and pattern method. She is a match to my mother and 3 of her other 4 siblings we tested, all in the 70-80 year old range. The match is 79. I have done a paper trail back to their MRCA, which would make her a 3rd cousin to my mother and their siblings (all full siblings). Out of all 5 siblings tested, this match only matches my mother at 9 cM, another 13, another 23, another 47 and the last not at all. Any thoughts on the low cMs? I cannot see in the paper trail how she could only be a half cousin or any other explanations.
Bonnie, this is a great question i hope people read it and the answer. DNA tends to be "normalized" which means that the CMs will tend to be close to the "step" or "node" in the pattern. That said, beyond 2C there ARE variances. out of 200 or so matches that are 2C1R or 3C i have TWO that are like this. mother matches at 278 i match ZERO. Mother matches 273 with another, i match ZERO. another she matches at 271 i match at 268! at the 2C1R and beyond (3C in your case) it CAN be a crazy low number or even 0! BUT, i have 2 of 200, thats 99:1 ratio or 1% of the time. (or 99% its normal). My initial response would be, if it was you only, that this was one of those 1%ers. HOWEVER, The fact FIVE tested with this anomaly tells me its not an anomaly but instead its probably indicative of something else, aka, the paper doesnt match the DNA. Example: i tested, mother tested. a Neice to my mother tested and i matched her at 9cms (my 1st cousin) and her 3 siblings had 0 cms match (well under 6 ancestry reported at that time). Ironically mother matched ZERO. i actually matched through my fathers side! This indicated Mother had a DNA surprise that i had to work through, which i did. if it was one or two i would say this was in that rare variance, BUT with FIVE? if its 99% unlikely, and it is occuring with all five, id want to scrap the paper trail and try to do that side via DNA triangulation. sounds to me like there is a surprise in there so tread lightly! its a soul punch and sometimes its easier to "accept" it as an anomaly (but in this case i doubt it is, i think there is more than 50% chance there is a surprise in that path.) good luck!
The probability chart (top right corner) needs a clearer or larger font. I'm viewing this on my macbook with a 13" screen and I had to increase the page size - which then cuts off some of the tool - to see 1c1r as an option but it looks like 1cr to me (even wearing my glasses for my stigmatism). I think the relationship choices should be a larger and maybe bolder text
I love this tool! I understand it better than that other one Blaine Bettinger was using, sorry Blaine! Lol. I was really looking for something like this to help explain the different family relationships such as 4xGreat-Aunt, etc... My question is: I am looking for my late mom's biological parents. She was adopted, but I found her family via my DNA matches since she never took a test. Can this tool be used for her line? I noticed in the video you only do the paternal line? My biggest top match is a woman who shares 1,094 cM! Ancestry says she would be my 1st-2nd cousin (I really wish Ancestry would separate out those 1-2 cousin lines) I see this tool also shows that too at 47%. So, that would mean we would look at one of her aunt/uncles? Still not quite understanding that part. Could you explain more? Thank you.
I love this tool, it kind of confirms that a couple of apx 2nd cousins are indeed multi path. Ahhh more research! LOL. Suggestion would be to make the bar graph larger. My eyes are old. But seriously, that is such a small thing. What you have already is amazing.
yeah i HAVE to do something withthe chart, we got the AGE entry improved, on to rest of the TO DO list from it :) lol Still no comments on color blind, either no genealogist are, or its not a problem or no one wants to mention they are, not sure, but i KNOW those greens are not the best :)
Great tool! I have three known half 1C1R's at 325, 311, and 309 cM. Not sure the tool represents the relationships for these values correctly. I was expecting the 1C1R circle to appear in blue, not yellow, to validate the "known." Unless the tool is telling me there is additional path(s) for them? Like the yellow bumps to red for multi-path, does the blue bump to yellow for relationships that start at only half? Seriously, awesome tool.
Thank you for your examples! I just pulled up the dataset for 300-325 it is 0 as is as are the other blues, i put the slice for the 300s in wrong spot! the Chart is right but the Balls in the pattern were WRONG! I have corrected that (Beta, probably will be a few more i messed up before it gets flushed out!) now to your scenario: A) I have updated the tool to properly reflect the pattern in relation to the data for that CMS B) Lets cross check with dnapainter.com/tools/sharedcmv4 . WE enter 325 and find that H1C1R histogram shows 83/56 (blue) to the left and right of our input number (139) and 374/1015 (1389) to the left and right of our number for a 1C1R (yellow) Now this is where my tool looks at things from 3 perspectives at once. the pattern shows Closer to 425 than to 212 (100 vs. 112) BUT it (now) shows it to be a 212 with additional pathing) The CHART confirms the error you mentioned as 1C wasn't even in the statistics for probability! (guess that was a low point in caffeinated drinks - Great catch!) What we have now in the BP tool is 2c, 1c1r, 1c2r, H1c1r, H1C, 2c1R in that order of likelihood. The SharedCM tool shows for: 1c1r (1389), 2C (1147), 1c2r (335), H1C (243), H1C1r (139), 2C1r (4). So the difference is 2c, 1c1r, 1c2r in bp tool and the same 3 in the sharedcms tool but first two flipped in order of 1c1r, 2c, 1c2r. difference being the inclusion of estimated multipathing in the BP Tool and not specifically addressed in the other tool. --- I fixed the error you pointed out and got it into the live production version immediately! (will be out of pocket Tuesday and Wednesday but Thursday i will review the break points and see if i messed up any other Chart to patterns (since it is manual not dynamic on data yet - only so many hours lol the sharedcm tool had Blaine, Johnny and Leah, this one has me. :) Thats why this is Beta though to flush out any mistakes i made! Thanks for catching that one! ---- Lastly, we are wanting to estimate MRCA location. the 2c and 2c1r (two of 3 top in both tools) would intersect withthe DNA matches Great Grandparents, the 1c1r would be just grandparents, so if you build their tree to their 8 great grandparents, you should intersect! if 1c1r, youll have that intersection already (so as you build tree, obviously id look at the 1c1r first on the way up to the GGPs. The Age probability will help you identify which 1c1r, 2c1r it may be. and if you know their age (their parents BDs vs. your parents BDs is a good clue for estimating, but looks as if its most likely (based on default age of 50) to be at YOUR GGP or 2GGP. AND there is likely another, perhaps on other parents side of 106 to be found at 2ggp, 3ggp OR perhaps 2 paths equaling this higher up. At 2ggp and beyond, ANCESTRY projects it to be multi and higher up, which is why they Timber the results at 90cm (3c which the mrca is 2ggp). Their paper on timbering discusses they believe at this point it is more likely to have a multi path than single path for the CMS given, so they "shrink" or "WEIGHT" numbers under 90. (which can effect 106s if there are 2 making it up) --- I am very happy you caught this! Even happier to correct it. Also Embarrassed i let it slip past me even into the Public Beta. (needed MORE COFFEE LOL) But Semper Porro (always forward) its fixed and its getting another step closer to solid as we flush things out in Beta! GREAT CATCH!
Thank you for all your hard work. I have learned so much for you and this video may have solved a family mystery. I was always told not to include endogamous lines in my searches since their numbers are too high. I think my DNA match, whose mother was abandoned as an infant in 1912, is from my double cousin line that I've excluded in searches. My brother matches her 230cM. I'm only 110cM. Our known 2nd cousins are 103 and 167. How is my brothers so high? More than 212 when the rest of us are closer to 106... It could be the double cousin line, right?!? Or am I doing this wrong?
At 2C1R dna can drop off significantly (or very little) from normal. it COULD be additional paths for him BUT id guess the match is a male? This explanation goes into the genome sequencing itself and outside of what i can probably simple explain. But do not worry about endogamy or pedigree collapse when searching! thats what the RED dots are for! i have 3 sisters married 3 brothers in my line, aka all 4 great grandparents on my fathers side are the same for me and for them so they get the "pathing" from each person passing dna along and thus appear twice as high as the same kinship (2C) from other paths including those from same GGPs but without the double (triple) marriage situation. They are 469-488-508 cm (3 matches) and they are all 2C in RED. others at 2C show as 208-233 in match list as would be expected for DUAL path only and one is a 110 (single path) thats what special about this, its all about the paths, not the kinships.
@@DNAFamilyTrees The DNA match is female. Her mother was the abandoned infant. I no sooner made hand made charts for Single Path and Dual Path to put into plastic sleeves for dry erase and then saw your new video. LOL. I went back and added dates of my ancestors and knowing the abandoned infants birth month and year was and to rule out some lines. That really helped me the most. In this area I have 3 generations in a row that are 30 years apart. After adding the dates for everyone it really seemed to pull it together. Now I just need a couple volunteers from the 2-3 possible children it should be to prove it. If I'm even right. LOL
Hi, love this tool! Can you make the age stay fixed unless you purposefully reset it? My age isn't changing that fast and I have to reset it each time I change the cm # of what I am testing. If you're doing a lot of check, it makes a difference. Thanks!
oh so if you "RESET" leave it the same? hmm.. i guess i could do another reset button that leaves age next to the full reset. Great point, i thought about that in my testing but didnt, but since you brought it up, yes i WILL make that adjustment! Thanks for the feedback!
OK, the Button you requested is now live, there is now a "Full Reset" as well as a "CMS Only Reset" option so that many can be run through at a time without selecting age and probability overlay each time.
I botched it up, sorry aobut that, somehow i messed up versioning and replaced the entry with slider. It is now corrected sorry about that, i messed it up.
HI Larry, Thanks for this wonderful tool! I do not fully understand it yet, but from what I think I understand I can see that it is going to be very useful to everyone working with DNA matches. My question is, can you make the tool work backwords as well as forwards? In other words, right now you can set an age and a CMS and find out where to look for your most likely MRCA. But I want to be able to put in my age and the particular MRCA ancestor on the ball chart who I am trying to identify and have it tell me what number of CMS I should look for in my matches list to best help me track down that ancestor. For example, I want to put in my age (76) and enter that I am attempting to find my 3xGGP dual path, and have it show me what number of CMS to look for in my matches. Then I also want to be able to put in my Aunt's age (96), and find out what number of CMS to look for in her list of matches to find the same person, but in her case he is her 2xGGP. I hope that is clear. Anyway, I love all of your youtubes and find them extremely helpful. Thanks again, Jane
Jane, That is What this one is supposed to do for you. When you enter your age adn the CMS of the match, the green overlay shows you were they are going to match you. By going up your ancestor line and down to that location that top spot will be the MRCA. Now to the specific, the ball and pattern CAN do what you want, put in the age BUT instead of entering the CMS you would COUNT to that kinship to know the estimated CMS. Thats what the blue, yellow and red balls do. for instance: I want to know the estimated CMS of a 2nd cousin. you know a 2nd cousin MRCA will be 1xGGP. So count up, parent 3400>>Grandparent 1700>>Great GrandParent 850>>(now down)Great Aunt/uncle 425>>1C1R 212>>2nd cousin 106! Now thats a single path (one MRCA) 2nd cousin, if a FULL kinship, aka dual path (yellow) its that same path through the OTHER 1xGGP (one 1xgreat grandmother, one 1x Great grandfather ) for 212! So the estimated CMS for a 2nd cousin is 106 (half) and 212) full. Ever time you traverse an arrow up or down you divide by 2 (start with you at 6800) this will give you the estimate. You can confirm you did that right if you enter 106 and also enter 212, youll see the 2nd cousin be blue and then yellow
This is so cool! I can’t wait to play some more at length but I did a quick test. I threw in my daughters’ and mother’s shared cMs of 2047, 1637, and 1553 respectively. The most likely relationship for all aunt/niece. What to make of that?
The only DNA line I'm sure of is my mothers line. There is some endogamy. Using leeds method I've worked out four lines. Cant get my siblings to test though even though I know we have different parent combinations. They just dont want to know. I will keep on looking. Hopefully this method helps me.
i think it will, the matches will stand out and the cms is a great wayto determine. hang in there! i had NO one help me with mine (until i figured it out, then dozens tested to either confirm/deny my assertion. I went from 1 match over 100 to dozens lol
yeah i had mine do that to me (even in video) darn slider, need to make that a box! if you refresh the page, it will work, something with the browser handlers does that every so often. definitely needs to be altered to be an age input box :)
Larry I am at a brick wall. I have DNA matches of 1st and 2nd cousins whom I do not know. I have made contact with them and still no trace of how we are related. The highest CM match is 636. Our families are from the same area but cannot find the connection. Any ideas of what to do next.
Very impressed; I do like it a lot. Suggestions for legibility, apart from what you have already mentioned: I'm not colour blind, but find black on darkish red lacks contrast. May I suggest a paler red?
Thank you for developing such a useful tool. I am fairly new to genealogy. I am looking for my great grandparents on my father's paternal side. I have a name for my grandfather, but it doesn't match anything in my matches. My hypothesis is that he either changed his name or was adopted. Using my DNA matches, I have come up with families for my great grandparents, but they were not married. I have narrowed it down to families, but I haven't been able to determine which person in the family is my great grandparent. I don't have any 1st cousins on my father's paternal side. The closest cM are 160 and 105 on the paternal line and 131 and 120 on the maternal line. Do I only look at the single path for all of my matches?
Mary, as i mentioned ot steve a trick i often use is to go the "other direction" so instead of dividing using the pattern i go the other way and multiply back. So we are looking for Grandparents in relationship to the DNA matches. If we take the 160 and 105 my initial assumption would be the 160 is a multipath 106 and the 105 is a single or dual 106. So id enter their numbers and use age overlay but .. this is where we get fun... This match is to YOU, right? so, we can get one step closer by looking at them to your Father. (if this is your fathers dna match then its his father and just adjust accordingly what im saying). this means we change the number from 106 to 225, this is the kinship to your father to that DNA match. Now lets double again, to get to the DNA matches parent (we dont know or care at this point which parent). this is your parent to their parent. Now we are going to do it again, once for each side, double then double again to get your grandparent to their grandparent. this gives you a VERY small area in which to focus to make a connection. If you know birth years of your grandparent and theirs (they have 4) then you can estimate if they are older, younger or about same generation (if you dont know you might estimate you to the dna match based on profile picture or informaiton that they shared). if same generation then the grandparents of each are likey sibling This would incline me to lok at your 2x ggp as the intersection. now back to the original 160 cms, we guessed from the doubling exercise that the mrca is likely at 2xggp. This fits the 160 as a multipath 3rd cousin, aka a 106 with multiple paths making it a 160. WE assumed multipath since over the 106 significantly, so all things lead me to think the mrc will be their 2xggp and your 2xggp at MRCA with other paths also present. I would look at 2xggp really hard :) build out those matches trees (yep, we gotta do the work for them) bulid those trees to their 2x or 3xggp. Then xref this with your DNA clustering to see which group of 2xGGPs it is for you. one more trick i do in this case is pick the dna match at 160, do shared matches, FILTER ON PUBLIC TREES, then right click on every tree in the list and open in new tab. (i do about 30 at a time) then i look at the trees for each going tab to tab to tab to see repeating surnames or identifiable duplications. in most cases there will be one or more overlaps, then focus there as those to overlapped each other and you intersected to both of them so its very likely you intersect on that duplicated kinship (marriage or remarriage name location). hope this helps.
@@DNAFamilyTrees Thank you so much for your response. I will have to take your recommendations step by step. The 2GG is my hypothesis as well. Since it isn't my GG and I have matches to 3 of the brothers who are less than the 160, am I correct that It should be the fourth brother who died at the age of 39, wasn't married and as far as I can tell, didn't have any other children at least none that I can find DNA for?
@@mjrose-r9y we cannot infer he didnt have children as there are people who dont take DNA tests for personal reasons, just as today people choosing to or not to get a shot. it simply means that IF THERE ARE kids, they havent tested. No kids is a real likely possibility but the data much to limited for that inference. it was SEVEN years after DNA testing came out that my relatives that provided the fact my tree was not biological after all and they were two years behind me so i went two years thinking everything was as I assumed. then one day, pop, like popcorn kernels this family line started popping into my match list until i knew with certainty that i had a DNA surprise. So i would say it just means no testers from that finite set. (we could infer some socio and economic things inside the USA or privacy things if they lived outside the USA. As for looking single, well, in finding a kinship we know ALL PEOPLE that DNA match have at least ONE path. More often Dual, and often multiple especially dependent on locations. So I would say initially look for that ONE path that connects then work back to see if more (youll know right away when you connect the one). those numbers on a single path would likely connect at 2c, your GGP and their GGP. This would be my first focus, however, if DUAL, if they are older then STILL their GGP, but if younger than you then still YOUR GGP but their 2GGP. In both Dual scenarios either YOUR GGP or THEIR GGP was the intersection so if you work out trees to get both of your lines GGPS, we can then infer that 1 or 2 of those GGPs will be the point of connectivity. on CMS i tend to focus on finding a SINGLE path first because to have dual or multi there must be a single so single is really the main focus, and i treat the dual as an additional single BECAUSE it may not (usually does, but may not) come from the spouse of the single! That second path can come from any direction, even from other parents of one or both of you. So yes, id focus on single to see if you can locate dual if present. (fyi, the 160 is MOST likely a 106 but dual or multi since at least 1 path is part of that 106, could be 2, but we can infer that the 160 has an additional 53 path from somewhere that makes it a dual or multi, we dont yet know.
I have a 2C1R who shares 567 cM with me. My great grandparents are the siblings to her grandparents so it doesn't even give me the option for that. Otherwise I like it. I like that it takes age into account.
If this is indeed a 2C1R and not a surprise story to be found, then there is a SERIOUS amount of pedigree collapse! For that to happen at that CM you would have to have all four great grandparents the same (sisters marrying brothers) AND have a close match on the other side of your tree. if this is not the case, then there is more to that story. Walk softly.
@@DNAFamilyTrees I know my lines cross I'm just not sure where yet. Her first cousin I share 327 with which I know can be in the normal range of a 1C1R. It was the 567 that threw me off for the longest time.
Still trying to understand all this, but I'm excited to learn. Suggestion: Can you put the numbers of cM's each generation and path divide or multiply up to inside the balls for quick reference?
LARRY PLEASE HELP!!! I am matching someone at 209, my full sister is matching the same person at 130, and my other full sister is matching the same person at 64cm........ What would you make of this???
💖💖💖 Speaking of Ancestry... I see you've connected a bunch of your matches in your tree. I made a chart as I did mine when that came out and the actual relationship is often WAY DIFFERENT from the most likely relationship according to the Shared Centimorgan Project. 🤔
Oh wow! email me those matches Cheri, id love to take a look (dnafamilytrees@gmail.com) I just responded to NYCSue, the answer to her i think would also likely fit your findings in similar ways.
Very helpful. I have 4 DNA matches on my paternal that are about 225 cM. They all show as shared matches in Ancestry. They all appear to be a generation below me. Only one has loaded to MyHeritage. Adoptions are present in 2 of them. It appears my MRCA is my paternal great grandfather but I am lost from there down. Side note: My 4th GGP's on my paternal side are also my 5th GGP's on my maternal side. Addition request: approximate age of target entry field
just based on this information, i would estimate this is a 212 with a bit extra from the other side as you mentioned. One trick i plan to do a video about is REVERSING the concept, divide by 2 going one way and instead DOUBLING the other way. Taking the clue they are younger, i would DOUBLE that 212 to 425 to see their parent to you. If they are one generation down then this would insinuate their parent being equal line. that would be 425 as single path cousin or multipath 2nd cousin. NOW we do that again, double that number again TWICE, once for them, once for you. so that is 850, then 1700. This would be your parent and their GRANDPARENT on equal lines. That insinuates a single path sibling. That then insinuates your grandparent is their grandparents parent in a single path scenario. NOW if indeed your Paternal great grandfather is the connection then the math really goes a bit crazy. Back at the 212 then they would be the #1 probability in the chart at 2c1r. Their Parent thus your 2nd cousin dual path. your parent to their grandparent would be cousins. your grandparent to their great grandparent would be siblings and thus connect at 1ggp as you mentioned. Which scenario is most likely? well if normalization is tending towards mid then for that 225 to be a 212, thats 13cms from a single additional path. That can be indicative of multipath 2c1r. BUT thats a pretty small variance at 13 and since you know of another path via the other parent that probably accommodates 13cms then id be inclined to think that for the MRCA path you seek its NOT multi path. So they are likley two gens down at 1c2r or one gen at 1c1r. The above scenario (1st one) demonstrates the single path option. I said i would go Yellow, Blue, then red as habit. There is NOT a yellow in the generation below, you have to go two down to get dual. So then the question becomes which is most likely single path or multi FOR ONLY THE ONE PARENT in question. since you know a legitimate path on other parent, then its unlikley its multi on this side. so again, id look first at single path 1c1r and intersecting at your Grandparent. Other factors in play, (identical twins, etc) that make the number act differently, but identical twins are 0.003 likely so 99.997 that its not one of those :) IF they are multipath Then this also means... its a 106 multipathed to 225 which is highly unlikely. remember blue is single, yellow is dual, red basically Quad. So if there is a MRCA with multi then its actualy more likely to be up at 2xggp statistically. BUT id look at GP first, Their GGP, cross reference with cluster information should identify for you which of the intersections it is. hope this helps more than confuses :)
I have a great-grandfather who was an identical twin. cM's are out of whack for my matches who are descendants of his twin. Just another scenario to consider when looking for another path for those matches who don't fit the expected cM range.
yes identical twins have have aunts show up as parents pretty often. But if you find that on a line you can simply REMOVE the division component to that descendancy line 1x and get a pretty good estimation for cms. My mothers adopting mother was an identical twin (she used to tell me stories how one was good in math the other loved english so while in university one took all the math classes (2x) and the other did the english (2x) lol
Gotta love it when the "experts" "know" more than you do. one thing i have learned is there are NO ABSOLUTES regardless of what paper trails say, regardless of what the DNA says, there is ALWAYS that exceptional condition! I have talked with Subscribers who one for example has had 6 children, obviously a women but FTDNA will NOT ALLOW her to designate herself(her own DNA) as such! She is one of the rare people whose Chromosomes actually changed from XX to XY! so FTDNA will only allow HER to upload as a MALE! (and she is not the only person whom i have spoken to with this change over time!) now i say rare, truth is the actually percentage it occurs with is UNKNOWN because over history we havent tested DNA like we have in recent years! And she doesnt feel or act like a male in any way shape of form so unless FTDNA told her and she investigated she like many others would NEVER have known!
thanks for pointing that out, was two data series combined into that chart. (i need to do follow up video on this specifically) there is a point in the ancestry dna match list that they change how numbers are shown right in mid list! Most if any people dont realize this! i do need to raise awareness on that.
Thank you Larry. This tool is great. I just want to ask about one test I ran. I have a 3rd cousin descended from a pair of 2nd great grandparents. I share 44 cM with him. This yields 3rd cousin, single path, or 4th cousin, multiple paths. My dad, who is his 2C1R, shares 237 cM with him. Results yield 2C1R multi-path. My sister matches him with 69 cM, also his 3rd cousin, and results yield 3C dual path or 3C1R multi-path (the highest probability). So, I'm thinking it looks like a multi-path relationship, but I haven't found any other shared ancestors through my genealogical research. THOUGH, he has an ancestry that involves a lot of intermarriage in a single community.
Richard, This is an excellent question, one i hope people read. If you have found your fathers MRCAs at 2ggp (dual) then your solid. This was created to narrow the place to look so that someone can get to an MRCA in shortest time and then build down to find Biological parents. At 2c1r DNA gets TRICKY! normalization tending towards mean gives us great estimation on where ot look BUT There be dragons so to speak at this point. It can vary WILDLY from 0 to Multipathing large numbers depending on each unique scenario. My mother matches someone as a 2C at 278 cms, i match them at under 6cms (they are not on my match list) another she matches at 268 cms i match at 258 cms. (now my 258 is obviously from another path via my father, but on the flip side not showing up is indicative that it was a non normal split all the way down that line, it happens, not often as other cousins on that same path match in a normal range. He may have a yet unknown DNA surprise in his line that he has yet to find out and correct (like i did with family #1 before i had my dna surprise and changed my tree accordingly) or it could be a strange split at every level (statistically its most likely a DNA surprise). if you put in 44 cms on the shared cms tool and then put 0cms in the same tool the answer is almost identical for what the possible kinships are! This tool is in conjunction with as it has an AGE band for probability, the ball and pattern that is highlighted based on normalized single, dual and multi (which statisitically is more accurate than not) and also chart that shows data from multiple sources rolled into one that at various cms ranges contradicts the normalized result. That, like the story i shared on my family above, can be from people not yet finding their true DNA family and believing their tree accurate or evidence of a bizarre split at each level which is possible (and shows frequently in the database, BUT i suspect its more evidence of unknown DNA surprises more than wild variants. as AtDNA has very few mutations even at 4 and 5 generations! AKA should not vary from normalization much. So for it to vary at every generation while statistically possible, I believe it FAR less likely than reported and the reports are people, who like me at one point, didnt know there was a DNA surprise. There have been people in comments question some results, but i can see that they are about to uncover a surprise and i let them do that at their own pace as i wish that on no one! IMPORTANT: If your dads match was a 237, and a known 2c1r to him, its very likely a multi path. (well he is, lets just go ahead with that) and if there is not duel marriage in the line or a remarriage or endogamy on that line, then its likely through your fathers other parent. his 237 is a 212 level. meaning 106 from each of those two and 35 from another (with slight variance of course) This would lend to 3ggp on other parents side maybe 4ggp and/or 5ggp on that side that in combination added to the 35ish range. It is very possible then for that line to NOT show to u being so far back and youd expect to be half of the remaining 212ish or 106ish. Here is where it gets tricky. TIMBERING. if say your DNA was just under 90, still close to 106cms that you would expect, Ancestry would TIMBER that to a much smaller number, perhaps as low as 44. To see the PRE-TIMBER cms score that is in line with your fathers you will need to click on the DNA match CMS in the dna match list, there you will find an UNWEIGHTED CMS number. I am guessing its 60-70ish cms. If so then while that drop off from your father to you seems giant its simply drop off on the one line and slightly lower than a perfect split on the known line. Let me know what the Unweighted CMS was for that match, i am very curious!
@@DNAFamilyTrees Hehe, Larry. You've given me a lot more work! I should have mentioned that these data are from FTDNA. They come from a study I did a few years ago to determine whether a family in Illinois was kin to my paternal maternal great grandmother. I started with a Y-DNA study of descendants of two known brothers in Illinois and descendants of my grandmother's two brothers. Sure enough, they matched. I then had a descendant of the Illinois branch of the family tested for autosomal DNA, and this fourth cousin also matched--we got 18 cM. From my great grandmother, we descend four generations to my daughter and match a descendant of my great grandmother's elder brother. I was using your new tool to analyze the relationships. The common ancestors are Thomas Liston and Letitia Sinnett of NB, Canada. They had three children. Each has descendants who have had DNA tests. The cousin in question is descended from their son James. My family is descended from their daughter Mary Elizabeth. The cousin is four generations from our common ancestors, and this would be a dual path (from Thomas and Letitia). Following is my family line anonymized with the cMs shared with the cousin and the documented relationships (on paper). Mary Liston Case gr gr aunt dual Hazel V Case 1C2R dual my dad 237 2C1R dual me 44 3C dual my dau 60 (!) 3C1R dual I don't think FTDNA timbers results, do they? My e-mail is triovlaif@gmail.com, I'd love to be able to show you visually what I am working with. Thank you so much for your lengthy reply.
My mom has 4 half siblings. No full siblings. 2 half siblings through her mom and two through her father. I have a match from my mom’s side of family that is showing 465 cM. So my 1st question…all the children of the 4 halfs will be my half 1st cousins? All relations of those people will be half-whatevers? So if my grandfather has only full siblings, their descendent will be Full relations with me? Or are they halfs too since my mom’s 2 siblings thru him are halfs…everyone is half? I tend to think that 1/2 relation is only of those people and the kids of grampa’s siblings are full. Help? This makes my brain hurt.
This one is confusing for alot of people. Descendants of your mother or Father will share ONE PATH to you, DESCENDANTS. Your mothers parents however, if not remarried the descendants of them would most likely be DUAL path through both great grandparents if that is the intersection, same on up the ancestor tree.
So Larry I always have to be difficult but here is a problem that I have puzzled over. I have a set of 3C. We all share the same second Great grandparents. There are three sisters from one descendant and one from another, they share great grandparents. I know them, have documents etc. One sister is 20 years older than her siblings so they are late 40ties she is already in her 70ties. Here is how they match me 13cM, 29cM and 42cM. Then the other cousin, who is their 2nd cousin, but my 3C, he matches me at 230cM. Yet we all share our second GREAT Grandparents, dual kinship, well documented. I know technically from DNA Painter that one can go from 0-234cM but I think it skrews with your ball and pattern tool. What do you think? Have you ever seen such a spread in cM for the same level of kinship?
yes it is possible to have that disparity. RARE, but possible. its like rolling snake eyes with dice 4 times in a row. it can happen, will happen sometimes but its rare. 1/36 x 1/36 x 1/36 x 1/36 for snake eyes 4 times in a row, or reduced 1 in 1,679,616. Not sure the actual odds of that disparity but after 2C1R (beyond 6 steps) aka 7 or more, it can range from 0 to nearly 250 for a 3C! But those are RARE occurrences but they DO happen, in fact i have that 2x in my tree with a 2C1R on my mothers side (dual path)! My mother matches 278 as 2c, they do not show on my match list! another 273 does not show on my match list! and another shows 268 to mom and i show 263 to the same person (turns out i have 2 more paths to them via my bio father)! At 7th step and beyond there can be WILD variances to the low side and even more rare, to the high side. That said, it is rare, so in this case, your matches at 12, 29, 42 cms matches would be all 3 on oddly low side unless single path (remeber its rare for one but then you roll the dice 3x and normality returns, or it should. While its possible to get snake eyes 3 in a row i darn sure wouldnt bet on it to occur LOL. your 230 is BROKEN high. like one of the highest recorded ever for 3c kinship. DNA painter shows only 9 data points at that range, while ball and pattern gives it 4% likelihood (much higher probability, BUT ONLY with multipathing!) (in dice terms this is like rolling a 3 with two dice, very low odds, but without multpathing it would be like rolling a 5 with 5 dice - Happens in Yahtzee, but id put my money on any number higher than 5 lol). its POSSIBLE sharon, but my gut tells me there is more to that to be found somewhere. To have that number at a 3C lvl there is more than the 2 paths in play for you and that match. with the intermarriages you told me about previously, that would be enough for me to think there is a "reconnection" by lines somewhere, perhaps still undocumented.
@@DNAFamilyTrees Lar thought exactly that same thing so tested my Mom and her sister. No cross match. Can only assume the cross is in the paternal line but only way to see that would be to test a 2nd cousin who is on the paternal paternal line. I tell you the endless aggravation. What about the three sisters? I also have another 2x3C matchs to this whole line same set up and they are at 24cM and 60cM. So everywhere from 13-230cM for the same line. The eldest son of my 2nd GG. Really?
Hi Larry, Just so gratefull for your lessons. You are amazing! I am so close to making a breakthrough, but I am still stuck and have been researching for decades to find a dad. May I contact you by email and hire your assistance please?
emailme at dnafamilytrees@gmail.com i will review what we have and maybe we get lucky and we can post your answer right away. if not Ill see what steps i can recommend for you to have success
If I am understanding correctly if I have a cousin match at 626 CM's then their parent would match at 1252 CM's or around that number, now would they also be a cousin match or half sibling?
Yes and no. Probably more because the parent would be a step closer. The only true 50/50 split is from you to your parents. All other relationships are based on what amount your parents gave you through recombination. The halving beyond parents is an approximate assumption. My sister is 1241 cMs to a 1st cousin My brother is 924 cMs and I’m 948 cMs. The average to a 1C when split half-half-dual path-half is 800-900 cMs and we are all above. If his sister, my other cousin, did DNA, it’s possible our shared DNA is under 800 cMs. Therefore if you have a cousin match at 626 cMs, it’s in the 1st cousin range so you should imagine that person as 850 cMs and their parent at around 1600-1700 cMs. If someone related to this person came in at 1252 cMs it could be the person’s sibling and you are mistakenly thinking it’s a parent. Oh it can be a bit confusing. I have a few 1C1R s that are low cMs and not falling into the B&P suggestions. this is why I suggested an advanced feature to include longest segment. It’s possible records and findings of full kinship are actually half kinships but until I find proof along that path, I just chalk up to low cMs.
626 shows as most likely a 1c1r (typical dual 1c being around 850) but yes, each time you go away from you, dived in half and if you go TOWARDS you, you would indeed double and that is a POWERFUL insight! because this is not only for THEIR parent, but you can do it again for YOUR parent and even better estimate kinship because the higher and closer it is, the less the margin for error. so in your example, you have lets say an unknown 626. you then could estimate their parent (unknown which one but one of them) is approximately double that at 1252 to you (always bearing in mind that that 626 is likely a low 850 or a high 425 ill refer to this in a minute). BUT you can then do the same for YOUR parent (you dont know which one but not important yet. DOUBLE that 1252 to 2504. This is the estimated kinship distance between YOUR parent and Their Parent. This gives you a VERY finite set of options with VERY clear likelihoods. (so in that example their parent and yours are most likely siblings if near the same age. now lets look again at the 626, lets then say its a high 425 (example i have a single path 1c who is 441, her daughter is half that at 222 but her son to me was 387! turns out this is due to the mcfarlin path that my mother ALSO has that hte set he had matched me more than the daughter who was less on that mcfarlin path than her brother) so if they were a high 425 (due to undiscovered additional pathing) then double that would be 850, double again for your parent to 1700. Now we still get a very finite set of options! the parents could be an Aunt/uncle (reciprocal niece/nephew depending on which was older or a single path sibling (half sibling) ) So from what youve said we know that the person you match at 626, their parent is your parents aunt/uncle (or niece/nephew) or is a sibling or half sibling (most likely single path sibling since unknown in connection. Additionally we could deduce that if their parent was your parents Aunt/uncle then one of the dna match's grandparents (their parents parent) is your parents Grandparent as well. (clustering should point to which side to look and look at males for missing lines first as its harder for a womans child to be born without notice than a males as often he may not even know. if they the parents are siblings again the DNA matches Grandparents will be the intersection but would be your parents parent. Now this is where that probability map helps based on ages of those involved, even though it may not eliminate aunt/uncle or niece nephew the green probability will tend up or down based on your age. that is the direction that recipricol kinship is most likely to be. hope this helps! i OFTEN use doubling to estimate because that helps laser focus by changing who your looking at! so looking at their parent and your parent is a GREAT way to make that number more concrete to help you focus on were to look! so while the 626 itself shows ALOT of options and the chart helps focus on most likely, go up one for their parent and one for your parent (double and double again) and use that number to connect your parents and that becomes a MUCH cleaner task, then take that knowledge go back to the 626 and see how that would fit. as i said, i do this ALL THE TIME to help me decide the best place to start looking! (so if the parents are say aunt uncle to each other then your most likely 1C1R to them (which is what the estimator shows as longest bar in the chart far longer than other bars, followed by single path 1C, but since a single path 1c would mean parents a single path aunt/uncle and 1700 is a dual, we know that probably not right if it was a high 425, now if it was a low 850 on the other hand that goes 2 nodes closer! doubling their parent would make 1700 and doubling yours would be 3400. This would mean your parent or theirs is the parent/child of the other. if within 15yrs of age of each other that is incredibly unlikely. So based on this, you pretty much have your answer of how they fit (and thats the beauty of this looking at it from path perspective :) when you prove it out, let me/us know what you find! i would treat them as a HIGH 425 and estimate from there which means they are highly likely to be 1c1r to you with your parents being A/U and N/N to each other. 626 (their parent being your dual path (full) 1C most likely. (but thats without age component which is why i added that, that could also be 1c1r and their parent be your Great A/U as there are two 1c1r locations in a tree. Now the 1C1R locations in the tree are ABOVE and BELOW your generational line SO... if the match is older than you, their parent is probably your great A/U and if younger probably 1C. Said in a more useful way, if older than you their Grandparents will be your MRCA (2 of their 4), if younger than you, YOUR grandparents (2 of 4) will be the MRCA. again hope that all made sense and helps.
@@DNAFamilyTrees Okay thank you so much this is very helpful, I have two siblings in my family who took a DNA test for me to help me out with my family history research, they matched at 1925 CM's, they have lots of known maternal matches in common like cousins they grew up with & sometimes lived with, the younger sibling has the 626 CM Cousin match this cousin does not match the older sibling or any other family members on the mothers or fathers side that I can see with the exception of the adult children of the younger sibling & they match at 334 CM's & 336 CM's, this person is a surprise adopted cousin match and their biological mother is only 9 years younger than the younger sibling they match with, the older sibling has paternal matches on Through-Lines with the highest match being 149 CM's they have all a 2x great grandparent in common, the younger sibling has none so far, I did find out that when the younger sibling match was a teenager in the 50's for about three years or so they had lived about 2 blocks away from where the biological mother of the adopted cousin match was living at the time of his birth in the 60's, that is the only connection I have been able to find with regards to the younger sibling & the surprise cousin match. P.S. no one in the family recognizes any of the last names associated with the surprise cousin match "IE." their birth/adopted name & their mothers or fathers names, nor any of the names of the DNA matches the adopted person has in common with the younger sibling & their adult children.
I am 78 my highest match 404cM. Thought this was a half 1st cousin but your tool is saying more likely 1st cousin 1 removed. Am I reading this right ? Perhaps I need to re-watch this video.
Hugh, that is correct, the most likely is a dual path 1C1R, 2nd most likely is arguably 2nd cousin multi paths or a 1st cousin single path. So if 1C1R or 2nd cousin this would mean this persons great grandparent is your grandparent. According to my tool that is the most likely and 2nd most likely MRCA (their Great Grandparent) --- i looked at the shared CM tool as well for confirmation, if you enter 404 and in the results click on H1C you see it has 180/243 on either side of the mean of 404, for 1C1R it has 1015/1196 on either side of 404. For a 2nd cousin (which would be mulit path 2c, sharedcms has 197/92 on either side of the mean of 404. So their results appear to confirm this assertion. -- What i think is important in determining MRCA is recognizing the required kinship to overlap. in this case 1c1r is 5x more likely to be the kinship (statistically) than H1C. 2nd cousin according to my chart is #2 most likely. This means statistically the #1 and #2 would mean that that persons great grandparent would be your grandparent. If you know your 4 grandparents then you have 2 of his 8 great grandparents, If he has the 8 and your looking for yours that means 2 of his are yours. If however he has all 8 GGP and you have your 4 GP and none match, there is a DNA surprise in store for someone. OR, AND THIS IS IMPORTANT, OR its a multipath 2nd cousin with 4 or more paths and not a clean GP/GPP intersection! This means you both would need to build out your trees another generation or two at which point if there isnt a surprise and this is simply multi path the names WILL cross at a generation or two above what was previously expected. That multiple path scenario is a Significant factor and according to my tool the 2nd most likely. in all cases YELLOW, 1C1R is statistically FAR more likely than a half first cousin. (Statistically. for the sharedcms tool the numbers would be yellow 2211 vs 423 for blue (h1c) and 289 red. Mine is more in line with 4583 yellow, 833 red 731 blue. Your looking at 1C1R being 5.6x more likely in mine, and 5.2x more likely in the sharedcms. in mine RED is 12% more likely than Blue according to the chart. the sharedcms tool has it 32% more likely to be blue than red. (but remember it also states they do not accommodate for endogamy or pedigree collapse in their stats) So i would recommend looking at Yellow first for sure *his ggp and your gp* THEN id look Blue (as i said in video i look at this practically too, a relationship of 2 people is 100% required in all scenarios. Then a divorce or death and remarriage relationship is subset of that and depending on location blue is more PROBABLE than a red. (but in some areas and family lines id look red before even looking at yellow if you know what I mean. I hope this wasn't confusing, but in short, yes, 1c1r is 5x more likely in both my tool here and in the sharedcms tool on dnapainter.
@@DNAFamilyTrees Hi Larry, Thanks for the reply. This connection was a surprise to me. My mother's father's side on paper does not have any DNA matches to me. I have been doing research for many years an the records show that by great grandfather spent many years in an asylum and his mother and grandmother did as well. The records show that this was heredity. I don't believe either my mother or I had this problem. My grandmother was living (in service) in the same small town on the 1911 census to my suspected unknown grandfather. He is the grandfather of this 404cM match. The great grandfather of this match would have been 67 at this time so unlikely to be involved with my grandmother who would have been 24. My mother was born February 1912. I guess is that my grandmother wanted a child but did not want the possibility of having one that was affected. I have further matches on the suspected line at 214.4 cM and 189.1cM on My Heritage. Highest probabilities being 2c1r and 3 cousin. Perhaps I may have to go back over my thought on this one based on your system but definitely re-watch your video a few more times to get my head around this. Thanks once again for your interest and thoughts.
@@hughwilliams7361 Since this 404 is on the "missing line" i would of course "cluster" all the matches with him. (now assuming ancestry that is easy and clean to to do). then I would "filter" that list by "public trees" which will give you matches with him of ONLY those with public trees. Then i would go down that list, rick click on every one of them and "open in new tab" the tree for every match (up to 25 at first). Then i would click tab to tab to tab looking at trees and just read the surnames on the farthest right of each tree aloud (or verbalize in mind). [thisis actually important as it helps trigger recognition of repeats] Not any repetitions that occur in surnames, you will typically get 1 to 3 that are noticed. this give you a place to focus for their connectivity, OFTEN you will spot a marriage between two people in seperate trees with different child names, this means you are very likely under that node since that is their MRCA and YOU also connect to each of them so it will be above or below that kinship. Use CMS and this pattern to determine, are you UNDER, AT, ABOVE that node. Then focus there until the answer found. if you dont spot overlaps in those 25, open the next 25 and read the names there. if still no overlaps, change the filter to "unlinked trees" and do the same (bit longer since you have to OPEN the trees on each of those profile pages to see the tree. to be honest ive never had to go beyond this point for a clue except in foreign countries with few documents and trees (like some research i did with an Egyptian line and even then there was overlap but just not useful overlap because i had no way to research what id found) Sounds like your on the right path and though process for the events, your close! Good luck!
Thank you Yvonne! no one had answered that yet! i did run it through the "simulators" and found that the colors chosen actually looked pretty good for all the variations of color blindness, thankfully! (would have been a bit to redo them so i was quite happy about that!) Thank you for letting me know! www.toptal.com/designers/colorfilter/ is what i used. Thank you for confirming it because what i used was a simulator not validated!
Thank you for this!!! I’m trying to figure it all out with my Ancestry matches. Quick question: I have 2 cousins who are siblings. Their mother is my mother’s first cousin. One of the siblings is 288 cM to me and the other sibling is 173 cM to me. Can you help me figure out what this means?
congrats on identifying their kinship! What this is likely to mean is that you have another path to them other than the one you found. This will be much farther up the tree and could branch at any node between you and them via the "other" parent at any one of those nodes. I have a 1c1r at 441, her daughter is 209 but her son is 358. I noticed some quad path cousins at higher than normal CMS and i matched ALL FOUR great grandparents on that side so i was curious why they would be nearly 100cms higher. I then found my mother ALSO had ties to two of the family lines of those 4! mcfarlin and brown MUCH MUCH farther up so it was 2 more shared paths to them! (which is why i am adamant about usage of terms single path, dual path, multipath vs. the arhaic terms of half kin and full kin and yeah there is endogamy. those terms aren't useful at all for research. On the example above my mother ALSO has the same two lines to them HOWEVER it is diluted (for a different time/video) but since i anticipated stronger passage male to male and female to female, and the CMS MRCA is vias the mother, i created a tree with fathers line and found another connection to the spouse of the MRCA. So there are dual paths but only ONE mrca in this case but it was closer and on other side of moms line but i inherited a good amount from the line that the Son in this inherited via his father (his daughter got it too but to a much smaller degree) At first glance one would think maybe half siblings, but it is not, just simple inheritance at a distance :0 but in all cases it was multipathing that accounted for the variance.
I have a 1st cousin 1x removed from my paternal line. We have spoken, she is in her 80s and we have matching trees. The problem is, our shared DNA is 234, which to me points to a half 1st cousin 1x removed. My paternal line is linked to Malta, and in my cousin's ethnicity report, she gets 6% Maltese (which is spot on for our tree). In the previous ethnicity report I had
well a 212 (234 with more from somewhere) would work to be at least a 1c1r. You provided her age without yours but since you provided her age i would then assume you felt it wold enough to mention thus 1 or 2 generations above you. Which you suspect (aka your age) is important for first estimating. Ill go with one generation down. This would mean HER grandparent(s) intersect your line as MRCA(s), at least one but obviously 1 for sure. I am guessing you also know this and has been identified since you have "matching trees". Her match is 212 to you along this line and she has 234 to you which means another path. Since your trees match and based on what you said, I deduce the 212 is a single path. This means there is another path between you and her that is much smaller. we know at least 22 cm estimate perhaps slightly more since it is rare ot have a perfect split at 212 (or conversely less also possible but for estimation we will run with 22). We "know" (or infer) this path is then a "single path" which helps where to look. 22cm single path appears for us at 3c1r with possible intersections above and below and since we "know" its above from previous information we can then infer the additional path will be most likely at here 2GGP and your 3GGP. (pretty far up but again we are looking for 22. NOW there is another option, EQUALLY likely is that your related via your OTHER parent and perhaps even HER other parent so you 4 "paths to evaluate" your same parent line via that same line farther up, that same parent for you to her other parents line, your other parent to her same parents line farther up and your other parent to her other parent. The match is most likely at her 2GGP (any of them!) and also any of your 3GGP (any of them from either side), it will take a bit of looking to find that match. OR that 22 could actually be DUAL from two slightly off from 13s rather than from a single 26 but good news is w would be looking at the same area but instead of 3c1r it would be a 4c. Your line would still be a 3GGP but on hers instead of 2GGP it would then be 3GGP. And lastly it could be simply a high 212 :) we look at the others are probable based on the ethnicity twist you included. But sounds like you really get this so i have no doubt youll figure it out before long! Key to this one will be the A) Ethnicity in combination with B) Clusters. Thats how youll decide which line to find that 22cms on.
Let me also add, that I have no close matches on my paternal line apart from this cousin. In fact, this cousin is the only match of 3 that are above 100cms (the other two are maternal), and the next highest match which I suspect is in the same paternal line shares 32cms, though there's no tree, so no way of finding any link.
how many people share DNA with this DNA match (even at lower CMS)? one trick is to go to that persons profile, click shared, THEN on the shared page, filter on "public trees". Then i right click and openin new tab every tree (30 at a time) to review. I look for duplicate names or marriages (or birth/death dates as people OFTEN changed name upon coming to America for a fresh start. Maybe you can glean something useful this way?
@@DNAFamilyTrees My 1c1r cousin shares no matches at all. The person who I think is most closely related on that side of the family also has no shared matches, but other lower matches have him in their shared matches, so he's part of the same cluster, but he's the highest CM match of them all. It's all quite odd, but I seem to recall it was not like this before a couple of updates ago. My father's paternal, and his grandmother's line seem like there's no hope to find any link in DNA. Malta is such a small country, and his grandmother was born in Konigsberg (now Kaliningrad) and I have been unable to find any records of her birth or parents - I imagine becuase of WW1 and 2. I'm British, so most of my tree is easily tracable, and I have linked up trees with many of my DNA matches from the British lines. It's just the ones outside of the UK that I cannot trace so well, and it seems like either the line didn't result in many more closely related current living relatives, or they just have not taken the DNA test. edit: In relation to your earlier question, I'm 35. My mother has taken the test, and she does not match with any of them.
442 is likely a “Half” first cousin aka that person’s grandparents (1 in 4) will be same as your matches Next possibility is a 1c1r in which case the match would be a generation older/younger which would determine whose grandparents (the others great grandparents) Long shot is 2nd cousin but only if there are intermarriages Not likely
My husband did 23andMe with 3 of his kids. All half siblings. They couldn’t get one’s dna so she couldn’t participate. But the other 2 matches at 2034 cm. I just did mine. I have a half sister as well. We matched at 2023 cm. So imo idk how a half sibling would even show up that low. He (1985) also has a 1C1R (1972) shares 519 cm his first cousin (1980) is 5 years older than him and shares 585 with this 1C1R. Idk where to start. All of our matches are people who I don’t have in my tree.
th-cam.com/video/3ilnR5Q8nb0/w-d-xo.html based on last post and this one i recommend this video. as for the half sibling, they are a "Single path" to you. aka via one parent. so your parent is 3400 to you and they half of that 1700 (rounded numbers for simplicity sake as the actual numbers take into account the amount of ancestral DNA passed to you via that parent that they also received.
This is really helpful, thank you so much! It's funny though, both my sister and I share 2,193cm and 2,244cms with our paternal aunt. It comes up as our aunt by way of duel path. However, my sister's daughter shares 1,298 cm with this our aunt ( her great/grand aunt ) and it shows a duel path relationship as 1st cousin or a multi path great aunt. I admit I'm just learning all of this, but this confuses me :)
This why i love this tool, it goes beyond the "they are related at this kinship" and tells us more about the relationship that can be useful i other research as well. This is simple but complex so ill try two ways : in the simplest format, you to your parent would be 3400. (one step) you to an Aunt would be 3 steps, up from mother to her mother and then down to the Aunt. thats 3400>>1700>>850. (this would be single path or "half" aunt.) if the Aunt has more than the 850 significantly this means there is an additional path to her so we add another 850 from the other parent. which gives us 1700, the mean or "average" expected CMS in your case and that of your sister you have yet even more CMS. another nearly full paths worth. This is typically due to additional "pathings" at a distance that can be on the same side as the same parent that came to you OR via a path coming through the other side of your parentage. The second scenario is EXTREME randomization from a grandparents to BOTH your parent and your Aunt. Instead of passing a typical near 50% from each of their parents on, They passed along 50% more of one of their parents lineages than the other to BOTH your parent when they were created and to your Aunt when she was created. While singularly this is uncommon its not so uncommon as to raise an eyebrow. for it to happen again is quite unusual. How unusual? well if you put the numbers in the Sharedcms tool, their data set for aunt/uncle has 2695 people only 12 are above the 2244 and the 2193 are only 53 above it so in perspective thats 0.004 and 0.019 which is EXTREMELY rare outside the case of twins or identical twins (Identical twins are 0.003 to compare to the 0.004 number for how rare that is. So while it is possible this is a "normal dual" and by normal i mean its on the edge of acceptable normalcy because thats a LOT OF EXTRA CMS or more likely, there is another factor yet undiscoverd at play. If this were mine, a couple hundred either way would not raise an eyebrow. but since a single path is 850 ad we are talking 493 and 544 cms higher, more than 1/2 of a single path and assuming Identical twins or twins scenario not a factor, again, possible but this is high enough id be HIGHLY suspicious of other Paths. (more likely more than one that add to that extra approx 500cms each. There are ways to find out where the paths are if there are indeed more or to confirm their existence. If it is a natural (while unusual) distribution, then your aunt (and your parent) will share more DNA (significantly more, very noticably more) CMS with one side of the cousins than the other side of the cousins. so if you have first cousins on both sides who we would tend to expect around the 850 mark, one side should show over 1000 and the other side would show closer to 600 to your mother or aunt. If this is apparent then its a natural split and you can use this information in researching lines! (and it will help when their numbers dont fit the norms (above or below). Now if one side is normal and theother side slightly elevated the extra path will be on the NORMAL side somewhere. If both sides are normal for cousins then the extra paths will be on the other parents side somewhere, very likely multiple paths up as many as 4-7 generation as a single inherited path of that extra amount (you/3400/1700/850/425) would go up to the 5th node and then the same concept of normalization apply. lets say the GGP on other side intersected, the same line then that would pass 450. OR up another would each pass 225 up another 4 passing 106. Since its highly unlikely your going to find a Dual or quad direct line, then multi paths tend to be much higher up and at varying spots across rather than direct line intersections. While this sounds "UNLIKELY" on the surface,l in reality its HIGHLY LIKELY! so much so that we have all heard the term on the concept "en.wikipedia.org/wiki/Six_degrees_of_separation" Six degrees of separation. This concept is that ALL PEOPLE, every friend you meet in the street is an average of 6 "nodes" from being friends from each other. That concept of "knowing" also applies to kinships (albeit NOT at 6 and no studies have conclusively proven this in kinships) the idea and principle of "knowing" does in fact apply to kinships in that if EVERY person, EVERY PERSON, is 6 handshakes away from each other, then its is probably that at this point, two random people who marry would be 6 handshakes away. and since communities in past were small, this DOES often translate into kinship! in fact the LDS has "genealogy" nights were they often have members do this as a sample to see how close married couples are related. my friend told me he was the closest with his wife at 7 generations, most were 8 and 9 generations (they had no idea). My father #1's parents were 7th cousins and never knew it (and thuse would fit the scenario your sharing now.) I have helped MANY people who have people in their trees that didnt know they were related (one as close as 2nd cousin!) Other tools simply show that your kinships "FIT" this goes a bit beyond that because i did this with the underlying principle that i wanted to find BIOLOGICAL parentage thus every clue like this is huge. sorry for the long response, hope it makes sense. id guess you have paths from other side or that the parents shared in common on that side may indeed be 7th cousins like my grandparents in family #1 was or like the friend who shared his kinships with me last week from their church exercise :) Both scenarios (which again 6 degrees of separation in mind) are both likely scenarios. But there is definately more probability of additional pathing than this being two 80/20 and 80/20 splits from one grandparent via parent to both your parent and your aunt and also from you to your Child similarly (again its possible, but thats not as likely as additional path possibility)
@@DNAFamilyTrees Thank you so much for such a detailed reply! While I don't completely understand ( yet lol ), I get the gist :) I can certainly understand why there could be multiple pathways further back. My sister and I are primarily Irish descent and both Ancestry and 23andme list counties Kerry & Cork as primary places where we share DNA. I'd guess that both my parent's ancestors come from those counties that are right next to each other. I still don't understand why both my sister and I get the correct relationship with my aunt via dual paths, yet her daughter doesn't. I'm going to save your reply for future reference though! Once again, thank you very much :)
I updated the tool to allow ENTRY of age rather than the slider, it was hot patched, just refresh the page and you should pick up the change, apologies to anyone who had it open at the time the hot fix went in.
I still have the slider.
@@Richard-zm6pt on the page, hold down shift key and hit F5 or click the page reload with shift key held down, either one should refresh the page for you and get the new page and overwrite what is stored locally in cache. if that doesnt work, let me know. That would mean your ISP is Caching the information and ill put code in the webpage to FORCE it to refresh
Maybe a future update (not nit-picking but thinking for the future) have the person put in the birth year of the tested person. Though there may have to be a whole 'nother thought when the possibility of both matches being deceased and over the probable age of someone who could have tested.
@@staceycoates1418 reason i didnt go with birth year was the fact it was more "extra" calculation as you would then convert year to age to get in same place (gens up and down based on age. plus 2 numbers versus 4 (has to be 4 due to century) and every key entered is a place of potential error aka 50% more
@@DNAFamilyTrees Hi Larry. My page still shows the age bar, yet I know that I did see the new version come up where you put the age, while I was in the site. If there is anything you can do to force the link on your site to refresh the page, that would be great. Thanks Teresa
So excited to try this out, but I will wait till I have finished watching the video..Your hints and pointers are always very helpful..Thank you!
Thank you VERY much, Larry. You designed an excellent interactive cM chart. It has already helped me. Well done you!!
Glad it helped!
Yay you are back. I use your tools all the time. Can't wait to try this since I have all the problemsn
Excellent Larry!
wow wow wow great tool,,, it should be labeled as a MUST HAVE,,, thnks for a great tool thats easy to understand.
thank you guss, i made it from my own need to keep it simple lol glad it helps!
Thank you! I have been using the one I printed and laminated. I can’t wait to try this one! You rock!
Awesome! Thank you!
Welcome back man… hope all is well
Thank you Rich!
RE: maternal vs. paternal DNA... I was always thinking of a road with houses on both sides of the street. I LOVE your zipper analogy. I'm using that one. It's neater, and the visual works better too.
Thank you H.A.K.!
Enjoyed the video. Thanks so much. Now to begin plugging cM's in. I can't wait to see what I find.
Have fun!
i cant wait to Hear what u found :)
I found the chart first, then this video both tonight. Still checking it out. I am color blind and the colors are fine for me
Thank you for the feeback on it Lee!
Fantastic - cannot wait to use here in old England :)
Love this Larry. Something I would like to see is a version that could be used offline. Sometimes a person could be in a situation where they don't have access to internet but would still like to work on their genealogy mysteries. Also, age slider doesn't work for me.
Cooper mentioned that too, it does get "stuck" especially on tablets and touch screens. try refreshing the page but i will prioritize the change to a box for inputting the age!
Ill have to go through the considerations on an offline version but i can definitely see the value in that!
Thanks for this tool!!
Oh Goodness, I am so excited for this tool!!!!! I am hoping it breaks through our wall at the 3G Grandparent level for my dad!
This looks awesome Larry, great job, well done! Is there a link to start using this new tool please?
In the description, may not be a obvious on mobile verions
Thanks, Larry! Still working to understand all of different passes and such. Would appreciate the probability chart to be bigger/easier to read. I had to increase the zoom and then can't see the chart. I'm 75 and my father was adopted. My highest match is 183 cM, so it is a struggle. Your videos help a lot!
183, whew that is one of the tough numbers, is that a low 212 or a multipath 106, that's hard to say.
being that your in 70s and average genealogist mid to upper 50s i think we can assume they are a generation down from you. (assumption they may be your age looking too!)
Since Dual most likely that leaves 2c1r s the most likely kinship.
Now to estimate we have to "guess" are they a low 212 or a multi 106. Since the dual is not multi at 2c1r, my first guess would then be a low 212 (i normally would guess other way but one gen down there is only dual. if however they are in same generation or TWO generations down then multpath 106 is most likely!) the method would be the same either way but we will run with statistical probability.
if dual then we double their cms of low 212 to be a 425 for their parent. this would mean the parent on same line as you. since not dual that means their parent likely to be a single path first cousin to you. Thus we infer that your grandparent, aka your adopted father's father.
Now lets look at this from your fathers perspective. we double to move the ball up from you to him to 850 and again to 1700 to move up from the parent of the DNA match to their Grandparent. at 1700 and due to the age mapping which we assume consistent for generations (might vary, but we have to have our best guess on where to focus) this would put your father and their grandparent as half siblings. Find the 4 grandparents of that person and ONE of them is the most likely MRCA. since more likely a male, you can now whittle that to two. With DNA clustering you may be able to tell which side of theirs you have matches to and thus find which is the sibling and therefore the parents.
That is where id look first. (and my first run through using the patterns. dividing down gives us kinship probability, BUT we want in all cases to really know the MRCA, so the kinship is just a step to that. SO, MULTIPLYing backwards takes us up each line, but NARROWS field and lesses margin for error tremendously thus honing in. Now we assumed a low 225, if this was a multipath, aka high, 106, then we do the same thing but it would require one more doubling to identify the place the MRCA would be.
Hope this helps! Let me know if you get it with this!
@@DNAFamilyTrees Thanks so much for this detailed analysis. From his profile he is definitely a generation below while his father is also 12 years younger than I am. 2C1R is what I have been thinking. I believe the MRCA had an NPE put up for adoption also based on other matches, ages and circumstances. The plot thickens. Again, I love your videos, but this concept has been a little difficult to wrap my head around. I'll keep at it and will let you know if I find true answers. And I still have to find another paternal grandparent! Keeps me busy.
@@susancdrexler could be two gens under based on what you just said it’s one for sure maybe two
The probability chart (top right corner) needs to move left and up to give you room to make it much bigger. Also, I have known 2c1r from 95cM down to 35cM. Only the 95cM shows up as the correct circle of relationship. Love that you made an interactive tool!
Your example is perfect! the 95 fit spot on with 2c1r and while the 35 didnt, look at the CHART. for 35 the very first column of the chart is ... 2c1r :)
if that is with ancestry, what is the "Unweighted" cms for that 35cm match? (just curious)
They apply a WEIGHT to numbers under 90 that "shrink" so the 95 number is an UNWEIGHTED result while the 35 is a WEIGHTED result and in fact is NOT an apples to apples within their own list! (most dont realize this!)
Thats why i included the chart as a way to cross reference!
This is the best example of the day so far! Thank you!
@@DNAFamilyTrees 39cM went up to 50cM. 35cM went up to 39cM. This one still falls outside the predicted area. It is in the column before the blue balls. I totally forgot about TIMBER.
@@quiltingtangent459 What your saying is really leading edge on this, i had no idea people would be catching on so fast!
OK, for those kinships that are KNOWN to be BEFORE the blue balls, this actually indicates a MULTI path!
I have updated the tool display now, BUT i havent yet put the red in FRONT, still getting people on board with the "multi path" concept itself. I think people are getting it, so may have to bump that video up in plans :)
@@DNAFamilyTrees Time to explain "multi path" before blue balls! Need examples, now I'm lost again. You are telling me these cousins somehow are related to me in multiple ways. Need next video. I see that you made the other chart clickable. If you make the numbers a font size or 2 larger it might be readable on the page.
@@quiltingtangent459 still working on making it readable without click but I think the ONLY way will be to hand modify EVERY one of them with a different font/size rather than it auto generated. That will have to come later if i am going ot stay on schedule for full release this year of all parts including MRCA predictor.
I have an interesting match that doesn't seem to be following the ball and pattern method. She is a match to my mother and 3 of her other 4 siblings we tested, all in the 70-80 year old range. The match is 79. I have done a paper trail back to their MRCA, which would make her a 3rd cousin to my mother and their siblings (all full siblings). Out of all 5 siblings tested, this match only matches my mother at 9 cM, another 13, another 23, another 47 and the last not at all. Any thoughts on the low cMs? I cannot see in the paper trail how she could only be a half cousin or any other explanations.
Bonnie, this is a great question i hope people read it and the answer.
DNA tends to be "normalized" which means that the CMs will tend to be close to the "step" or "node" in the pattern. That said, beyond 2C there ARE variances. out of 200 or so matches that are 2C1R or 3C i have TWO that are like this. mother matches at 278 i match ZERO. Mother matches 273 with another, i match ZERO. another she matches at 271 i match at 268!
at the 2C1R and beyond (3C in your case) it CAN be a crazy low number or even 0!
BUT, i have 2 of 200, thats 99:1 ratio or 1% of the time. (or 99% its normal).
My initial response would be, if it was you only, that this was one of those 1%ers. HOWEVER, The fact FIVE tested with this anomaly tells me its not an anomaly but instead its probably indicative of something else, aka, the paper doesnt match the DNA.
Example: i tested, mother tested. a Neice to my mother tested and i matched her at 9cms (my 1st cousin) and her 3 siblings had 0 cms match (well under 6 ancestry reported at that time). Ironically mother matched ZERO. i actually matched through my fathers side! This indicated Mother had a DNA surprise that i had to work through, which i did.
if it was one or two i would say this was in that rare variance, BUT with FIVE? if its 99% unlikely, and it is occuring with all five, id want to scrap the paper trail and try to do that side via DNA triangulation. sounds to me like there is a surprise in there so tread lightly! its a soul punch and sometimes its easier to "accept" it as an anomaly (but in this case i doubt it is, i think there is more than 50% chance there is a surprise in that path.)
good luck!
The probability chart (top right corner) needs a clearer or larger font. I'm viewing this on my macbook with a 13" screen and I had to increase the page size - which then cuts off some of the tool - to see 1c1r as an option but it looks like 1cr to me (even wearing my glasses for my stigmatism). I think the relationship choices should be a larger and maybe bolder text
100% agree! there definitely needs to be a change/improvement in that display!
I love this tool! I understand it better than that other one Blaine Bettinger was using, sorry Blaine! Lol. I was really looking for something like this to help explain the different family relationships such as 4xGreat-Aunt, etc... My question is: I am looking for my late mom's biological parents. She was adopted, but I found her family via my DNA matches since she never took a test. Can this tool be used for her line? I noticed in the video you only do the paternal line? My biggest top match is a woman who shares 1,094 cM! Ancestry says she would be my 1st-2nd cousin (I really wish Ancestry would separate out those 1-2 cousin lines) I see this tool also shows that too at 47%. So, that would mean we would look at one of her aunt/uncles? Still not quite understanding that part. Could you explain more? Thank you.
I love this tool, it kind of confirms that a couple of apx 2nd cousins are indeed multi path. Ahhh more research! LOL. Suggestion would be to make the bar graph larger. My eyes are old. But seriously, that is such a small thing. What you have already is amazing.
yeah i HAVE to do something withthe chart, we got the AGE entry improved, on to rest of the TO DO list from it :) lol
Still no comments on color blind, either no genealogist are, or its not a problem or no one wants to mention they are, not sure, but i KNOW those greens are not the best :)
How does one access this tool? Where do I get it? Larry, you’re the best! 💫‼️💯
@@familygeno3970 link In video description
Great tool! I have three known half 1C1R's at 325, 311, and 309 cM. Not sure the tool represents the relationships for these values correctly. I was expecting the 1C1R circle to appear in blue, not yellow, to validate the "known." Unless the tool is telling me there is additional path(s) for them? Like the yellow bumps to red for multi-path, does the blue bump to yellow for relationships that start at only half? Seriously, awesome tool.
Thank you for your examples! I just pulled up the dataset for 300-325 it is 0 as is as are the other blues, i put the slice for the 300s in wrong spot! the Chart is right but the Balls in the pattern were WRONG! I have corrected that (Beta, probably will be a few more i messed up before it gets flushed out!)
now to your scenario:
A) I have updated the tool to properly reflect the pattern in relation to the data for that CMS
B) Lets cross check with dnapainter.com/tools/sharedcmv4 .
WE enter 325 and find that H1C1R histogram shows 83/56 (blue) to the left and right of our input number (139) and 374/1015 (1389) to the left and right of our number for a 1C1R (yellow)
Now this is where my tool looks at things from 3 perspectives at once. the pattern shows Closer to 425 than to 212 (100 vs. 112) BUT it (now) shows it to be a 212 with additional pathing)
The CHART confirms the error you mentioned as 1C wasn't even in the statistics for probability! (guess that was a low point in caffeinated drinks - Great catch!)
What we have now in the BP tool is 2c, 1c1r, 1c2r, H1c1r, H1C, 2c1R in that order of likelihood.
The SharedCM tool shows for: 1c1r (1389), 2C (1147), 1c2r (335), H1C (243), H1C1r (139), 2C1r (4).
So the difference is 2c, 1c1r, 1c2r in bp tool and the same 3 in the sharedcms tool but first two flipped in order of 1c1r, 2c, 1c2r.
difference being the inclusion of estimated multipathing in the BP Tool and not specifically addressed in the other tool.
---
I fixed the error you pointed out and got it into the live production version immediately! (will be out of pocket Tuesday and Wednesday but Thursday i will review the break points and see if i messed up any other Chart to patterns (since it is manual not dynamic on data yet - only so many hours lol the sharedcm tool had Blaine, Johnny and Leah, this one has me. :) Thats why this is Beta though to flush out any mistakes i made! Thanks for catching that one!
----
Lastly, we are wanting to estimate MRCA location. the 2c and 2c1r (two of 3 top in both tools) would intersect withthe DNA matches Great Grandparents, the 1c1r would be just grandparents, so if you build their tree to their 8 great grandparents, you should intersect! if 1c1r, youll have that intersection already (so as you build tree, obviously id look at the 1c1r first on the way up to the GGPs. The Age probability will help you identify which 1c1r, 2c1r it may be. and if you know their age (their parents BDs vs. your parents BDs is a good clue for estimating, but looks as if its most likely (based on default age of 50) to be at YOUR GGP or 2GGP. AND there is likely another, perhaps on other parents side of 106 to be found at 2ggp, 3ggp OR perhaps 2 paths equaling this higher up. At 2ggp and beyond, ANCESTRY projects it to be multi and higher up, which is why they Timber the results at 90cm (3c which the mrca is 2ggp). Their paper on timbering discusses they believe at this point it is more likely to have a multi path than single path for the CMS given, so they "shrink" or "WEIGHT" numbers under 90. (which can effect 106s if there are 2 making it up)
---
I am very happy you caught this! Even happier to correct it. Also Embarrassed i let it slip past me even into the Public Beta. (needed MORE COFFEE LOL)
But Semper Porro (always forward) its fixed and its getting another step closer to solid as we flush things out in Beta!
GREAT CATCH!
Thank you for all your hard work. I have learned so much for you and this video may have solved a family mystery. I was always told not to include endogamous lines in my searches since their numbers are too high. I think my DNA match, whose mother was abandoned as an infant in 1912, is from my double cousin line that I've excluded in searches. My brother matches her 230cM. I'm only 110cM. Our known 2nd cousins are 103 and 167. How is my brothers so high? More than 212 when the rest of us are closer to 106... It could be the double cousin line, right?!? Or am I doing this wrong?
At 2C1R dna can drop off significantly (or very little) from normal.
it COULD be additional paths for him BUT id guess the match is a male? This explanation goes into the genome sequencing itself and outside of what i can probably simple explain.
But do not worry about endogamy or pedigree collapse when searching! thats what the RED dots are for! i have 3 sisters married 3 brothers in my line, aka all 4 great grandparents on my fathers side are the same for me and for them so they get the "pathing" from each person passing dna along and thus appear twice as high as the same kinship (2C) from other paths including those from same GGPs but without the double (triple) marriage situation. They are 469-488-508 cm (3 matches) and they are all 2C in RED. others at 2C show as 208-233 in match list as would be expected for DUAL path only and one is a 110 (single path)
thats what special about this, its all about the paths, not the kinships.
@@DNAFamilyTrees The DNA match is female. Her mother was the abandoned infant. I no sooner made hand made charts for Single Path and Dual Path to put into plastic sleeves for dry erase and then saw your new video. LOL. I went back and added dates of my ancestors and knowing the abandoned infants birth month and year was and to rule out some lines. That really helped me the most. In this area I have 3 generations in a row that are 30 years apart. After adding the dates for everyone it really seemed to pull it together. Now I just need a couple volunteers from the 2-3 possible children it should be to prove it. If I'm even right. LOL
Hi, love this tool! Can you make the age stay fixed unless you purposefully reset it? My age isn't changing that fast and I have to reset it each time I change the cm # of what I am testing. If you're doing a lot of check, it makes a difference. Thanks!
oh so if you "RESET" leave it the same? hmm.. i guess i could do another reset button that leaves age next to the full reset. Great point, i thought about that in my testing but didnt, but since you brought it up, yes i WILL make that adjustment!
Thanks for the feedback!
OK, the Button you requested is now live, there is now a "Full Reset" as well as a "CMS Only Reset" option so that many can be run through at a time without selecting age and probability overlay each time.
Just downloaded the tool and I'm still seeing the slider even after updating. Great tool just thought you'd like to know.
I botched it up, sorry aobut that, somehow i messed up versioning and replaced the entry with slider. It is now corrected sorry about that, i messed it up.
HI Larry, Thanks for this wonderful tool! I do not fully understand it yet, but from what I think I understand I can see that it is going to be very useful to everyone working with DNA matches.
My question is, can you make the tool work backwords as well as forwards? In other words, right now you can set an age and a CMS and find out where to look for your most likely MRCA.
But I want to be able to put in my age and the particular MRCA ancestor on the ball chart who I am trying to identify and have it tell me what number of CMS I should look for in my matches list to best help me track down that ancestor.
For example, I want to put in my age (76) and enter that I am attempting to find my 3xGGP dual path, and have it show me what number of CMS to look for in my matches.
Then I also want to be able to put in my Aunt's age (96), and find out what number of CMS to look for in her list of matches to find the same person, but in her case he is her 2xGGP.
I hope that is clear.
Anyway, I love all of your youtubes and find them extremely helpful.
Thanks again,
Jane
Jane, That is What this one is supposed to do for you. When you enter your age adn the CMS of the match, the green overlay shows you were they are going to match you. By going up your ancestor line and down to that location that top spot will be the MRCA.
Now to the specific, the ball and pattern CAN do what you want, put in the age BUT instead of entering the CMS you would COUNT to that kinship to know the estimated CMS.
Thats what the blue, yellow and red balls do. for instance:
I want to know the estimated CMS of a 2nd cousin. you know a 2nd cousin MRCA will be 1xGGP.
So count up, parent 3400>>Grandparent 1700>>Great GrandParent 850>>(now down)Great Aunt/uncle 425>>1C1R 212>>2nd cousin 106!
Now thats a single path (one MRCA) 2nd cousin, if a FULL kinship, aka dual path (yellow) its that same path through the OTHER 1xGGP (one 1xgreat grandmother, one 1x Great grandfather ) for 212!
So the estimated CMS for a 2nd cousin is 106 (half) and 212) full.
Ever time you traverse an arrow up or down you divide by 2 (start with you at 6800) this will give you the estimate.
You can confirm you did that right if you enter 106 and also enter 212, youll see the 2nd cousin be blue and then yellow
This is so cool! I can’t wait to play some more at length but I did a quick test. I threw in my daughters’ and mother’s shared cMs of 2047, 1637, and 1553 respectively. The most likely relationship for all aunt/niece. What to make of that?
email me at dnafamilytrees@gmail.com
What was the result? Interesting for sure.
Brilliant!!!!!
The only DNA line I'm sure of is my mothers line. There is some endogamy. Using leeds method I've worked out four lines. Cant get my siblings to test though even though I know we have different parent combinations. They just dont want to know. I will keep on looking. Hopefully this method helps me.
i think it will, the matches will stand out and the cms is a great wayto determine. hang in there! i had NO one help me with mine (until i figured it out, then dozens tested to either confirm/deny my assertion. I went from 1 match over 100 to dozens lol
The age doesn’t change with the slide.
yeah i had mine do that to me (even in video) darn slider, need to make that a box! if you refresh the page, it will work, something with the browser handlers does that every so often. definitely needs to be altered to be an age input box :)
Larry I am at a brick wall. I have DNA matches of 1st and 2nd cousins whom I do not know. I have made contact with them and still no trace of how we are related. The highest CM match is 636. Our families are from the same area but cannot find the connection. Any ideas of what to do next.
Very impressed; I do like it a lot. Suggestions for legibility, apart from what you have already mentioned: I'm not colour blind, but find black on darkish red lacks contrast. May I suggest a paler red?
Thank you Bob!
Thank you for developing such a useful tool. I am fairly new to genealogy. I am looking for my great grandparents on my father's paternal side. I have a name for my grandfather, but it doesn't match anything in my matches. My hypothesis is that he either changed his name or was adopted. Using my DNA matches, I have come up with families for my great grandparents, but they were not married. I have narrowed it down to families, but I haven't been able to determine which person in the family is my great grandparent. I don't have any 1st cousins on my father's paternal side. The closest cM are 160 and 105 on the paternal line and 131 and 120 on the maternal line. Do I only look at the single path for all of my matches?
Mary, as i mentioned ot steve a trick i often use is to go the "other direction" so instead of dividing using the pattern i go the other way and multiply back.
So we are looking for Grandparents in relationship to the DNA matches. If we take the 160 and 105 my initial assumption would be the 160 is a multipath 106 and the 105 is a single or dual 106. So id enter their numbers and use age overlay but .. this is where we get fun... This match is to YOU, right? so, we can get one step closer by looking at them to your Father. (if this is your fathers dna match then its his father and just adjust accordingly what im saying).
this means we change the number from 106 to 225, this is the kinship to your father to that DNA match. Now lets double again, to get to the DNA matches parent (we dont know or care at this point which parent). this is your parent to their parent. Now we are going to do it again, once for each side, double then double again to get your grandparent to their grandparent.
this gives you a VERY small area in which to focus to make a connection. If you know birth years of your grandparent and theirs (they have 4) then you can estimate if they are older, younger or about same generation (if you dont know you might estimate you to the dna match based on profile picture or informaiton that they shared). if same generation then the grandparents of each are likey sibling
This would incline me to lok at your 2x ggp as the intersection.
now back to the original 160 cms, we guessed from the doubling exercise that the mrca is likely at 2xggp. This fits the 160 as a multipath 3rd cousin, aka a 106 with multiple paths making it a 160. WE assumed multipath since over the 106 significantly, so all things lead me to think the mrc will be their 2xggp and your 2xggp at MRCA with other paths also present.
I would look at 2xggp really hard :) build out those matches trees (yep, we gotta do the work for them) bulid those trees to their 2x or 3xggp.
Then xref this with your DNA clustering to see which group of 2xGGPs it is for you.
one more trick i do in this case is pick the dna match at 160, do shared matches, FILTER ON PUBLIC TREES, then right click on every tree in the list and open in new tab. (i do about 30 at a time) then i look at the trees for each going tab to tab to tab to see repeating surnames or identifiable duplications. in most cases there will be one or more overlaps, then focus there as those to overlapped each other and you intersected to both of them so its very likely you intersect on that duplicated kinship (marriage or remarriage name location).
hope this helps.
@@DNAFamilyTrees Thank you so much for your response. I will have to take your recommendations step by step. The 2GG is my hypothesis as well. Since it isn't my GG and I have matches to 3 of the brothers who are less than the 160, am I correct that It should be the fourth brother who died at the age of 39, wasn't married and as far as I can tell, didn't have any other children at least none that I can find DNA for?
@@mjrose-r9y we cannot infer he didnt have children as there are people who dont take DNA tests for personal reasons, just as today people choosing to or not to get a shot. it simply means that IF THERE ARE kids, they havent tested. No kids is a real likely possibility but the data much to limited for that inference. it was SEVEN years after DNA testing came out that my relatives that provided the fact my tree was not biological after all and they were two years behind me so i went two years thinking everything was as I assumed.
then one day, pop, like popcorn kernels this family line started popping into my match list until i knew with certainty that i had a DNA surprise. So i would say it just means no testers from that finite set. (we could infer some socio and economic things inside the USA or privacy things if they lived outside the USA.
As for looking single, well, in finding a kinship we know ALL PEOPLE that DNA match have at least ONE path. More often Dual, and often multiple especially dependent on locations.
So I would say initially look for that ONE path that connects then work back to see if more (youll know right away when you connect the one).
those numbers on a single path would likely connect at 2c, your GGP and their GGP. This would be my first focus, however, if DUAL, if they are older then STILL their GGP, but if younger than you then still YOUR GGP but their 2GGP. In both Dual scenarios either YOUR GGP or THEIR GGP was the intersection so if you work out trees to get both of your lines GGPS, we can then infer that 1 or 2 of those GGPs will be the point of connectivity.
on CMS i tend to focus on finding a SINGLE path first because to have dual or multi there must be a single so single is really the main focus, and i treat the dual as an additional single BECAUSE it may not (usually does, but may not) come from the spouse of the single! That second path can come from any direction, even from other parents of one or both of you.
So yes, id focus on single to see if you can locate dual if present. (fyi, the 160 is MOST likely a 106 but dual or multi since at least 1 path is part of that 106, could be 2, but we can infer that the 160 has an additional 53 path from somewhere that makes it a dual or multi, we dont yet know.
@@DNAFamilyTrees Thank you so much for all of your help. I really appreciate all of the information that you share.
Ive tried the link above a few times and it doesnt seem to be working. Any help would be appreciate
christie which link is not working? so i can fix it
I have a 2C1R who shares 567 cM with me. My great grandparents are the siblings to her grandparents so it doesn't even give me the option for that. Otherwise I like it. I like that it takes age into account.
If this is indeed a 2C1R and not a surprise story to be found, then there is a SERIOUS amount of pedigree collapse!
For that to happen at that CM you would have to have all four great grandparents the same (sisters marrying brothers) AND have a close match on the other side of your tree.
if this is not the case, then there is more to that story.
Walk softly.
@@DNAFamilyTrees I know my lines cross I'm just not sure where yet. Her first cousin I share 327 with which I know can be in the normal range of a 1C1R. It was the 567 that threw me off for the longest time.
Still trying to understand all this, but I'm excited to learn. Suggestion: Can you put the numbers of cM's each generation and path divide or multiply up to inside the balls for quick reference?
maybe the ability to toggle kinships/numbers, hmm... interesting idea, i like it!
Yes I was thinking the same thing. I don't have those numbers memorized like Larry does.
LARRY PLEASE HELP!!! I am matching someone at 209, my full sister is matching the same person at 130, and my other full sister is matching the same person at 64cm........ What would you make of this???
sent u reply email
💖💖💖
Speaking of Ancestry... I see you've connected a bunch of your matches in your tree. I made a chart as I did mine when that came out and the actual relationship is often WAY DIFFERENT from the most likely relationship according to the Shared Centimorgan Project. 🤔
Oh wow! email me those matches Cheri, id love to take a look (dnafamilytrees@gmail.com) I just responded to NYCSue, the answer to her i think would also likely fit your findings in similar ways.
Very helpful. I have 4 DNA matches on my paternal that are about 225 cM. They all show as shared matches in Ancestry. They all appear to be a generation below me. Only one has loaded to MyHeritage. Adoptions are present in 2 of them. It appears my MRCA is my paternal great grandfather but I am lost from there down. Side note: My 4th GGP's on my paternal side are also my 5th GGP's on my maternal side.
Addition request: approximate age of target entry field
just based on this information, i would estimate this is a 212 with a bit extra from the other side as you mentioned.
One trick i plan to do a video about is REVERSING the concept, divide by 2 going one way and instead DOUBLING the other way.
Taking the clue they are younger, i would DOUBLE that 212 to 425 to see their parent to you. If they are one generation down then this would insinuate their parent being equal line. that would be 425 as single path cousin or multipath 2nd cousin.
NOW we do that again, double that number again TWICE, once for them, once for you. so that is 850, then 1700.
This would be your parent and their GRANDPARENT on equal lines. That insinuates a single path sibling.
That then insinuates your grandparent is their grandparents parent in a single path scenario.
NOW if indeed your Paternal great grandfather is the connection then the math really goes a bit crazy. Back at the 212 then they would be the #1 probability in the chart at 2c1r. Their Parent thus your 2nd cousin dual path. your parent to their grandparent would be cousins. your grandparent to their great grandparent would be siblings and thus connect at 1ggp as you mentioned.
Which scenario is most likely? well if normalization is tending towards mid then for that 225 to be a 212, thats 13cms from a single additional path. That can be indicative of multipath 2c1r. BUT thats a pretty small variance at 13 and since you know of another path via the other parent that probably accommodates 13cms then id be inclined to think that for the MRCA path you seek its NOT multi path. So they are likley two gens down at 1c2r or one gen at 1c1r. The above scenario (1st one) demonstrates the single path option.
I said i would go Yellow, Blue, then red as habit. There is NOT a yellow in the generation below, you have to go two down to get dual. So then the question becomes which is most likely single path or multi FOR ONLY THE ONE PARENT in question. since you know a legitimate path on other parent, then its unlikley its multi on this side. so again, id look first at single path 1c1r and intersecting at your Grandparent.
Other factors in play, (identical twins, etc) that make the number act differently, but identical twins are 0.003 likely so 99.997 that its not one of those :)
IF they are multipath Then this also means... its a 106 multipathed to 225 which is highly unlikely. remember blue is single, yellow is dual, red basically Quad. So if there is a MRCA with multi then its actualy more likely to be up at 2xggp statistically. BUT id look at GP first, Their GGP, cross reference with cluster information should identify for you which of the intersections it is.
hope this helps more than confuses :)
I have a great-grandfather who was an identical twin. cM's are out of whack for my matches who are descendants of his twin. Just another scenario to consider when looking for another path for those matches who don't fit the expected cM range.
yes identical twins have have aunts show up as parents pretty often. But if you find that on a line you can simply REMOVE the division component to that descendancy line 1x and get a pretty good estimation for cms.
My mothers adopting mother was an identical twin (she used to tell me stories how one was good in math the other loved english so while in university one took all the math classes (2x) and the other did the english (2x) lol
@@DNAFamilyTrees Funny story!
Yep, my 3rd cousins look like 2C"s and Ancestry won't let me designate them properly.
Gotta love it when the "experts" "know" more than you do. one thing i have learned is there are NO ABSOLUTES regardless of what paper trails say, regardless of what the DNA says, there is ALWAYS that exceptional condition!
I have talked with Subscribers who one for example has had 6 children, obviously a women but FTDNA will NOT ALLOW her to designate herself(her own DNA) as such! She is one of the rare people whose Chromosomes actually changed from XX to XY! so FTDNA will only allow HER to upload as a MALE! (and she is not the only person whom i have spoken to with this change over time!)
now i say rare, truth is the actually percentage it occurs with is UNKNOWN because over history we havent tested DNA like we have in recent years! And she doesnt feel or act like a male in any way shape of form so unless FTDNA told her and she investigated she like many others would NEVER have known!
The chart in the upper left corner some matches show red and blue- where some just show blue.
thanks for pointing that out, was two data series combined into that chart. (i need to do follow up video on this specifically) there is a point in the ancestry dna match list that they change how numbers are shown right in mid list! Most if any people dont realize this! i do need to raise awareness on that.
Thank you Larry. This tool is great. I just want to ask about one test I ran.
I have a 3rd cousin descended from a pair of 2nd great grandparents. I share 44 cM with him. This yields 3rd cousin, single path, or 4th cousin, multiple paths. My dad, who is his 2C1R, shares 237 cM with him. Results yield 2C1R multi-path. My sister matches him with 69 cM, also his 3rd cousin, and results yield 3C dual path or 3C1R multi-path (the highest probability). So, I'm thinking it looks like a multi-path relationship, but I haven't found any other shared ancestors through my genealogical research. THOUGH, he has an ancestry that involves a lot of intermarriage in a single community.
Richard, This is an excellent question, one i hope people read. If you have found your fathers MRCAs at 2ggp (dual) then your solid. This was created to narrow the place to look so that someone can get to an MRCA in shortest time and then build down to find Biological parents.
At 2c1r DNA gets TRICKY! normalization tending towards mean gives us great estimation on where ot look BUT There be dragons so to speak at this point. It can vary WILDLY from 0 to Multipathing large numbers depending on each unique scenario.
My mother matches someone as a 2C at 278 cms, i match them at under 6cms (they are not on my match list) another she matches at 268 cms i match at 258 cms. (now my 258 is obviously from another path via my father, but on the flip side not showing up is indicative that it was a non normal split all the way down that line, it happens, not often as other cousins on that same path match in a normal range. He may have a yet unknown DNA surprise in his line that he has yet to find out and correct (like i did with family #1 before i had my dna surprise and changed my tree accordingly) or it could be a strange split at every level (statistically its most likely a DNA surprise).
if you put in 44 cms on the shared cms tool and then put 0cms in the same tool the answer is almost identical for what the possible kinships are!
This tool is in conjunction with as it has an AGE band for probability, the ball and pattern that is highlighted based on normalized single, dual and multi (which statisitically is more accurate than not) and also chart that shows data from multiple sources rolled into one that at various cms ranges contradicts the normalized result.
That, like the story i shared on my family above, can be from people not yet finding their true DNA family and believing their tree accurate or evidence of a bizarre split at each level which is possible (and shows frequently in the database, BUT i suspect its more evidence of unknown DNA surprises more than wild variants. as AtDNA has very few mutations even at 4 and 5 generations! AKA should not vary from normalization much. So for it to vary at every generation while statistically possible, I believe it FAR less likely than reported and the reports are people, who like me at one point, didnt know there was a DNA surprise.
There have been people in comments question some results, but i can see that they are about to uncover a surprise and i let them do that at their own pace as i wish that on no one!
IMPORTANT:
If your dads match was a 237, and a known 2c1r to him, its very likely a multi path. (well he is, lets just go ahead with that) and if there is not duel marriage in the line or a remarriage or endogamy on that line, then its likely through your fathers other parent. his 237 is a 212 level. meaning 106 from each of those two and 35 from another (with slight variance of course) This would lend to 3ggp on other parents side maybe 4ggp and/or 5ggp on that side that in combination added to the 35ish range.
It is very possible then for that line to NOT show to u being so far back and youd expect to be half of the remaining 212ish or 106ish.
Here is where it gets tricky. TIMBERING. if say your DNA was just under 90, still close to 106cms that you would expect, Ancestry would TIMBER that to a much smaller number, perhaps as low as 44. To see the PRE-TIMBER cms score that is in line with your fathers you will need to click on the DNA match CMS in the dna match list, there you will find an UNWEIGHTED CMS number. I am guessing its 60-70ish cms.
If so then while that drop off from your father to you seems giant its simply drop off on the one line and slightly lower than a perfect split on the known line.
Let me know what the Unweighted CMS was for that match, i am very curious!
@@DNAFamilyTrees Hehe, Larry. You've given me a lot more work! I should have mentioned that these data are from FTDNA. They come from a study I did a few years ago to determine whether a family in Illinois was kin to my paternal maternal great grandmother. I started with a Y-DNA study of descendants of two known brothers in Illinois and descendants of my grandmother's two brothers. Sure enough, they matched. I then had a descendant of the Illinois branch of the family tested for autosomal DNA, and this fourth cousin also matched--we got 18 cM.
From my great grandmother, we descend four generations to my daughter and match a descendant of my great grandmother's elder brother. I was using your new tool to analyze the relationships.
The common ancestors are Thomas Liston and Letitia Sinnett of NB, Canada.
They had three children. Each has descendants who have had DNA tests.
The cousin in question is descended from their son James. My family is descended from their daughter Mary Elizabeth. The cousin is four generations from our common ancestors, and this would be a dual path (from Thomas and Letitia).
Following is my family line anonymized with the cMs shared with the cousin and the documented relationships (on paper).
Mary Liston Case gr gr aunt dual
Hazel V Case 1C2R dual
my dad 237 2C1R dual
me 44 3C dual
my dau 60 (!) 3C1R dual
I don't think FTDNA timbers results, do they?
My e-mail is triovlaif@gmail.com, I'd love to be able to show you visually what I am working with.
Thank you so much for your lengthy reply.
My mom has 4 half siblings. No full siblings. 2 half siblings through her mom and two through her father. I have a match from my mom’s side of family that is showing 465 cM. So my 1st question…all the children of the 4 halfs will be my half 1st cousins? All relations of those people will be half-whatevers? So if my grandfather has only full siblings, their descendent will be Full relations with me? Or are they halfs too since my mom’s 2 siblings thru him are halfs…everyone is half? I tend to think that 1/2 relation is only of those people and the kids of grampa’s siblings are full. Help? This makes my brain hurt.
This one is confusing for alot of people. Descendants of your mother or Father will share ONE PATH to you, DESCENDANTS. Your mothers parents however, if not remarried the descendants of them would most likely be DUAL path through both great grandparents if that is the intersection, same on up the ancestor tree.
How can we get this? I’ve been trying to find a bio father for a cousin for over a year.
www.dnafamilytrees.com/bp/indexV8.html
So Larry I always have to be difficult but here is a problem that I have puzzled over. I have a set of 3C. We all share the same second Great grandparents. There are three sisters from one descendant and one from another, they share great grandparents. I know them, have documents etc. One sister is 20 years older than her siblings so they are late 40ties she is already in her 70ties. Here is how they match me 13cM, 29cM and 42cM. Then the other cousin, who is their 2nd cousin, but my 3C, he matches me at 230cM. Yet we all share our second GREAT Grandparents, dual kinship, well documented. I know technically from DNA Painter that one can go from 0-234cM but I think it skrews with your ball and pattern tool. What do you think? Have you ever seen such a spread in cM for the same level of kinship?
yes it is possible to have that disparity. RARE, but possible. its like rolling snake eyes with dice 4 times in a row. it can happen, will happen sometimes but its rare. 1/36 x 1/36 x 1/36 x 1/36 for snake eyes 4 times in a row, or reduced 1 in 1,679,616. Not sure the actual odds of that disparity but after 2C1R (beyond 6 steps) aka 7 or more, it can range from 0 to nearly 250 for a 3C! But those are RARE occurrences but they DO happen, in fact i have that 2x in my tree with a 2C1R on my mothers side (dual path)!
My mother matches 278 as 2c, they do not show on my match list! another 273 does not show on my match list! and another shows 268 to mom and i show 263 to the same person (turns out i have 2 more paths to them via my bio father)!
At 7th step and beyond there can be WILD variances to the low side and even more rare, to the high side. That said, it is rare, so in this case, your matches at 12, 29, 42 cms matches would be all 3 on oddly low side unless single path (remeber its rare for one but then you roll the dice 3x and normality returns, or it should. While its possible to get snake eyes 3 in a row i darn sure wouldnt bet on it to occur LOL. your 230 is BROKEN high. like one of the highest recorded ever for 3c kinship. DNA painter shows only 9 data points at that range, while ball and pattern gives it 4% likelihood (much higher probability, BUT ONLY with multipathing!)
(in dice terms this is like rolling a 3 with two dice, very low odds, but without multpathing it would be like rolling a 5 with 5 dice - Happens in Yahtzee, but id put my money on any number higher than 5 lol).
its POSSIBLE sharon, but my gut tells me there is more to that to be found somewhere.
To have that number at a 3C lvl there is more than the 2 paths in play for you and that match. with the intermarriages you told me about previously, that would be enough for me to think there is a "reconnection" by lines somewhere, perhaps still undocumented.
@@DNAFamilyTrees Lar thought exactly that same thing so tested my Mom and her sister. No cross match. Can only assume the cross is in the paternal line but only way to see that would be to test a 2nd cousin who is on the paternal paternal line. I tell you the endless aggravation. What about the three sisters? I also have another 2x3C matchs to this whole line same set up and they are at 24cM and 60cM. So everywhere from 13-230cM for the same line. The eldest son of my 2nd GG. Really?
Hi Larry, Just so gratefull for your lessons. You are amazing! I am so close to making a breakthrough, but I am still stuck and have been researching for decades to find a dad. May I contact you by email and hire your assistance please?
emailme at dnafamilytrees@gmail.com i will review what we have and maybe we get lucky and we can post your answer right away. if not Ill see what steps i can recommend for you to have success
If I am understanding correctly if I have a cousin match at 626 CM's then their parent would match at 1252 CM's or around that number, now would they also be a cousin match or half sibling?
Yes and no. Probably more because the parent would be a step closer.
The only true 50/50 split is from you to your parents. All other relationships are based on what amount your parents gave you through recombination.
The halving beyond parents is an approximate assumption.
My sister is 1241 cMs to a 1st cousin
My brother is 924 cMs and I’m 948 cMs.
The average to a 1C when split half-half-dual path-half is 800-900 cMs and we are all above.
If his sister, my other cousin, did DNA, it’s possible our shared DNA is under 800 cMs.
Therefore if you have a cousin match at 626 cMs, it’s in the 1st cousin range so you should imagine that person as 850 cMs and their parent at around 1600-1700 cMs. If someone related to this person came in at 1252 cMs it could be the person’s sibling and you are mistakenly thinking it’s a parent.
Oh it can be a bit confusing.
I have a few 1C1R s that are low cMs and not falling into the B&P suggestions. this is why I suggested an advanced feature to include longest segment. It’s possible records and findings of full kinship are actually half kinships but until I find proof along that path, I just chalk up to low cMs.
626 shows as most likely a 1c1r (typical dual 1c being around 850) but yes, each time you go away from you, dived in half and if you go TOWARDS you, you would indeed double and that is a POWERFUL insight! because this is not only for THEIR parent, but you can do it again for YOUR parent and even better estimate kinship because the higher and closer it is, the less the margin for error.
so in your example, you have lets say an unknown 626. you then could estimate their parent (unknown which one but one of them) is approximately double that at 1252 to you (always bearing in mind that that 626 is likely a low 850 or a high 425 ill refer to this in a minute). BUT you can then do the same for YOUR parent (you dont know which one but not important yet. DOUBLE that 1252 to 2504. This is the estimated kinship distance between YOUR parent and Their Parent.
This gives you a VERY finite set of options with VERY clear likelihoods. (so in that example their parent and yours are most likely siblings if near the same age.
now lets look again at the 626, lets then say its a high 425 (example i have a single path 1c who is 441, her daughter is half that at 222 but her son to me was 387! turns out this is due to the mcfarlin path that my mother ALSO has that hte set he had matched me more than the daughter who was less on that mcfarlin path than her brother)
so if they were a high 425 (due to undiscovered additional pathing) then double that would be 850, double again for your parent to 1700.
Now we still get a very finite set of options! the parents could be an Aunt/uncle (reciprocal niece/nephew depending on which was older or a single path sibling (half sibling) )
So from what youve said we know that the person you match at 626, their parent is your parents aunt/uncle (or niece/nephew) or is a sibling or half sibling (most likely single path sibling since unknown in connection.
Additionally we could deduce that if their parent was your parents Aunt/uncle then one of the dna match's grandparents (their parents parent) is your parents Grandparent as well. (clustering should point to which side to look and look at males for missing lines first as its harder for a womans child to be born without notice than a males as often he may not even know.
if they the parents are siblings again the DNA matches Grandparents will be the intersection but would be your parents parent.
Now this is where that probability map helps based on ages of those involved, even though it may not eliminate aunt/uncle or niece nephew the green probability will tend up or down based on your age. that is the direction that recipricol kinship is most likely to be.
hope this helps! i OFTEN use doubling to estimate because that helps laser focus by changing who your looking at! so looking at their parent and your parent is a GREAT way to make that number more concrete to help you focus on were to look!
so while the 626 itself shows ALOT of options and the chart helps focus on most likely, go up one for their parent and one for your parent (double and double again) and use that number to connect your parents and that becomes a MUCH cleaner task, then take that knowledge go back to the 626 and see how that would fit.
as i said, i do this ALL THE TIME to help me decide the best place to start looking! (so if the parents are say aunt uncle to each other then your most likely 1C1R to them (which is what the estimator shows as longest bar in the chart far longer than other bars, followed by single path 1C, but since a single path 1c would mean parents a single path aunt/uncle and 1700 is a dual, we know that probably not right if it was a high 425, now if it was a low 850 on the other hand that goes 2 nodes closer! doubling their parent would make 1700 and doubling yours would be 3400. This would mean your parent or theirs is the parent/child of the other. if within 15yrs of age of each other that is incredibly unlikely.
So based on this, you pretty much have your answer of how they fit (and thats the beauty of this looking at it from path perspective :)
when you prove it out, let me/us know what you find!
i would treat them as a HIGH 425 and estimate from there which means they are highly likely to be 1c1r to you with your parents being A/U and N/N to each other. 626 (their parent being your dual path (full) 1C most likely. (but thats without age component which is why i added that, that could also be 1c1r and their parent be your Great A/U as there are two 1c1r locations in a tree.
Now the 1C1R locations in the tree are ABOVE and BELOW your generational line SO... if the match is older than you, their parent is probably your great A/U and if younger probably 1C.
Said in a more useful way, if older than you their Grandparents will be your MRCA (2 of their 4), if younger than you, YOUR grandparents (2 of 4) will be the MRCA.
again hope that all made sense and helps.
@@DNAFamilyTrees Okay thank you so much this is very helpful, I have two siblings in my family who took a DNA test for me to help me out with my family history research, they matched at 1925 CM's, they have lots of known maternal matches in common like cousins they grew up with & sometimes lived with, the younger sibling has the 626 CM Cousin match this cousin does not match the older sibling or any other family members on the mothers or fathers side that I can see with the exception of the adult children of the younger sibling & they match at 334 CM's & 336 CM's, this person is a surprise adopted cousin match and their biological mother is only 9 years younger than the younger sibling they match with, the older sibling has paternal matches on Through-Lines with the highest match being 149 CM's they have all a 2x great grandparent in common, the younger sibling has none so far, I did find out that when the younger sibling match was a teenager in the 50's for about three years or so they had lived about 2 blocks away from where the biological mother of the adopted cousin match was living at the time of his birth in the 60's, that is the only connection I have been able to find with regards to the younger sibling & the surprise cousin match. P.S. no one in the family recognizes any of the last names associated with the surprise cousin match "IE." their birth/adopted name & their mothers or fathers names, nor any of the names of the DNA matches the adopted person has in common with the younger sibling & their adult children.
I am 78 my highest match 404cM. Thought this was a half 1st cousin but your tool is saying more likely 1st cousin 1 removed. Am I reading this right ? Perhaps I need to re-watch this video.
Hugh, that is correct, the most likely is a dual path 1C1R, 2nd most likely is arguably 2nd cousin multi paths or a 1st cousin single path.
So if 1C1R or 2nd cousin this would mean this persons great grandparent is your grandparent. According to my tool that is the most likely and 2nd most likely MRCA (their Great Grandparent)
---
i looked at the shared CM tool as well for confirmation, if you enter 404 and in the results click on H1C you see it has 180/243 on either side of the mean of 404, for 1C1R it has 1015/1196 on either side of 404. For a 2nd cousin (which would be mulit path 2c, sharedcms has 197/92 on either side of the mean of 404. So their results appear to confirm this assertion.
--
What i think is important in determining MRCA is recognizing the required kinship to overlap. in this case 1c1r is 5x more likely to be the kinship (statistically) than H1C. 2nd cousin according to my chart is #2 most likely. This means statistically the #1 and #2 would mean that that persons great grandparent would be your grandparent. If you know your 4 grandparents then you have 2 of his 8 great grandparents, If he has the 8 and your looking for yours that means 2 of his are yours.
If however he has all 8 GGP and you have your 4 GP and none match, there is a DNA surprise in store for someone. OR, AND THIS IS IMPORTANT, OR its a multipath 2nd cousin with 4 or more paths and not a clean GP/GPP intersection!
This means you both would need to build out your trees another generation or two at which point if there isnt a surprise and this is simply multi path the names WILL cross at a generation or two above what was previously expected.
That multiple path scenario is a Significant factor and according to my tool the 2nd most likely.
in all cases YELLOW, 1C1R is statistically FAR more likely than a half first cousin. (Statistically. for the sharedcms tool the numbers would be yellow 2211 vs 423 for blue (h1c) and 289 red. Mine is more in line with 4583 yellow, 833 red 731 blue. Your looking at 1C1R being 5.6x more likely in mine, and 5.2x more likely in the sharedcms. in mine RED is 12% more likely than Blue according to the chart. the sharedcms tool has it 32% more likely to be blue than red. (but remember it also states they do not accommodate for endogamy or pedigree collapse in their stats)
So i would recommend looking at Yellow first for sure *his ggp and your gp* THEN id look Blue (as i said in video i look at this practically too, a relationship of 2 people is 100% required in all scenarios. Then a divorce or death and remarriage relationship is subset of that and depending on location blue is more PROBABLE than a red. (but in some areas and family lines id look red before even looking at yellow if you know what I mean.
I hope this wasn't confusing, but in short, yes, 1c1r is 5x more likely in both my tool here and in the sharedcms tool on dnapainter.
@@DNAFamilyTrees Hi Larry, Thanks for the reply. This connection was a surprise to me. My mother's father's side on paper does not have any DNA matches to me. I have been doing research for many years an the records show that by great grandfather spent many years in an asylum and his mother and grandmother did as well. The records show that this was heredity. I don't believe either my mother or I had this problem. My grandmother was living (in service) in the same small town on the 1911 census to my suspected unknown grandfather. He is the grandfather of this 404cM match. The great grandfather of this match would have been 67 at this time so unlikely to be involved with my grandmother who would have been 24. My mother was born February 1912. I guess is that my grandmother wanted a child but did not want the possibility of having one that was affected. I have further matches on the suspected line at 214.4 cM and 189.1cM on My Heritage. Highest probabilities being 2c1r and 3 cousin. Perhaps I may have to go back over my thought on this one based on your system but definitely re-watch your video a few more times to get my head around this. Thanks once again for your interest and thoughts.
@@hughwilliams7361 Since this 404 is on the "missing line" i would of course "cluster" all the matches with him. (now assuming ancestry that is easy and clean to to do).
then I would "filter" that list by "public trees" which will give you matches with him of ONLY those with public trees. Then i would go down that list, rick click on every one of them and "open in new tab" the tree for every match (up to 25 at first).
Then i would click tab to tab to tab looking at trees and just read the surnames on the farthest right of each tree aloud (or verbalize in mind). [thisis actually important as it helps trigger recognition of repeats]
Not any repetitions that occur in surnames, you will typically get 1 to 3 that are noticed.
this give you a place to focus for their connectivity, OFTEN you will spot a marriage between two people in seperate trees with different child names, this means you are very likely under that node since that is their MRCA and YOU also connect to each of them so it will be above or below that kinship.
Use CMS and this pattern to determine, are you UNDER, AT, ABOVE that node.
Then focus there until the answer found.
if you dont spot overlaps in those 25, open the next 25 and read the names there.
if still no overlaps, change the filter to "unlinked trees" and do the same (bit longer since you have to OPEN the trees on each of those profile pages to see the tree.
to be honest ive never had to go beyond this point for a clue except in foreign countries with few documents and trees (like some research i did with an Egyptian line and even then there was overlap but just not useful overlap because i had no way to research what id found)
Sounds like your on the right path and though process for the events, your close! Good luck!
I asked my sister in law who is color blind, and she can see all the colors including green on the chart, so that is positive.
Thank you Yvonne! no one had answered that yet! i did run it through the "simulators" and found that the colors chosen actually looked pretty good for all the variations of color blindness, thankfully! (would have been a bit to redo them so i was quite happy about that!)
Thank you for letting me know!
www.toptal.com/designers/colorfilter/ is what i used. Thank you for confirming it because what i used was a simulator not validated!
Thank you for this!!! I’m trying to figure it all out with my Ancestry matches. Quick question: I have 2 cousins who are siblings. Their mother is my mother’s first cousin. One of the siblings is 288 cM to me and the other sibling is 173 cM to me. Can you help me figure out what this means?
congrats on identifying their kinship!
What this is likely to mean is that you have another path to them other than the one you found.
This will be much farther up the tree and could branch at any node between you and them via the "other" parent at any one of those nodes.
I have a 1c1r at 441, her daughter is 209 but her son is 358. I noticed some quad path cousins at higher than normal CMS and i matched ALL FOUR great grandparents on that side so i was curious why they would be nearly 100cms higher.
I then found my mother ALSO had ties to two of the family lines of those 4! mcfarlin and brown MUCH MUCH farther up so it was 2 more shared paths to them! (which is why i am adamant about usage of terms single path, dual path, multipath vs. the arhaic terms of half kin and full kin and yeah there is endogamy. those terms aren't useful at all for research.
On the example above my mother ALSO has the same two lines to them HOWEVER it is diluted (for a different time/video) but since i anticipated stronger passage male to male and female to female, and the CMS MRCA is vias the mother, i created a tree with fathers line and found another connection to the spouse of the MRCA.
So there are dual paths but only ONE mrca in this case but it was closer and on other side of moms line but i inherited a good amount from the line that the Son in this inherited via his father (his daughter got it too but to a much smaller degree)
At first glance one would think maybe half siblings, but it is not, just simple inheritance at a distance :0 but in all cases it was multipathing that accounted for the variance.
Is this tool still online?
I have a 1st cousin 1x removed from my paternal line. We have spoken, she is in her 80s and we have matching trees.
The problem is, our shared DNA is 234, which to me points to a half 1st cousin 1x removed.
My paternal line is linked to Malta, and in my cousin's ethnicity report, she gets 6% Maltese (which is spot on for our tree). In the previous ethnicity report I had
well a 212 (234 with more from somewhere) would work to be at least a 1c1r. You provided her age without yours but since you provided her age i would then assume you felt it wold enough to mention thus 1 or 2 generations above you.
Which you suspect (aka your age) is important for first estimating. Ill go with one generation down. This would mean HER grandparent(s) intersect your line as MRCA(s), at least one but obviously 1 for sure.
I am guessing you also know this and has been identified since you have "matching trees".
Her match is 212 to you along this line and she has 234 to you which means another path. Since your trees match and based on what you said, I deduce the 212 is a single path. This means there is another path between you and her that is much smaller. we know at least 22 cm estimate perhaps slightly more since it is rare ot have a perfect split at 212 (or conversely less also possible but for estimation we will run with 22).
We "know" (or infer) this path is then a "single path" which helps where to look. 22cm single path appears for us at 3c1r with possible intersections above and below and since we "know" its above from previous information we can then infer the additional path will be most likely at here 2GGP and your 3GGP. (pretty far up but again we are looking for 22.
NOW there is another option, EQUALLY likely is that your related via your OTHER parent and perhaps even HER other parent so you 4 "paths to evaluate" your same parent line via that same line farther up, that same parent for you to her other parents line, your other parent to her same parents line farther up and your other parent to her other parent.
The match is most likely at her 2GGP (any of them!) and also any of your 3GGP (any of them from either side), it will take a bit of looking to find that match.
OR that 22 could actually be DUAL from two slightly off from 13s rather than from a single 26 but good news is w would be looking at the same area but instead of 3c1r it would be a 4c. Your line would still be a 3GGP but on hers instead of 2GGP it would then be 3GGP.
And lastly it could be simply a high 212 :)
we look at the others are probable based on the ethnicity twist you included.
But sounds like you really get this so i have no doubt youll figure it out before long! Key to this one will be the A) Ethnicity in combination with B) Clusters. Thats how youll decide which line to find that 22cms on.
Let me also add, that I have no close matches on my paternal line apart from this cousin. In fact, this cousin is the only match of 3 that are above 100cms (the other two are maternal), and the next highest match which I suspect is in the same paternal line shares 32cms, though there's no tree, so no way of finding any link.
how many people share DNA with this DNA match (even at lower CMS)? one trick is to go to that persons profile, click shared, THEN on the shared page, filter on "public trees". Then i right click and openin new tab every tree (30 at a time) to review. I look for duplicate names or marriages (or birth/death dates as people OFTEN changed name upon coming to America for a fresh start.
Maybe you can glean something useful this way?
@@DNAFamilyTrees My 1c1r cousin shares no matches at all.
The person who I think is most closely related on that side of the family also has no shared matches, but other lower matches have him in their shared matches, so he's part of the same cluster, but he's the highest CM match of them all. It's all quite odd, but I seem to recall it was not like this before a couple of updates ago.
My father's paternal, and his grandmother's line seem like there's no hope to find any link in DNA.
Malta is such a small country, and his grandmother was born in Konigsberg (now Kaliningrad) and I have been unable to find any records of her birth or parents - I imagine becuase of WW1 and 2.
I'm British, so most of my tree is easily tracable, and I have linked up trees with many of my DNA matches from the British lines. It's just the ones outside of the UK that I cannot trace so well, and it seems like either the line didn't result in many more closely related current living relatives, or they just have not taken the DNA test.
edit: In relation to your earlier question, I'm 35.
My mother has taken the test, and she does not match with any of them.
I have a unknown parent case and a 442 that I can not figure out. The 442 is the closes match on bio dad side 🤷
442 is likely a “Half” first cousin aka that person’s grandparents (1 in 4) will be same as your matches
Next possibility is a 1c1r in which case the match would be a generation older/younger which would determine whose grandparents (the others great grandparents)
Long shot is 2nd cousin but only if there are intermarriages
Not likely
My husband did 23andMe with 3 of his kids. All half siblings. They couldn’t get one’s dna so she couldn’t participate. But the other 2 matches at 2034 cm. I just did mine. I have a half sister as well. We matched at 2023 cm. So imo idk how a half sibling would even show up that low.
He (1985) also has a 1C1R (1972) shares 519 cm his first cousin (1980) is 5 years older than him and shares 585 with this 1C1R. Idk where to start. All of our matches are people who I don’t have in my tree.
th-cam.com/video/3ilnR5Q8nb0/w-d-xo.html based on last post and this one i recommend this video.
as for the half sibling, they are a "Single path" to you. aka via one parent. so your parent is 3400 to you and they half of that 1700 (rounded numbers for simplicity sake as the actual numbers take into account the amount of ancestral DNA passed to you via that parent that they also received.
This is really helpful, thank you so much! It's funny though, both my sister and I share 2,193cm and 2,244cms with our paternal aunt. It comes up as our aunt by way of duel path. However, my sister's daughter shares 1,298 cm with this our aunt ( her great/grand aunt ) and it shows a duel path relationship as 1st cousin or a multi path great aunt. I admit I'm just learning all of this, but this confuses me :)
This why i love this tool, it goes beyond the "they are related at this kinship" and tells us more about the relationship that can be useful i other research as well.
This is simple but complex so ill try two ways :
in the simplest format, you to your parent would be 3400. (one step) you to an Aunt would be 3 steps, up from mother to her mother and then down to the Aunt. thats 3400>>1700>>850. (this would be single path or "half" aunt.) if the Aunt has more than the 850 significantly this means there is an additional path to her so we add another 850 from the other parent. which gives us 1700, the mean or "average" expected CMS
in your case and that of your sister you have yet even more CMS. another nearly full paths worth. This is typically due to additional "pathings" at a distance that can be on the same side as the same parent that came to you OR via a path coming through the other side of your parentage.
The second scenario is EXTREME randomization from a grandparents to BOTH your parent and your Aunt. Instead of passing a typical near 50% from each of their parents on, They passed along 50% more of one of their parents lineages than the other to BOTH your parent when they were created and to your Aunt when she was created. While singularly this is uncommon its not so uncommon as to raise an eyebrow. for it to happen again is quite unusual.
How unusual? well if you put the numbers in the Sharedcms tool, their data set for aunt/uncle has 2695 people only 12 are above the 2244 and the 2193 are only 53 above it so in perspective thats 0.004 and 0.019 which is EXTREMELY rare outside the case of twins or identical twins (Identical twins are 0.003 to compare to the 0.004 number for how rare that is.
So while it is possible this is a "normal dual" and by normal i mean its on the edge of acceptable normalcy because thats a LOT OF EXTRA CMS or more likely, there is another factor yet undiscoverd at play.
If this were mine, a couple hundred either way would not raise an eyebrow. but since a single path is 850 ad we are talking 493 and 544 cms higher, more than 1/2 of a single path and assuming Identical twins or twins scenario not a factor, again, possible but this is high enough id be HIGHLY suspicious of other Paths. (more likely more than one that add to that extra approx 500cms each.
There are ways to find out where the paths are if there are indeed more or to confirm their existence.
If it is a natural (while unusual) distribution, then your aunt (and your parent) will share more DNA (significantly more, very noticably more) CMS with one side of the cousins than the other side of the cousins.
so if you have first cousins on both sides who we would tend to expect around the 850 mark, one side should show over 1000 and the other side would show closer to 600 to your mother or aunt.
If this is apparent then its a natural split and you can use this information in researching lines! (and it will help when their numbers dont fit the norms (above or below).
Now if one side is normal and theother side slightly elevated the extra path will be on the NORMAL side somewhere.
If both sides are normal for cousins then the extra paths will be on the other parents side somewhere, very likely multiple paths up as many as 4-7 generation as a single inherited path of that extra amount (you/3400/1700/850/425) would go up to the 5th node and then the same concept of normalization apply. lets say the GGP on other side intersected, the same line then that would pass 450. OR up another would each pass 225 up another 4 passing 106. Since its highly unlikely your going to find a Dual or quad direct line, then multi paths tend to be much higher up and at varying spots across rather than direct line intersections.
While this sounds "UNLIKELY" on the surface,l in reality its HIGHLY LIKELY! so much so that we have all heard the term on the concept "en.wikipedia.org/wiki/Six_degrees_of_separation" Six degrees of separation. This concept is that ALL PEOPLE, every friend you meet in the street is an average of 6 "nodes" from being friends from each other. That concept of "knowing" also applies to kinships (albeit NOT at 6 and no studies have conclusively proven this in kinships) the idea and principle of "knowing" does in fact apply to kinships in that if EVERY person, EVERY PERSON, is 6 handshakes away from each other, then its is probably that at this point, two random people who marry would be 6 handshakes away. and since communities in past were small, this DOES often translate into kinship! in fact the LDS has "genealogy" nights were they often have members do this as a sample to see how close married couples are related. my friend told me he was the closest with his wife at 7 generations, most were 8 and 9 generations (they had no idea). My father #1's parents were 7th cousins and never knew it (and thuse would fit the scenario your sharing now.) I have helped MANY people who have people in their trees that didnt know they were related (one as close as 2nd cousin!)
Other tools simply show that your kinships "FIT" this goes a bit beyond that because i did this with the underlying principle that i wanted to find BIOLOGICAL parentage thus every clue like this is huge.
sorry for the long response, hope it makes sense. id guess you have paths from other side or that the parents shared in common on that side may indeed be 7th cousins like my grandparents in family #1 was or like the friend who shared his kinships with me last week from their church exercise :) Both scenarios (which again 6 degrees of separation in mind) are both likely scenarios. But there is definately more probability of additional pathing than this being two 80/20 and 80/20 splits from one grandparent via parent to both your parent and your aunt and also from you to your Child similarly (again its possible, but thats not as likely as additional path possibility)
@@DNAFamilyTrees Thank you so much for such a detailed reply! While I don't completely understand ( yet lol ), I get the gist :) I can certainly understand why there could be multiple pathways further back. My sister and I are primarily Irish descent and both Ancestry and 23andme list counties Kerry & Cork as primary places where we share DNA. I'd guess that both my parent's ancestors come from those counties that are right next to each other. I still don't understand why both my sister and I get the correct relationship with my aunt via dual paths, yet her daughter doesn't. I'm going to save your reply for future reference though! Once again, thank you very much :)
It is still 50 even after refreshing
ok ill get that change out pretty soon :) its not just you :)