RootsChat.Com
General => Ancestral Family Tree DNA Testing => Topic started by: Will Stevenson on Wednesday 03 May 17 15:56 BST (UK)
-
What is going on. I have very different DNA result for all 3 companies that I have used. This is now making my doubt this whole DNA thing. Can someone please try and explain to me what is going on ? Here are my results.
Thanks.
My Family Tree (Autosomal) DNA came out as 99% British.
I also did Ancestry and my DNA and results were,
Great Britain 39%
Scandinavia 26%
Europe West 13%
Ireland 9%
Iberian Peninsula 8%
Low Confidence Region
Finland/Northwest Russia 2%
Europe East 1%
Italy/Greece 1%
Then there are my 23 and Me Results. Different again
British & Irish
55.3%
Scandinavian
11.8%
French & German
10.8%
Broadly Northwestern European
21.2%
Eastern European
0.8%
Sub-Saharan African
< 0.1%
Middle Eastern & North African
< 0.1%
-
What you have discovered is what we all find out eventually and that is that DNA cannot give you your ethnic inheritance. What these DNA testing companies give you is their best guess based on the algorithms they use and their sometimes very small sample populations on which they base their definitions of one ethnicity or another.
Of course it gets better as periodically the companies recalculate ethnicities and you can suddenly find your alleged ethnicity changes significantly.
The truth is the science is no where near as advanced as it needs to be to make the claims the companies are making. They are in short snake oil companies selling near worthless products to the gullible. That said I still think its worth testing as you get an idea and maybe even a surprise or two and the accumulating collection of tests results may lead to better predictions in the future but in the meantime these ethnicity DNA results should just be treated as a bit of fun and not taken at all seriously.
This is just my personal view so anyone please do not be offended if you do not agree with it
-
Thanks David. All opinions are important. ;)
-
The different results don't surprise me.
In any case, knowing "your" DNA profile is not really all that helpful with tracing your own direct ancestry -
- but it's a nice, not-so-little earner!
-
Each company uses different algorithms and different reference populations. They also cover different time periods. Most people do get very different results from each company. These admixture reports are a bonus feature. The main reason for testing is to get the DNA matches.
I wrote a blog post comparing my results and I included links to various resources for further reading:
https://cruwys.blogspot.co.uk/2015/05/comparing-admixture-results-from.html (https://cruwys.blogspot.co.uk/2015/05/comparing-admixture-results-from.html)
-
Thanks for this, and especially for the link to Debbie's blog. I've just done an autosomal test with Ancestry so while I wait for the results I thought I'd do some reading on the subject.
While I'm more interested in cousin matches and finding people who are researching the same ancestors the ethnic inheritance question is also interesting. Like Debbie my documented roots are very English - 31/32 of my 3x great grandparents were born in England with the remaining one born in Wales. On branches where I've gone back further everyone has been born in England. It'll be interesting what Ancestry tell my about my ethnic inheritance!
-
I think I've just saved some money...just when I was getting almost on the part-way road to maybe getting perhaps tested. Possibly.
Martin
-
While I think it is great if people want to do this for themselves, it seems expensive and also rather unsettling that someone in an official status (or not)... might one day be able to demand or even buy these results and have my DNA.
A few distant cousins have found me because they have the DNA of other cousins who have already contacted me...but truthfully, more contacts have been made from being right here on Rootschat or from other online information. J.J.
-
Thanx for this post, i was thinking of getting mine done but hesitated for these same reasons, looks like the companies just want our money. Don't think i will bother now.
-
I would never want to take it away from those who would like to have it done, though... Our choices are all ours to make. :) I liked Martin's answer I was "almost - maybe - perhaps - possibly " going to get it done, hehe...
-
After 30 years of research and publishing nine 600 page traditional family history books, I found myself in being reactive to requests to add more distant relatives each year (around 4,000 new cousins came in each year that were unsolicited). These additions were becoming far removed from the surnames of my ancestors and I was making minimal progress on my brick walls of ancestors. Around ten years ago I decided to dive into YDNA testing as it seem to have promise of breaking down brick walls. So I decided to only concentrate on my parent's surnames at birth (Casey and Brooks).
Here is what I have learned to date on my Brooks line: The two oldest children of my oldest proven ancestor were for some reason omitted from an extensive estate of my ancestor, Robert Brooks, Sr. I knew for sure that these two young men actually lived in the same household (via personal property tax listings that revealed their names when poll tax was collected). Also, these two signed several marriage bonds for those included in the will and after the father died, nine of these children moved from Virginia to South Carolina and lived next to where these two omitted sons lived. For years, we just thought that these two older sons must have received funding from their father or moved away to South Carolina and were left out of the will.
There had always been undocumented family history that the maiden of this ancestor, Robert Brooks, Sr. married Brambly "Wade." The given name of Brambly was listed in many deeds but we could not verify this marriage via marriage records or Bible records. So this family lore was only casually mentioned in my sketch of this couple with the additional information that several Wade men were determined to be neighbors of Robert Brooks, Sr.
So along came YSTR testing. As a Brooks admin for several years (now pretty inactive), I became quite discouraged that every fourth Brooks tester was yet another different genetic cluster that was not related to any other genetic cluster in the last 1,000 years. I quickly became much more aware that names based on geography or trade just was not very reliable for relatedness. So, I tested the second Jordan Brooks in the south that lived in the south between 1780 to 1830 who resided within few miles of my ancestor, Jordan Brooks, b. 1765. YDNA testing revealed no genetic connection in the last 4,000 years. This means that common surnames and geography are even not enough to be very reliable. I had already compiled 500 descendants of this second Jordan Brooks - I really had a hard time letting this connection go (and I had become the focal point for the documentation of this line as well).
YSTR markers were upgraded to 111 markers and YSNP testing made many matches even more closely related. Plus the number Wade testers that matched continued to grow and continued to be way under 1,000 years. So, I now believe that Robert Brooks, Sr. did indeed marry Brambly Wade around 1730 but Wade was probably her married name - not her maiden name. They also later named a son Wade Brooks. This would explain why the two oldest sons were omitted from the will as adopted children rarely received in inheritance. This is really only very strong evidence supporting this kind of connection - but as more genetic information continues to become available for even more recent connections this adoption will some day be pretty well confirmed due to YDNA testing.
-
Part 2 -
For my Casey line, we have between 50 and 75 males living in western South Carolina prior to 1800 and less than a dozen have been connected via traditional research (over 50 researchers have spent time in these counties and archives with very little being added over the last 30 years). More than half these connected lines come from one of the wealthy families that were published during the 1876 centennial. We now have around twenty-five Casey testers from South Carolina tested to date. Here are discoveries:
1) All but one Casey line is genetically related in the last 400 to 600 years, possibly much more recent. The other line is related at about 1,400 years ago - so genealogy will probably never connect these two lines since surnames not used for 400 years of this time frame. But this did validate what we suspected, that these lines were indeed closely related - except for one line.
2) A lot of the Internet genealogy has extended this line back to Virginia based on a manuscript that published in the 1960s. Several of the Virginia lines have been tested are not related in the last 2,500 years. So genetic evidence has really discounted this Virginia connection - yet no updates in Ancestry.com to date.
3) Very early on, we knew that our genetic cluster was extremely isolated from all other testers - anyone who matched our very unique signature with very unique marker values was probably going to be related in the last few hundred years. We had a very active Hanvey researcher that matched our line. After 1830, all his tests matched the Casey cluster and before this date they all properly matched a well defined Hanvey signature. This person did have Casey probate records that gave property to his Hanvey ancestors and we have now concluded that his ancestor was born a Casey, orphaned and then informally adopted by a kind Hanvey neighbor. YDNA is really good at revealing these types of adoptions.
4) Early on, we also discovered that we had one YSTR mutation where half the lines belong to one very unique value and the other half belonged to an even more unique marker value (all ten only belong to our cluster out of 50,000 YSTR testers under haplogroup R). So this is a very well defined branch that splits our cluster into six lines belonging to the older branch and five or so belonging to the younger branch. Due to significant geographic ties, I had compiled around 2,000 descendants of four possibly Casey related lines. Three belonged to my younger branch - but one belonged to the older branch. So I now know that this other line is much more distantly related by several generations. Very recently, a new line that I have never researched now belong to my younger branch.
5) We finally found a solid match that is around 400 to 600 years old. His surname is Kersey and to everyone's surprise, his line has been in England since the earlier 1600s traced back to Oxford, England. We suspect that his line moved from Ireland to England and switched his name from Casey to the English surname Kersey probably to hide his Irish heritage.
-
Part 3
6) We definitely know that Casey is a Irish surname - the surname according Clan history were supposed to be descendants of King Cas - a direct ancestor of King Brian Boru. This supposed to be the largest sept of the Casey clan. We are very fortunate to have Sir Conor O'Brien tested who is the 40th male descendant to hold the title of being a direct descendant of King Brian Boru. This title has been granted to each generation over the last 1,000 years. Since the South Carolina Casey line is the most closely related to King Brian Boru line, we are now pretty certain that the South Carolina Casey matches Clan history as being a descendant of King Cas. Of course this connection is around 1,100 years old but it does validate the Clan history. Also, the second largest sept of the Casey clan was supposed to be distantly related several hundred years earlier and our second largest genetic Casey cluster is around 1,400 years old.
7) We now have 560 testers at 67 markers that are known or predicted to be L226. We know that King Brian Boru was part of the Dal Cais tribe that originated primarily in County Clare. Around 40 or 50 L226 testers have traced their lines back to counties in Ireland. Around 80 % of these testers list only five southern counties in Ireland as birthplaces. These are Clare, Tipperary, Cork, Kerry and Limerick - all neighboring counties in Munster, Ireland (southern part of Ireland). We also know that the great majority of surnames in the project are Dal Cais surnames according to the history books about the Dal Cais. This information is very relevant as we can know narrow our Irish research to only five counties of Ireland.
8) Around one year ago, we had our second NPE (adoption, etc.) join our genetic cluster. Due to the extreme isolation of YSTR markers (and recent YSNP branches), this person immediately concluded that he must have Casey origins. But this Meredith line was born in the early 1800s in Virginia (the only line tied to Virginia). This is kind of an outlier for our cluster.
9) Around two months ago, we had another major breakthrough as we finally had our first Casey to join this cluster to have no known ties to South Carolina and the only Casey line in this cluster to have very early ties to North Carolina. However, this tester is in the same time frame as the Kersey tester and we have concluded that his line is probably not connected in the last 400 years and probably migrated from Ireland directly to North Carolina independent of the South Carolina migration from Ireland to America in the 1750s.
10) I was the first L226 tester to take the Next Generation Sequencing test which reveal 55 mutations that was unique to me. Over the last two years, eight of these private YSNPs are now permanent branches on the tree of mankind. Over the next few months, there is a very good chance to add two new branches based on my private YSNPs (one that is over 1,500 years old and one in the last 400 years old). You can either be reactive and let others advance this serious research or you can be very proactive. Two years ago, the most recent YSNP branch was L226 which became prolific in offspring around 1,500 years. As of today, we now have 52 branches under (one third are genealogical branches under 1,000 years old and mostly dominated by one surname). We are now finding a branch about every other week now.
11) For my particular part of Caseys under L226, we known have two genealogical branches - FGC5647 (400 to 600 years old) and FGC5639 (300 to 400 years old). If you test positive for FGC5647, you probably have a Casey ancestor and originate from one of five counties in the last several hundred years. With the new North Carolina Casey tester, we could soon add a third genealogical YSNP branch soon.
Many may think that this is minimal progress - but most of these discoveries would never happen with traditional genealogical research. This is just the start of very long discovery process. The long term promise of YDNA testing in truly unbelievable. In only two or three years, YSNP branches will average 1.5 generations and with 500 YSTRs added, each of our ancestors on our pedigree charts will average three or four unique mutations unique to this particular ancestor. Pretty exciting times.
-
I just now saw that Mr Casey, just above me on this forum, has Ellison in his family tree. My maiden name was Ellison. Furthest back is John Ellison born in PA and moved to Green and Cocke Co's, TN.
I have tested at FTDNA and was gifted a 23 and Me kit and then I took my raw data to My Heritage. Most of the results are fairly close to one another except for a few things, like the amount of Scandinavian, Iberian or trying to place the odd bits that I have in my DNA. I've seen Middle East, North Africa, Near East and 23 and Me put me in West Africa, which surprised me. DNA Land just was honest and said, well for about 6 or 7% of you, we haven't a clue! :D None of the major admixture surprises me and it's the odd bits that fascinate me the most. I'm happy to not be 100% European and wishing I had that crystal ball to see my deep ancestry and find out just what those odd bits are for sure. AND I must say, if I'd never done DNA for my paternal aunt, one of our family mysteries never would have been solved. I'm happy I spent my money on doing my DNA and the DNA of a few relatives.
-
Nylnasus - My Ellison line is quite remote in my pedigree chart. Here is what I known about my ancestor Polly Ellison who married my ancestor, Ambler Casey in 1809 in Roane County, TN:
http://www.rcasey.net/acrobat/cas0503n.pdf#Page=276
It has been over ten years since I have updated this family history but I have been the FTDNA admin for the Casey project for over ten years and have around 25 testers who are relatives of Ambler Casey.
http://www.rcasey.net/DNA/R_L226/Haplotrees/L226_Home.pdf#Page=28
Unfortunately, I know very little about my Ellison line (including YDNA testing).
-
Part 2 -
...1) All but one Casey line is genetically related in the last 400 to 600 years, possibly much more recent. The other line is related at about 1,400 years ago - so genealogy will probably never connect these two lines since surnames not used for 400 years of this time frame. But this did validate what we suspected, that these lines were indeed closely related - except for one line.
I am a novice at DNA testing but wouldn't everyone in north western Europe have been related in some way '1,400 years ago' ?
3) Very early on, we knew that our genetic cluster was extremely isolated from all other testers - anyone who matched our very unique signature with very unique marker values was probably going to be related in the last few hundred years. We had a very active Hanvey researcher that matched our line. After 1830, all his tests matched the Casey cluster and before this date they all properly matched a well defined Hanvey signature. This person did have Casey probate records that gave property to his Hanvey ancestors and we have now concluded that his ancestor was born a Casey, orphaned and then informally adopted by a kind Hanvey neighbor. YDNA is really good at revealing these types of adoptions.
Wouldn't illegitimacy be a more plausible explanation than being 'orphaned and then informally adopted by a kind Hanvey neighbor...'?
5) We finally found a solid match that is around 400 to 600 years old. His surname is Kersey and to everyone's surprise, his line has been in England since the earlier 1600s traced back to Oxford, England. We suspect that his line moved from Ireland to England and switched his name from Casey to the English surname Kersey probably to hide his Irish heritage.
'The name 'Kersey' seems to heavily cluster (1881) in the Suffolk area of England.
It could also be that the some of the 'Casey's' in Ireland originated from an English 'Kersey'
Why would someone in England 'hide' their 'Irish heritage'? Irish and English names are both found in old records in both countries
-
Part 2 -
...1) All but one Casey line is genetically related in the last 400 to 600 years, possibly much more recent. The other line is related at about 1,400 years ago - so genealogy will probably never connect these two lines since surnames not used for 400 years of this time frame. But this did validate what we suspected, that these lines were indeed closely related - except for one line.
I am a novice at DNA testing but wouldn't everyone in north western Europe have been related in some way '1,400 years ago' ?
Not at all, just looking at the major very old haplogroups in UK and Ireland, you have R1b, R1a, I, J and E - all of these are well over 5,000 years old. There are dozens of YSTR signatures that are predictable for YSNPs and are fairly prolific in numbers. Most of these are from 1,500 to 2,500 years old. Some of major haplogroups, L21, U106, DF27, U152 and other very large haplogroups are five to twenty percent of UK & Ireland and are generally 4,000 to 5,000 years old.
3) Very early on, we knew that our genetic cluster was extremely isolated from all other testers - anyone who matched our very unique signature with very unique marker values was probably going to be related in the last few hundred years. We had a very active Hanvey researcher that matched our line. After 1830, all his tests matched the Casey cluster and before this date they all properly matched a well defined Hanvey signature. This person did have Casey probate records that gave property to his Hanvey ancestors and we have now concluded that his ancestor was born a Casey, orphaned and then informally adopted by a kind Hanvey neighbor. YDNA is really good at revealing these types of adoptions.
Wouldn't illegitimacy be a more plausible explanation than being 'orphaned and then informally adopted by a kind Hanvey neighbor...'?
Illegitimacy is a certain percentage as well as rape and many other ways - but war, famine, disease, accidents left many orphans. When men died with a young family, the wife really needed to marry again pretty quickly to provide a shelter and food for herself and children - resulting in a lot of informal adoptions where the children just assumed the male's surname early in life.
5) We finally found a solid match that is around 400 to 600 years old. His surname is Kersey and to everyone's surprise, his line has been in England since the earlier 1600s traced back to Oxford, England. We suspect that his line moved from Ireland to England and switched his name from Casey to the English surname Kersey probably to hide his Irish heritage.
'The name 'Kersey' seems to heavily cluster (1881) in the Suffolk area of England.
It could also be that the some of the 'Casey's' in Ireland originated from an English 'Kersey'
Why would someone in England 'hide' their 'Irish heritage'? Irish and English names are both found in old records in both countries.
My particular Kersey in England is L226 where 90 % of testers list Ireland as there country of origin and L226 is around 2,500 years old (when counting branch equivalents). That is so many generations that any other component would become close to zero. It is very common for Irish individuals who were living in England to change to more Anglo sounding names. Gaelic names were converted in mass to English centric names once the English began to colonize Ireland. Casey is not the original version of the surname which was Gaelic "O'Cathasaigh" and became O'Casey. Even the O portion was dropped later as well.
After Cromwell invaded Ireland, things change radically. At least half of the land owned by Irish was transferred to English landowners, so Irish people had to adapt to this new environment. A lot of English surnames are now used by Irish people today. Under L226, you have Lynch, Butler, Smith, Thomas, etc. who are L226 (very Irish) but carry English surnames. There are also many people with Irish surnames that have haplogroups that are rare in Ireland and probably come from other European influences as well. For many haplogroups, it is pretty hard to tell where their origins are, so it could either way for these haplogroups.
-
Not at all, just looking at the major very old haplogroups in UK and Ireland, you have R1b, R1a, I, J and E - all of these are well over 5,000 years old. There are dozens of YSTR signatures that are predictable for YSNPs and are fairly prolific in numbers. Most of these are from 1,500 to 2,500 years old. Some of major haplogroups, L21, U106, DF27, U152 and other very large haplogroups are five to twenty percent of UK & Ireland and are generally 4,000 to 5,000 years old.
Hi again, thanks for the replies
Being 'related' does not only rely on Y haplo dna. Y haplo is only one direct line
Illegitimacy is a certain percentage as well as rape and many other ways - but war, famine, disease, accidents left many orphans. When men died with a young family, the wife really needed to marry again pretty quickly to provide a shelter and food for herself and children - resulting in a lot of informal adoptions where the children just assumed the male's surname early in life.
The above is conjecture. It is not proof the person in particular was 'adopted'.
The most likely cause for a different surname match is still illegitimacy, it was not unusual and especially in the 18th century
My particular Kersey in England is L226 where 90 % of testers list Ireland as there country of origin and L226 is around 2,500 years old (when counting branch equivalents). That is so many generations that any other component would become close to zero.
I have tried to find more info about 'L226' and the only place I could find it mentioned is in two Irish American family history thread blogs/projects.
It is very common for Irish individuals who were living in England to change to more Anglo sounding names. Gaelic names were converted in mass to English centric names once the English began to colonize Ireland. Casey is not the original version of the surname which was Gaelic "O'Cathasaigh" and became O'Casey. Even the O portion was dropped later as well.
It would not have mattered where they lived or whether someone was English or Irish. All surnames were anglicised. French sounding English surnames were also anglicised. Most people did not know how to spell their own names and so their names were recorded as heard by the recorder and how he thought it should be spelled. It wasn't a 'mass conversion', it was a gradual process
After Cromwell invaded Ireland, things change radically. At least half of the land owned by Irish was transferred to English landowners, so Irish people had to adapt to this new environment.
Yes civil war tends to do that. The English peasantry had to adapt too land enclosure too...
A lot of English surnames are now used by Irish people today.
I see many American family history researchers who say this.
Under L226, you have Lynch, Butler, Smith, Thomas, etc. who are L226 (very Irish) but carry English surnames. There are also many people with Irish surnames that have haplogroups that are rare in Ireland and probably come from other European influences as well. For many haplogroups, it is pretty hard to tell where their origins are, so it could either way for these haplogroups.
The surnames you mention are common and found all over Britain and Ireland
Again, I am not sure what the significance of 'L226' is or why it means that a tester with this result is 'very Irish'. The implication seems to be here that a tester who does not get this special L226 result is therefore NOT 'very Irish'. That they might be 'just a little bit Irish'? That, even though they might have an Irish, or even English/Scots/Welsh surname and been in Ireland since time immemorial, that they do not quite pass this 'Irishness' test?
I think my point is, as I mentioned in the other topic, that the people of Britain and Ireland are very mixed together, they are the same people. They share the same/similar distribution of Y haplo groups. I think this why people here are sceptical about the tests and how much they can tell us that we don't already know by the paper trail, and also from what we already know about history and migration.
One of my grandchildren has 'Casey' ancestry, this is why I am trying to understand it all btw.
-
I have tried to find more info about 'L226' and the only place I could find it mentioned is in two Irish American family history thread blogs/projects.
Here is my web site on the R-L226 haplogroup (the three links in the haplogroup pull down menu are updated weekly but the other content is not updated very often):
http://www.rcasey.net/DNA/R_L226/R_L226_Private.html (http://www.rcasey.net/DNA/R_L226/R_L226_Private.html)
Another good web site is maintained by the lead admin for the R-L226 project - our YSTR signature was the third signature (YSTR pattern) that is believed to be 90 % Irish in origin and between 1,500 to 2,500 years old (the older date includes L226 equivalent branches). This is why this web site states Irish Type III:
http://www.irishtype3dna.org/index.php (http://www.irishtype3dna.org/index.php)
I am only aware of three haplogroups that are very Irish in origin: R-M222 (largest - 3X larger than L226), R-L226 (second largest) and CTS4466 (third largest - pretty close to L226). This represents only five percent of Irish testers, so haplogroups can only identify five percent of the Irish testers, the rest are pretty mixed. M222 and L226 are connected to the first two kings to unite/conquer the entire island of Ireland and enjoy the spoils of war to become prolific.
A lot of English surnames are now used by Irish people today.
I see many American family history researchers who say this.
This is based only on R-L226 genetic testers only and the known family history associated with these testers. To date, for 90 % of the testers who predicted or verified to be R-L226, they list Ireland as their place of residence for any European countries. In fact, R-L226, has 80 % of the testers who show a county level residence are from only five counties in southern Ireland: Clare, Tipperary, Cork, Kerry and Limerick. Since R-L226 was proven to include King Brian Boru and his ancestry, the original origin is probably limited to County Clare. One of our testers is Sir Conor O'Brien, the official title holder of the O'Brien surname - title that has been passed to only to male O'Brien descendants over the last 1,000 years (it goes through a formal approval process every generation for around 40 generations). The rise in prolific offspring of R-L226 tracks very closely with the rise to power of Dal gCais tribe, the ancestors of King Brian Boru.
One of my grandchildren has 'Casey' ancestry, this is why I am trying to understand it all btw.
Here is my web site on the Casey DNA project:
http://www.rcasey.net/DNA/Casey/CaseyDNAProject.html (http://www.rcasey.net/DNA/Casey/CaseyDNAProject.html)
If you grandchild can prove her Casey line back to 1870 (with ties to the five counties mentioned for R-L226) and can locate a living Casey descendant of this line, I will send this living male descendant a FTDNA 37 marker test kit at no charge ($169 value). I descend from one of the earliest larger migration of Irish to America in the 1740s during the large crop failures driven by significant climate change for several years. They were residing in western South Carolina in the early 1750s and by 1800, there were around 50 adult Casey males living in three counties of South Carolina.
Three of our Casey genetic lines are R-L226 and represent around one-third of all Casey's in the world today. We also have a few Casey's that were R-M222 as well. The rest have a very varied haplogroups and around 20 % have not tested enough YSNPs to know which haplogroup they belong to. This based on the 75 Casey testers to date which is very biased with American testers with very few Irish testers. However, the number of people of Irish descent are spread through out the British colonies:
Casey tracks the averages: Ireland 1X, USA 8X, England 1.2X, Australia/NZ 1X and Canada 0.5X
Hogan tracks the averages: Ireland 1X, USA 7.6X, Australia/NZ 1.3X, England 1X and Canada 1X
McNamara tracks close: Ireland 1X, USA 5.6X, Australia/NZ 1.8X, England 1.2X and Canada 0.6X
http://forebears.io/surnames/hogan (http://forebears.io/surnames/hogan)
-
I don't think these tests are worth taking. Mine came back virtually all British, even going back thousands of years. You only have to look at my dad to see he's not from British stock! So what's going on here?
-
The geographic analysis done by all genetic testing companies (and even at GEDMATCH) is based on very unreliable technology. It does a pretty reasonable job at the continental level but there is a lot of misleading marketing on the accuracy and consistency of geographic mixtures. The advanced forums are full of complaints about these parts of these tests. This unfortunately really detracts from the matching part of the atDNA tests which can be useful. But even these are now being marketed beyond their reasonable limits of accuracy.
Also, the simplistic tools of just raw matching based on shared segments makes individuals think there are relationships when there are not. You really have to triangulate segments to match your ancestors to really know which matches belong to each line. Most people just do not want to spend the time and funds for this. They do not want to spend the time to download raw results for upload to GEDMATCH for much better tools.
YDNA is being over-marketed as well - but at least this testing will pay off in the next few years and for some lucky testers are already began to yield some truly significant progress. But again the FTDNA matching system is way too simplistic in nature and most do not want to spend the significant effort in understanding better ways to analyze. We really need better tools but end users want them for free. As the actual testing continues to go down in price, the software and IT costs are now probably more than the actual testing. Eventually, these two parts will have to be separated unless you are happy with the minimal unreliable tools that vendors currently offer.
-
Yes I agree with you Robert, I do think these tests are being over-marketed and they are beyond reasonable levels of accuracy at the moment, it may well be better to let the tests develop over time and then re-take them. Then we should hopefully receive some with more meaningful results!
-
Yes I agree with you Robert, I do think these tests are being over-marketed and they are beyond reasonable levels of accuracy at the moment, it may well be better to let the tests develop over time and then re-take them. Then we should hopefully receive some with more meaningful results!
I do not understand why people think that niche companies like genetic genealogy testing companies will be any different from car advertisers. The atDNA companies are much worse than the YDNA testing companies though. But all this hype does get people interested and eventually they get more serious about atDNA segment matching and advanced YDNA testing of YSNPs.
There is little doubt that in 10 to 20 years, YDNA testing will be key to genealogical research and will become primary sources of information as the accuracy and coverage improves. We already know that if you test Y5610 (YSNP) positive that you are direct descendant of King Brian Boru - so there is some serious progress being made already. However, not all geographies are evenly tested and many lines have barely survived leaving few males to test. Under Y5610, we also have five YSNP branches and nine YSTR based branches that help sort out all of the O'Briens.
I was the first Next Generation Sequencing test for R-L226 and now have seven permanent branches of my Irish ancestry that was discovered via my first test. You can either wait for the everyone else to discover your branches for you or you can be proactive in this major advance in genealogy. Eventually, every male person on our ancestry chart will average five or so unique YDNA mutations to identify all male descendants - so the future looks extremely bright in the near future.
-
Ah well, no male directs in my line left. Save me a fortune that will!
-
Ah well, no male directs in my line left. Save me a fortune that will!
You are stuck with the traditional genealogical option until you can discover your male ancestor that has living descendants. My great-grandparents that were Shelton's daughtered out as well. But I was able to go back one more generation (born in 1811) and found his brothers as well and now have over 1,000 male descendants charted. atDNA could solve this problem if you ancestor is not too high in your pedigree chart.
Having is no male descendants is one major limitation in YDNA but even greater is that many lines barely survived over the last 1,000 years and only have a handful of male descendants. These lines are subject to daughtering out as well. So test now on mother's YDNA line or get busy with traditional research to get through your brick wall to find 100s of willing YDNA testers or atDNA test around your brick wall.
-
So glad I read this thread, you've saved me €90. I was going to do the DNA as some family members have already done but with so many different results it's obviously a great waste if money. Thank you
-
Personally it depends entirely on what you want from a test really.
If you want an accurate report of your ethnicity then yes a waste of time.
However, using the results (whatever the ethnicity) to link up with others who share your DNA and then possibly break down brick walls in your paper research is really useful and not a waste of time.
So far have been able to help two other families achieve a breakthrough in their research - now I am just hoping that my next match will achieve the same for me.
Pheno
-
I don't think these tests are worth taking. Mine came back virtually all British, even going back thousands of years. You only have to look at my dad to see he's not from British stock! So what's going on here?
Where is your father's ancestry from?
-
The supposed "ethnicity" is not the reason I'm interested in genetic genealogy. I think many of the "naysayers" do not understand genetic genealogy very well. I find the same naysayers post fairly frequently on this topic on here despite having no actual experience in the field. I'm not forcing them them to take a test. I don't understand why they are such misery guts about others choosing to pay for a genetic genealogy test with their own money.
Autosomal testing has already been very useful in my family and over time as more people test it will be even more so. It has confirmed who two of my gtgt-grandparents were - in each instance the name of their mother was incorrect on their death certificate. One had married when father's or parents' names weren't required, so her death certificate was the only document that could link it back. Clearly there was some sort of estrangement in the family and she wasn't mentioned at all in the death notice when her mother died but her other siblings were. She was born before official registration commenced in Scotland but is on one census before they emigrated.
The other's age was incorrect on the ship manifest when he emigrated, so that made finding his birth in England harder. I had a hunch I'd connected them to their births back in Britain and then DNA matches with a 3C and a 3C1R confirmed this.
For me, none of the success relied on ethnicity calculators, but it has helped a relative of a relative.
She found she had some non-European ancestry which is not from the man she thought was her father. She has many matches with people with that non-European ancestry.
-
The best ethnic methodology (if you have a pretty robust of pedigree chart of a 100 direct ancestors charted), use the surnames in your pedigree chart and go to io.surname to look up the geographic origins of these surnames - much more accurate than atDNA estimates. Over time however, atDNA will improve as the methodology is fine tuned and improved.
I have noticed that these boards have a lot of negative comments due to high expectations that DNA will reveal a lot pretty quickly. Just like genealogy, genetic genealogy takes an investment of time and funds to get there and there is luck involved as well. As usual, those with more means (those that won battles and had enough not to starve to death) were more prolific (M222, L226 and CTS4466) - while other lines have barely survived over the last 1,000 years where progress will be much slower due to smaller sample sizes of testers.
-
I have the same problem- but what I found was that, the more i tested with different companies ( free ones mostly) I started to get a bigger picture of my DNA.. i sort of got a mean of all the results together. my results summed up were
ancestry dna
almost all irish- french - tiny bit of Spanish scandinavian
23and me
almost all irish- french- tiny bit of ashkenazi jewish
family tree dna
irish eastern europe
livingdna
irish and 5 percent indian???
my heritage
Irish, scandinavian 7 percent west asia ???
DNALand
irish, scandinavian, Finnish, spanish, european jewish, india/ west asian
Then gedmatch managed to break these down even further... I sort of cross referenced all the tests, and researched the movement of people around these parts, to understand why i was getting such a different genetic makeup from all these websites.
I still don't really know though, hope the science gets better to the point where they are agreeing more. some sites say I'm up to 20 percent west asian- then others say completely european !! aaahhhhh
overall I think the DNA tests are worth doing- even if it is new science- as I didn't even know I was Irish so....
-
This is Roberta Estes recent Blog on DNA
https://dna-explained.com/2018/05/11/pass-the-dna-please/
Kim
-
A year on from making a couple of comments above on this thread, as I await the arrival of my test kit and having done lots of research, I now realise that these DNA tests are good for DNA matches but of little more than guidelines if not just entertainment as for ethnicity estimates. I do realise that these marketing strategies can be a bit misleading but I still look forward to hearing what my ethnicity estimates are. I anticipate 90% northern English.
Martin
-
Martin, I'm willing to bet you won't get 90% Northern England.
1. Depends who you tested with, I suppose, but Ancestry, Ftdna and MyHeritageDNA do not appear to have such a category.
2. You are highly unlikely to get what you think you should be with whichever company you test with, as shown by the many queries rootschat get about ethnicity.
3. You are highly unlikely to actually be 90% Northern English.
Regards Margaret
Correction. Ancestry do have a Northern English category. My ancestry is mainly Southern English and it does say that for me, but no percentage is given.
Regards Margaret
-
I will report back in 6-8 weeks!
Martin