Welcome to City-Data.com Forum!
U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Genealogy
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
Reply Start New Thread
 
Old 11-22-2013, 03:42 PM
 
Location: New Mexico
4,938 posts, read 7,299,736 times
Reputation: 8524

Advertisements

Random users are not included in the reference populations. Reference populations are a combination of previous genetic studies and 23andMe users who they believe with high confidence have a long history within a particular region. This would not include any Americans (for the European populations).

Also, what you self-report about your ethnicity will have no effect whatsoever on your Ancestry Composition. It is based on their reference samples and has nothing to do with your paper trail. There have been many reports of users finding out from Ancestry Composition that they are of a different ethnicity than what they thought they were (because of mistaken paternity, etc.).

If you don't like your "non-specific" ancestry, then blame your ancestors for moving around too much, don't blame 23andMe.
Reply With Quote Quick reply to this message

 
Old 11-22-2013, 08:38 PM
 
Location: Baltimore, MD
5,291 posts, read 5,958,411 times
Reputation: 10818
Quote:
Originally Posted by suzy_q2010 View Post
People who are participating are trying to find matches to help fill out their family trees. If you give fictional names for ancestors, that will not work. Other participants will waste time trying to contact you, thinking they have found someone with whom they share a common ancestor. You will have DNA matches with a fictional family tree. Please do not do that.

I have no intention of providing a family tree or surnames in the 2nd test.

It is not necessary to provide fictional information to remain anonymous. Just decline any requests to share info, though it is frustrating to find close cousins who do not want to share.

By anonymous, I meant anonymous to 23andMe.

The same with the health information. By giving fake info, you skew the results of matching certain genes to certain traits. I actually suspect that your account could be flagged if their were enough discrepancies between your genetic material and your fake medical history.

I understand health risks and would not indicate I have an illness that is caused by a single autosomal dominant or recessive gene. Do you honestly believe that 23andMe's association of certain genes to "certain traits" based on surveys submitted by their subscribers is statistically valid? Really?

If enough people did what you are thinking about doing, it would make the whole exercise worthless for all of us.

While the deep ancestry results are interesting, they really are not all that helpful if you are researching your family tree and you happen to be mainly European. The gene panels just cannot reliably separate people whose ancestors came from a relatively tight geographic area 500 years ago. It is not reasonable at this point to expect that. In fact, the ancestor you think came from England might actually have immigrated there from somewhere else (say Germany).

Or, more likely, it is because 23andMe fails to identify 93% of the French and German DNA.

I really think you need to rein in your expectations a bit about what deep ancestry means. The more people who participate and provide accurate information, the more detailed the breakdown can become in the future.

You're missing the point. I don't have high expectations regarding the ancestry composition - I was simply pointing out that in my case 23andMe's ancestry composition was worthless.

If you want to retest, send another kit. Just do not make up a fake family and a fake medical history to go with it.
I may retest, but my evil (fraternal) twin wants to try the test, too. BTW, my younger sister has tested, but no one except me (and the company, of course) can see her results. I don't question the validity of our health risks, although we share none. But the traits? When we shared our results, both of us were laughing so hard we had tears running down our cheeks. I wonder of there is a gene for that??
Reply With Quote Quick reply to this message
 
Old 11-22-2013, 08:52 PM
 
Location: Baltimore, MD
5,291 posts, read 5,958,411 times
Reputation: 10818
Quote:
Originally Posted by creepy View Post
"I went from 996 to 960. Why?" that is happening to me too. This is my basic understanding and more people test and the tests get better at discerning relations you will have more accurate stats.
Hey Creepy,

I found the answer on a Rootsweb list serve. One guy actually has around 3,000 matches. If you do not issue an invitation to share within a certain amount of time, 23andMe will drop the lowest matches. So, some of the posters immediately issue an invitation to all new matches in order to preserve every match on their list. Some folks with limited time have chosen to immediately send an invitation to their "weakest" matches, knowing that their highest matches will stay on their list indefinitely.
Reply With Quote Quick reply to this message
 
Old 11-23-2013, 10:55 AM
 
Location: Illinois
3,169 posts, read 5,139,745 times
Reputation: 5617
Quote:
Originally Posted by lenora View Post
95.5% European
77.9 % Northern European
9.5% British & Irish
0.1% Finnish
68.2% Nonspecific Northern European
0.0% Southern European
0.0% Eastern European
0.0% Ashkenazi
17.7% Nonspecific European

3.7% Sub-Saharan African
? West African
? East African
? Central & South African
3.7% Nonspecific Sub-Saharan African

0.8% Unassigned

Conservative estimate:
94.5% European
1.0% British & Irish
43% Nonspecific Northern European
50.5% Nonspecific European

3.5% Sub-Saharan
3.5% Nonspecific Sub-Saharan

2.0% Unassigned
_____________________________________________
Worthless. Frankly, even the Conservative Estimate, at 90% confidence raises issues. I'd love to see what my results would be with a confidence level of 95% or higher. I'm guessing my Nonspecific Northern European estimate would significantly decrease or entirely disappear.

"The default Standard Estimate corresponds to 75% confidence. You have the option to make the estimate more strict (the Conservative Estimate is at 90% confidence) or more lenient (the Speculative Estimate is at 50% confidence)."

I would have posted my Speculative Estimate, but a confidence level of 50% is equivalent to a coin toss. The Standard Estimate is not much better (statistically speaking). 23andme should stick to health risk analysis and family finder. Seriously.
If it will make you feel any better much of my Euro is non-specific. And honestly, when considering Dr. McDonald and Gedmatch's information I am expecting to see largely non-specific African and Asian once my new AC is issued. I believe it is due to my slave ancestors being from various African countries but largely east of West Africa (read somewhere in Central Africa), including some ancestry in East, North and South Africa. I am not hyping myself on the new breakdown. I'd be surprised if many of my questions are answered.

Last edited by CMichele; 11-23-2013 at 11:11 AM..
Reply With Quote Quick reply to this message
 
Old 11-23-2013, 11:19 AM
 
Location: Canada
7,653 posts, read 5,424,607 times
Reputation: 8791
My ancestry composition is vague as well:

66.8% - Nonspecific European
16.2% - Eastern European
9.2% - Nonspecific Northern European
0.7% - French & German
3.1% - Balkan
3.6% - Nonspecific Southern European
0.1% - East Asian
0.3% - Unassigned

My conclusion: My ancestors 500+ years ago roamed around Europe more than I thought.
Reply With Quote Quick reply to this message
 
Old 11-23-2013, 02:08 PM
 
Location: Baltimore, MD
5,291 posts, read 5,958,411 times
Reputation: 10818
Quote:
Originally Posted by CMichele View Post
If it will make you feel any better much of my Euro is non-specific. And honestly, when considering Dr. McDonald and Gedmatch's information I am expecting to see largely non-specific African and Asian once my new AC is issued. I believe it is due to my slave ancestors being from various African countries but largely east of West Africa (read somewhere in Central Africa), including some ancestry in East, North and South Africa. I am not hyping myself on the new breakdown. I'd be surprised if many of my questions are answered.
Did your 23andMe results pick up the regions in Africa? According to FTDNA, I am 95.45% +/- 0.11% Orcadian and 4.55% +/- 0.11% African. For me, the bonus was the identification of both the east and west coasts of Africa, with Yoruba, Bantu Kenya, and Mandenka identified as the subregions. I also found it refreshing that FTDA provided a margin of error for the West Europe (Orcadian) and African components.

Interestingly, Geno 2.0, known for its particularly deepclad analysis, identified German, Iberian and Sub-Saharan. For my father, they identified British and German. Of course, their reference populations are very different from that of 23andMe's.

Best of luck
Reply With Quote Quick reply to this message
 
Old 11-23-2013, 03:05 PM
 
Location: Illinois
3,169 posts, read 5,139,745 times
Reputation: 5617
Quote:
Originally Posted by lenora View Post
Did your 23andMe results pick up the regions in Africa? According to FTDNA, I am 95.45% +/- 0.11% Orcadian and 4.55% +/- 0.11% African. For me, the bonus was the identification of both the east and west coasts of Africa, with Yoruba, Bantu Kenya, and Mandenka identified as the subregions. I also found it refreshing that FTDA provided a margin of error for the West Europe (Orcadian) and African components.

Interestingly, Geno 2.0, known for its particularly deepclad analysis, identified German, Iberian and Sub-Saharan. For my father, they identified British and German. Of course, their reference populations are very different from that of 23andMe's.

Best of luck
No, not yet. This upgrade is what prompted this thread. They made new changes to AC and the new customers are seeing the new AC first. I am still waiting. Right now I am just SSA and a bit of North African.

Once they recalculate my results I should see West, East, and South African. According to Gedmatch my South and East African is just over 12% and 5% respectively.
Reply With Quote Quick reply to this message
 
Old 11-24-2013, 11:27 AM
 
Location: Baltimore, MD
5,291 posts, read 5,958,411 times
Reputation: 10818
Quote:
Originally Posted by CMichele View Post
No, not yet. This upgrade is what prompted this thread. They made new changes to AC and the new customers are seeing the new AC first. I am still waiting. Right now I am just SSA and a bit of North African.

Once they recalculate my results I should see West, East, and South African. According to Gedmatch my South and East African is just over 12% and 5% respectively.
Ahh, I did not realize that nobody received a breakdown of the Sub-Saharan component. Thank you.
Reply With Quote Quick reply to this message
 
Old 11-24-2013, 11:47 AM
 
Location: Baltimore, MD
5,291 posts, read 5,958,411 times
Reputation: 10818
Well, I am not a happy camper. It appears that 23andMe is now in the process of "upgrading" to version 4. The newer version will not be compatible with FTDNA, Gedmatch, etc. I was going to have my elderly father spit for 23andMe, but am now reluctant to do so. The newer version probably is more valuable in terms of identifying medically significant alleles, but not so much ancestry. (My 87 year old father does not need to know his health risks.) It looks like the total SNPS tested will be around 600,000, about the same as Geno and FTDNA. I wonder how many of the approximate 600,000 SNPS will be ancestral SNPS.

23andMe's New Chip Improves Efficiency | The 23andMe Blog

OTOH, if there are additional health related SNPS that are of interest to me, I may submit a new sample for that alone.
Reply With Quote Quick reply to this message
 
Old 11-24-2013, 01:04 PM
 
Location: Illinois
3,169 posts, read 5,139,745 times
Reputation: 5617
Quote:
Originally Posted by lenora View Post
Well, I am not a happy camper. It appears that 23andMe is now in the process of "upgrading" to version 4. The newer version will not be compatible with FTDNA, Gedmatch, etc. I was going to have my elderly father spit for 23andMe, but am now reluctant to do so. The newer version probably is more valuable in terms of identifying medically significant alleles, but not so much ancestry. (My 87 year old father does not need to know his health risks.) It looks like the total SNPS tested will be around 600,000, about the same as Geno and FTDNA. I wonder how many of the approximate 600,000 SNPS will be ancestral SNPS.

23andMe's New Chip Improves Efficiency | The 23andMe Blog

OTOH, if there are additional health related SNPS that are of interest to me, I may submit a new sample for that alone.
If you opted to have them store your sample for further studies is it even necessary to spit again? I would think that you would simply get the new information.
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply
Please update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:


Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Genealogy
Similar Threads

All times are GMT -6.

© 2005-2024, Advameg, Inc. · Please obey Forum Rules · Terms of Use and Privacy Policy · Bug Bounty

City-Data.com - Contact Us - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 - Top