Using mtDNA to ID bio parents; adoption

[armandthecorsair]

Here's the deal: My mother's father's mother, my great grandmother, was adopted. Her biological parents were entirely unknown to her and her adoptive parents. Until now my strategy has been to search the 1910 census for households with a child matching her name and age, and then follow that family to the 1920 census, hoping to find the same child absent (she was adopted mid-decade). No results. Now, obviously I can't test for this purpose, but my great grandmother had several daughters, and one of their daughters (my 1st cousin 1x removed) has agreed to provide a sample. My goal is to match with someone whose tree contains a matrilineal surname close in time and place, and then follow the branches to evaluate the probables. E.g. a common descendant of my 3rd great grandmother, ID her daughters genealogically, and find which of those lived in same or nearby counties, may have died in childbirth, etc. Would the level of testing provided by mtDNAPlus - or any test for that matter - enable me to do this? Any obvious flaws in my proposed methodology? Helpful hints or personal experience with this type of situation?

[tomcat]

I suggest you choose the Geno 2.0 rather than FTDNA's mtDNA Plus. Geno 2.0 samples 3,300 mtDNA SNP's that will take you further in haplogroup definition than the HVR SNP's sampled by Plus. Geno will also give you 130,000 autosomal and X chromosome SNP's that includes a set of ancestry informative markers, the opportunity to download your data for aftermarket analyses, and the chance of contacting other Geno test takers in what promises to be one of largest and most wide ranging databases of any provider.

The more expensive option is to do both mtDNA FMS and Family Finder that will give you complete coverage of mtDNA, wider, if less selective, coverage of the autosomes and X, data download and access to a smaller, but growing, matching database.

If you, or another member of your immediate family also does Geno or Family Finder your or your family member matches to the first cousin once removed will ID segments than descend from that line.

[MMaddi]

My advice is to order the Family Finder test, not the mtDNA test or Geno 2.0.

In the case of mtDNA testing, the databases are not as large as those for yDNA testing. So, you would be relying on the hope that someone who is in the same maternal line as your adopted great-grandmother will have tested and be in the database to match. And even if you do find an mtDNA match, you have no assurance that the common ancestor didn't live several hundred or more years ago. The resolution for mtDNA matches is just not that great, at least compared to yDNA matches. Of course, if you did do the mtDNA test on the appropriate relative, you might luck out and get a relatively rare haplogroup/subclade match with someone whose ancestor lived in the same town as your great-grandmother; that might point to a recent common maternal line ancestor. But I wouldn't count on that sort of luck.

In the case of Geno 2.0, the components of the test are mainly oriented to deep ancestry. In the case of the yDNA and autosomal/x SNPs tested, they will tell you nothing about ancestry in the last few hundred years. In fact, since you're looking for a match with your great-grandmother's line, the yDNA SNPs are useless; she didn't have any yDNA. It's true that Geno 2.0 will test more mtDNA SNPs than the low level (HVR1/HVR2) tests at FTDNA. But you still have the problems I discussed in the previous paragraph.

Your best bet is the Family Finder test. This tests autosomal SNPs and can find cousins in the database, as long as they've tested, of course. If there are 1st or 2nd cousins in the database, they will almost certainly be found by Family Finder. In the case of 3rd cousins, most of them in the database will also be found by Family Finder. What you must do is test the oldest descendant of your great-grandmother that you can find. With each generation from your great-grandmother, long segments received from her (the basis for matching with cousins) will increasingly be broken up by recombination. So, for instance, you want to test any children or at least grandchildren of your great-grandmother who are willing to test. The only drawback for you with the Family Finder is that you won't know off the bat in which line the common ancestor can be found. Any matches may be in lines other than those of your great-grandmother. It's still your best bet. You can certainly possibly use geographical locations of your matches' ancestors to match up with where your great-grandmother lived. That may narrow down in which line you're matching.

[tomcat]

Here's the deal: My mother's father's mother, my great grandmother, was adopted.

If your mother is still alive she could take either the 23&Me, Geno 2.0 or AncestryDNA test as one half of her X inheritance came from her father, who inherited his single X from his mother, your great grandmother, and carried that X unchanged. In other words, one of your mother's X chromosomes is a direct inheritance from her paternal grandmother.

Matches to such an X would be more genealogically informative than any partial, or even full match, to mtDNA.

Problem is, although Family Finder samples X it doesn't offer matching to X. Only the three labs noted above offer matching to X.

[tomcat]

And if your mother has a USofA pedigree, AncestryDNA would be the better bet than 23&Me or Geno2.0.

[MMaddi]

If your mother is still alive she could take either the 23&Me, Geno 2.0 or AncestryDNA test as one half of her X inheritance came from her father, who inherited his single X from his mother, your great grandmother, and carried that X unchanged. In other words, one of your mother's X chromosomes is a direct inheritance from her paternal grandmother.

Matches to such an X would be more genealogically informative than any partial, or even full match, to mtDNA.

Problem is, although Family Finder samples X it doesn't offer matching to X. Only the three labs noted above offer matching to X.

The x SNPs tested by Geno 2.0 are only ancestry informative markers, which can't be used to find someone with whom you share a recent common ancestor. They're solely useful for admixture analysis.

But you do have a point with testing at 23andMe, since they do include shared segments on the x to match customers. Plus, their database is significantly larger than the Family Finder database. And they also have all the other advantages of Family Finder that I mentioned above.

As for Ancestry's autosomal test, I wouldn't recommend it until they allow raw data downloads (to compare results at gedmatch.com) and some sort of chromosome browser, which allows you to make some sense of which matches may also match each other, which is useful information.

[tomcat]

As for Ancestry's autosomal test, I wouldn't recommend it until they allow raw data downloads (to compare results at gedmatch.com) and some sort of chromosome browser, which allows you to make some sense of which matches may also match each other, which is useful information.

They will allow downloads starting next year. They don't have a chromosome browser, yet. What they do have, is thousands of genealogically-interested users, many with USofA pedigrees, and a fully public matching regime like Family Finder but unlike 23&Me. The downsides are that they are just getting into autosomal testing and they are a subscription service but a subscription comes with access to all their online records.

[tomcat]

Here's the deal: My mother's father's mother, my great grandmother, was adopted. Her biological parents were entirely unknown to her and her adoptive parents.

Have you tried to get her original birth record? Getting info depends on the laws of the state in which she was adopted. State law might allow anything from full disclosure, to disclosure of non-identifying information to no disclosure at all.

[MFWare]

If your mother is still alive she could take either the 23&Me, Geno 2.0 or AncestryDNA test as one half of her X inheritance came from her father, who inherited his single X from his mother, your great grandmother, and carried that X unchanged. In other words, one of your mother's X chromosomes is a direct inheritance from her paternal grandmother.

Matches to such an X would be more genealogically informative than any partial, or even full match, to mtDNA.

Problem is, although Family Finder samples X it doesn't offer matching to X. Only the three labs noted above offer matching to X.

I agree that the OP should test with 23andMe. However, he should not put all of his eggs into the X-chromosome bag. FT-DNA Family Finder reports a lot of matching chromosome segments in the 2-3 cM range. 23andMe reports only down to 7 cM or so. It ignores the tiny fragments that are so popular with some FT-DNA customers. Its rule of thumb is that a match of two segments of 10 cM or more is a certain match. A match of one segment that is 10 cM or more is a probably match, but it is not certain. Most matches will be on only one segment. If the OP matches on his X, then chances are that this is his only matching segment with 23andMe. Although a match on a man's X-chromosome points to his maternal side, he has only 1 in 23 chances of a maternal match appearing there--and a 1 in 46 chance of any particular match on the X because none of his paternal matches will appear on the X.

Whether the OP tests with FT-DNA Family Finder or 23andMe, he can upload his X-Chromosome data to GEDmatch.com for additional inspection. However, however GEDmatch.com compares the X-Chromosome only to other specific X-Chromosomes and not to the entire database. Therefore, the OP must first identify candidates for X-matches and separately compare each to his own.

AncestryDNA does not provide access to raw data. It also makes no mention of X-Chromosome matching. GEDmatch.com is currently not prepared to accept Ancestry.com DNA data of any kind. Therefore, I recommend that the OP not test with AncestryDNA for the time being.

[tomcat]

From AncestryDNA FAQ:

"The AncestryDNA test uses microarray-based autosomal DNA testing, which surveys a person’s entire genome at over 700,000 locations ... "

"The test analyzes your autosomal DNA, which includes the entire genome—all 23 chromosomes— ... "

"If you and your DNA match both have family trees linked to DNA results, the pins on the map indicate the birth locations of the ancestors in your tree (up to about 10 generations of direct line ancestors), the birth locations of ancestors in your DNA match's tree, and overlapping birth locations that appear in both family trees. Some pins have numbers to indicate multiple ancestors within a specific region/city/county. ... "

More information:
link removed

[tomcat]

If your mother is still alive she could take either the 23&Me, Geno 2.0 or AncestryDNA test as one half of her X inheritance came from her father, who inherited his single X from his mother, your great grandmother, and carried that X unchanged. In other words, one of your mother's X chromosomes is a direct inheritance from her paternal grandmother.

The X your mother received from her father, that he received and carried unrecombined from his mother, is a recombined product of the two X's of that mother, one of which came from her mother and the other from her father. As such, one of your mother's X's offers a read on the ancestry/identity of your target great grandmother and her parents.

The mtDNA of your great grandmother, available from the first cousin once removed, will only give a read on the target great grandmother's maternal line.

If your mother has a full sister available to test, a comparison of their X results will immediately disclose the X that came from their father (and through him from his mother) as the paternal X of the two sisters will be identical.

[armandthecorsair]

Thanks for your input, everyone. I think I'll have my cousin take the Family Finder, then upgrade to Full Sequence mt if there's a persisting need.

tomcat, if I understood you correctly, 23&me has a matching feature to compare my mother and aunt's x chromosomes. Is the theory that their father's genetic contribution will be consitent with respect to each daughter, but their mother's will show more mutation/variance, and the discrepancy will allow me to parse the X? i.e. what they get from their father they'll have in common.

[tomcat]

tomcat, if I understood you correctly, 23&me has a matching feature to compare my mother and aunt's x chromosomes. Is the theory that their father's genetic contribution will be consitent with respect to each daughter, but their mother's will show more mutation/variance, and the discrepancy will allow me to parse the X? i.e. what they get from their father they'll have in common.

Yes, the paternal X will be identical in both sisters whereas the maternal X will be different, to some degree, in each sister.

AncestryDNA, Geno2.0 and 23&Me all do X matching. FTDNA does not do X matching but does allow the importation of 23&Me data.

[tomcat]

If mother and aunt also took the FTDNA test you might find useful comparisons to your cousin whom you plan to test with FTDNA.