[Update: On March 2, 2017, Roberta Estes updated her Native American Haplogroup blogpost and eliminated all references to M haplogroups—years after I informed of her mistake. I was glad to see that finally.]
This was not the first post that I wanted to write on Madagascar, the land of my ancestors, but I felt it necessary to do so. In the future, I will be writing about my Malagasy ancestors and how they ended up in colonial NY and NJ.
(Just a reminder, there are hyperlinks wherever you see RED highlighted text.)
In early October, I attended The Genealogy Event in NYC that featured a lot of well-known genetic genealogists, including CeCe Moore. In her talk about what goes on behind the scenes of PBS’s Finding Your Roots, she discussed Ben Jealous’s Malagasy mtDNA and how slave ships directly imported Malagasy slaves into VA. I immediately, and proudly, told her that I, too, was a descendant of Malagasy slaves directly imported into NYC/NJ in the late 1600s-early 1700s. Ever since my cousin Andrea, a direct matrilineal descendant of our shared 2nd great-grandmother, found out her mtDNA was M23, the two of us have researched everything Madagascar. Surely, we both felt the call of our ancestors. Basically, in finding our mtDNA M23 ancestors, we felt our ancestors calling out to us—-telling us to speak for them, urging us to tell the world about how they arrived in NYC as slaves, under what conditions they lived and labored in NY/NJ, etc. In all of my blog posts, I have tried to do my best to appease our ancestors. How can we not listen to them? So, when CeCe asked me to be the Co-Administrator of FTDNA’s new Malagasy Roots Project, I happily accepted. My mama didn’t raise no fool. Besides, I firmly believe that my ancestors would be a little annoyed with me if I hadn’t accepted the position. And we can’t let that happen. No, we can’t.
As a descendant of Madagascar slaves brought to this country, I am particularly disturbed to see M23, a haplogroup found only in Madagascar, be placed under the rubric of “New Native American Mitochondrial Haplogroups” by Roberta Estes, a person who is well-known in the field of genetic genealogy. In no way, shape, or form, do I want people to be misled into thinking that this haplogroup has anything to do with it being a Native American one. Her hypothesis goes against current literature on M23. As a result of several of her recent blog posts, I have included references to the Malagasy origins of M23 at the end of this blog post.
On September 18, 2014, Roberta posted “Native American Mitochondrial Haplogroups” on her blog, DNAeXplained-Genetic Genealogy. These known Native American founder haplogroups were A,B,C,D, and X. I had no problem with her designation of these haplogroups as being Native American ones, as there is enough literature to back up her claim and I was already aware of those Native American haplogroups. To be honest, I only read the beginning of her blog post back on Sept. 18th which didn’t mention haplogroup M.
Even the haplogroup diagram, at the beginning of her blog post, made no mention of haplogroup M.
It was only further down her blog post, when she listed all the Native American haplgroups alphabetically, that I now see mention of the M haplogroup.
Another one of Roberta’s blog post on September 24,2014 caught my eye with the title “New Native Mitochondrial DNA Haplogroups Extrapolated from Anzick Match Results.” The methodology she seems to have used was to take the autosomal matches of living people from Gedmatch’s One-to-Many list who matched the ancient DNA of Clovis Anzick, look at the mtDNA of those matches, and then automatically deem them Native American.
Roberta wrote:
“Given that, and given the autosomal ethnicity analysis of several individuals, and given that mitochondrial haplogroups A, B, C, and D are not known to be routinely found in the European population, I decided to extract all of the associated mitochondrial DNA haplogroups. Furthermore, parts of haplogroup X are known to be Native, and haplogroup M, which is quite rare, has long been suspected, but unproven.
In some cases, looking at the Anzick matches, we know that because of the very high level of Native heritage, the individual is either not admixed or only very slightly admixed. In other words, it makes perfect sense that their mitochondrial DNA is indeed Native as well as their Y haplogroup. At nearly 100% Native, both of those lines would have to be Native.”
In the same blog post, she continues:
We found repeated instances of many mitochondrial haplogroups not previously identified as Native. In fact, with the exception of a couple subgroups of the M and X haplogroups, all of the Native haplogroups were found repeatedly in these samples.
A big pause is needed here. A very BIG pause. I am at a complete loss as to WHY she went to great lengths to extrapolate so much from an ancient sample and compare it to living people. I am aware of the article “Mitochondrial Haplogroup M Discovery in Prehistoric North Americans”, but even the authors of that article state:
“The discovery of haplogroup M in the Americas is consistent with the hypothesis of a single colonization for the Americas since this haplogroup is found in Southern Siberia, the presumed homeland of the ancestors of North Americans (Bonatto and Salzano, 1997; Meriwether et. Al., 1995a). However, it also demonstrates the limitations of using genetic data solely from contemporary populations to infer the events and early population history of the Americas. Using genetic data from contemporary populations to infer early prehistoric demographic events is even less accurate when the population history has been variable over time….Therefore data based on living Northwestern North America might bias interpretations of population prehistory in the Americas (p. 646-647).”
Second, it doesn’t necessarily follow that because someone matches Clovis Anzick autosomally that their mtDNA is a Native American given. For example, I have mtDNA H1ag1 which is a European mtDNA, however, when the first Clovis Anzick matches came in, I matched Clovis Anzick at just over 7 cMs. Likewise, someone with mtDNA M23, like anyone of my 5 DNA tested M23 cousins, could have Native American ancestry from a completely different source other than their mtDNA. Her hypothesis just doesn’t add up. Besides, Roberta herself mentions over and over again that haplogroup M has not been proven to be Native American. In fact, there is also a great body of research about the East Asian to East African geographical distribution of haplogroup M.
The same day that Roberta published her blog post on “Native American Mitochondrial Haplogroups,” she also crossed posted it in the Facebook Group, Native American Ancestry Explorer:DNA, Genetics, Genealogy, and Anthropology. I immediately posted a comment indicated that I thought M23 was only found in Madagascar and I asked her if M23 was now associated with being Native American. I must admit I was a little taken back because her inclusion of M23 as a Native American haplogroup went against everything I have read about M23—-that M23 is only found in Madagascar.
Her response back to me was:
What I gleaned from her response was that she included haplogroup M on her list because M was found in a Native burial BEFORE Full Sequencing of that mtDNA. I still am not sure if she was referring to haplogroup M in general, or M23 in particular, but anyone who has taken a Full Sequence mtDNA test knows that this test is the most definitive test regarding a person’s mtDNA. Again, how can you include haplogroup M on a Native American Mitochondrial Haplogroup list if the one sample referred to has not been fully sequenced? What if the sample was a mistake or was related to a different subgroup? Roberta herself states that she spoke to a scientist who would have loved to have more full-sequencing and more advanced haplogroup designations. At the same time, she also states that haplogroup M is “waiting in the wings” for more confirmation that it is a Native American haplogroup????
Roberta then asked me if my mtDNA ancestors had Native American ancestry. As you can see, I clearly pointed out that my M23 ancestors were “mulatto”, a classification that also included Native Americans. However, I thought I was clear in differentiating between my M23 Madagascar ancestry and the fact that my family also has Native American ancestry that comes from a different source. As you can see, her response back to me was just a ” You know, it can never be easy, can it 🙂 Thanks.” I decided to let the matter rest a few months ago. I just discussed her position among friends and let it go. In retrospect, I should have been more adamant in questioning her. I just didn’t hear my ancestors calling out to me then. Not hearing them was a big mistake on my part!
Early this past Sunday, December 7th, when I logged onto FB and checked the Native American Ancestry group posts, I then noticed Roberta had updated her “Native American Mitochondrial Haplogroup” list and I immediately felt déjàvu. But, this time, I heard my ancestors calling out to me LOUD and CLEAR to set the record straight. So, I immediately responded back to her.
As you can see, I was more to the point and asked her directly if she was saying that M23 was not a Malagasy haplogroup, but was a Native American one. I even attached a well-known, accepted, and peer reviewed article indicating the Madagascar origins of M23. Up until that day, she only listed her own blog post as a reference for M23. My response was followed by TL Dixon asking her more pointed questions, as he had also done last September, not only about M23, but also about other haplogroup subclades also found in Madagascar, like B41a1a.
In addition, later on Sunday, I started reaching out to genetic genealogists like CeCe Moore and Claudio Bravi, who has been analyzing Native American haplogroups since 1993, as well as James Lick, asking them about the origins of M23. They all agreed that M23 was only found in Madagascar, a fact I already knew. Somehow, I wanted a confirmation from others before I wrote this blog post.
On Tuesday, December 9th, I again responded to Roberta’s post in the Native American Ancestry FB Group. This time I also cut and pasted my response to her blog. Roberta did respond to my post on her blog:
After reading her response, I went back to her blog and re-read it. I also started reading the responses to her “Native American Mitochondrial Haplogroup” posting. I was happy to see that on Monday, December 8th, Angie Bush, a well-known molecular genealogist, also stated that M23 had a Madagascar origin and she also posted the link to the same article I had made reference to a day earlier in the Native American Ancestry Explorer FB Group.
Angie wrote:
Her response to Angie was more detailed:
Roberta finally linked the article on M23 having Madagascar origins after Angie referenced it to her. She now indicated M23 as being a “Madagascar Motif” when it is in fact the Madagascar haplogroup unquestionably. Angie also let Roberta know about the FTDNA Malagasy Roots Project as well. That being said, I still find it highly problematic that Roberta still links her “Anzick Provisional Extract”, along with the peer reviewed article that Angie and I both referenced to her, to the M23 haplogroup on her “Native American Mitochondrial Haplogroup” list.
In conclusion, I am left with the following unanswered questions:
1) How does one arbitrarily decide to designate mtDNA haplogroups as Native American based on autosomal DNA comparisons to an ancient DNA sample—with some comparisons at very small segments?
2) How does one initially ignore a body of literature about the Madagascar origins of M23 and, after finally acknowledging its origins, still decide to link it to being a “potential” Native American haplogroup?
3) Why insist on repeatedly stating that haplgroup M isn’t proven to be a Native American haplogroup, but still link certain subclade M haplogroups to them being “possible” Native American haplogroups?
4) How does one attempt to publish a hypothesis on “New Native American Mitochondrial Haplogroups” without the hypothesis being analytically challenged and peer reviewed?
As a genealogy/DNA blogger and speaker and, as someone who is also tri-racial, my obligation is to correct the misinformation out there, pinpoint inaccurate statements automatically assumed to be facts, and elucidate the flawed analyses/methodologies that I come across as they relate to my own genealogy/family research. I want information out in the public realm that is reliable as it is true. I don’t know the answers to these questions. But, what I do know is that the M23 haplogroup is not a “Native American Mitochondrial Haplogroup.” My M23 mtDNA ancestors called me out and told me so. So, I am now telling the world.
References to M23 being a Madagascar Haplogroup:
http://www.biomedcentral.com/1471-2164/10/605
http://www.webmedcentral.com/article_view/2237
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1199379/
http://www.nature.com/ejhg/journal/v21/n12/full/ejhg201351a.html
http://www.africatoaotearoa.otago.ac.nz/haplogroups/9-mtdna/14-mtdna-m
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0080932
If you have any of the Malagasy mtDNA or Y-DNA haplgroups below, please consider taking a FTDNA Full Sequence mtDNA test or a Y-37 DNA test and then join the Malagasy Roots Project. Please click on the title link below for more details.
Teresa – thank you for writing about this issue. I am not a Native American expert by any stretch of the imagination. When dealing with anthropology and population genetics, I think that it is very wise for genetic genealogists to rely on peer-reviewed scientific literature. I also believe that more of us should take the time to read this literature and understand/interpret it properly. Thank you again!
Good post, Teresa! Thanks for this clarification. Since there are a number of scholarly publications on this subject, I don’t understand why someone would think otherwise and re-publish different information.
Teresa, my tests have revealed no discernible NA, yet my haplo M32c, exclusive to Madagascar. My brothers test R1b1 (European). What is the reason Roberta decided to take this on? Is there a large enough sample of Malagasys in Madagascar to debunk her argument? I did Jack’s Utility on my FTDNA and 23andme tests and they both show I have defining markers for Madagascar and Indonesia. I don’t get why she wants to take this with no research to back up her assertions?
Very thorough coverage, Teresa. Thank you for clarifying the origins of M23.
Awesome job Teresa!
Informative post, thanks for writing it!
That was EXCELLENT! I solicited her (Roberta) feedback about two or three weeks ago regarding the plethora of literature i’ve been reading that halpogroup E1b1a (being African) and E1b1b (being labeled as Middle Eastern or Near Eastern at the exclusion of African) though both descended from the same branch E1b1 – no answer.
I can help in that area. I’ve conducted my own research due to non-answers from “experts”
https://druidikal.wordpress.com/2014/12/08/dnadventures-premiere-partie/
https://druidikal.wordpress.com/2013/05/06/new-subclade-found-in-haplogroup-e/
Roberta Estes is a fraud and has been caught lying and skewing results many times. Her blog says, “Roberta Estes has been a professional scientist and business owner for 25+ years, (BS Computer Science, MBA, graduate work in Geographic Information Systems), as well as an obsessed genealogist since 1978. ”
Nowhere does it state the name of the educational institutions she attended.
Thanks for sharing this excellent blog, Teresa Vega. I definitely think there needs to be some clarity in terms of Malagasy-specific haplogroup assignments, none of which have been proved to be Native American. Here I will note that B4a1a1a (Malagasy) and B2 (Native American) are both subgroups of B4. However B4a1a1a and B2 are exclusive to its respective populations. [see http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2987306/%5D. I do think the Malagasy population is fairly new to the ethnic admixture world and we’re learning it could much more common than we expect. Malagasy-specific haplogroups are much younger than it Polynesian-motif predecessor, B4a1a1. With the latter there is some controversy raised by this Polynesian-motif haplogroup found in two of 14 skeletons of a now-extinct Brazilian tribe. It is unclear if the haplogroups are from an ancient Polynesian or more recent Malagasy origin because full mtDNA sequences were not performed in the analysis to get the best resolution. One explanation could be that before the Brazilian tribe became extinct, they mixed with enslaved persons with some sort of Austronesian/Indonesian, Malagasy, or Polynesian ancestry when the Portuguese transported slaves from the Indian Ocean islands and East Africa during early part of Trans-Atlantic slave trade. They arrived before the Brazilian tribe died out. Yet, the jury is still out…. As more people from the Americas test we are seeing more and more Southeast Asian, Austronesian, and Oceanian genetic and uniparental markers that can’t be explained as Native American in origin. Many people will certainly confuse Malagasy for highly contentious and sought-after Native American ancestry. LOL. However in order to properly identify possible Malagasy ancestry, we must know about the characteristics and mutations that make them unique. The information must be properly researched, identified and documented in accordance to rigorous scientific standards, inclusive of peer review. Equally important such information must be presented or reported to the public as accurately, intelligibly, and objectively as possible. …
Love this post Cousin Teresa! Thank you for bringing the truth to the light and continually informing us of the Malagasy contribution to African American genetics and genealogy. You know your stuff when it comes to the Malagasy aspect of DNA. We appreciate you always for your educating us!
Thanks so much Teresa for this post! I wasn’t sure what to make of the M23 appearing among the Anzick remains and appreciate your insights on Malagasy ancestry!
In the first place, I have never felt comfortable using the GEDMatch tools to compare with the ancient DNA profiles, so I’d be reluctant to draw any conclusions on that basis.
In the second place, I took a quick peek at the article you linked about the presence of M in Native Americans. It was published in 2007, and I suspect the findings should be revisited. The following details aren’t really relevant for your main interest of M23 specifically, but the diagnosis of M was based on an older technique called RFLP. It uses enzymes to probe for the presence or absence of a mutation at a specific location, and the China Lake samples tested positive with the probe used for haplogroup M. However, the HVR1 mutations of the China Lake specimens were at positions 16093, 16213, and 16223. In my personal database of GenBank full mitochondrial sequences, this suite of mutations is found in 15 haplogroup X2a1 (a Native American haplogroup), published after the article was written. The authors did not use a RFLP probe for haplogroup X, and it’s possible that the RFLP technique gave a false positive for haplogroup M for some reason that would come to light only with full sequencing.
In my view, this further weakens the association of M with Native American haplogroups.
Thank you, Ann, for your comment.
My apologies if this is a dumb question, but are you suggesting there is no way that there could be Madagascan ancestry involved in the evolution of native americans? That no ancient peoples ever left Madagascar and intermingled with those that ended up peopling the Americas?
David-
Roberta Estes compared living individuals to Clovis Anzick, an ancient boy that lived about 13,000 in Montana. One of those matches was a person, I assume, with a mtDNA of M23. M23 is only found in Madagascar and originated around 2,000 +/- years ago. It is impossible for M23 to have existed around the time that Clovis lived.
Now, if you look at the past 500 years ago, we know that Malagasy slaves were taken to the ?US, Caribbean, and South America where some mixed with the native peoples there. This is not ancient history though.
Teresa