Locating African American haplogroups within Africa

Charting the ancestry of .. (Fig. 1b)

Charting the ancestry of .. (Table 2)

Charting the Ancestry of African Americans (Salas et al., 2005)

____________________

Charting the Ancestry of African Americans (Salas et al., 2005)

“Here, we make use of an African database of 4,860 mtDNAs, which include 948 mtDNA sequences from west-central Africa and a further 154 from the southwest, and compare these for the first time with a publicly available database of 1,148 African Americans from the United States that contains 1,053 mtDNAs of sub-Saharan ancestry. We show that >55% of the U.S. lineages have a West African ancestry, with <41% coming from west-central or southwestern Africa. These results are remarkably similar to the most up-todate analyses of the historical record.

____________________

***

In this blogpost i will be reviewing three scientific papers dealing with the haplogroups identified among African Americans and correlating their frequencies with possible African source regions. These studies can only represent the state of knowledge at the date of their publication naturally 😉 But they should be suggestive of the approximate regional distribution of African ancestry among African Americans.

– African-American mitochondrial DNAs often match mtDNAs found in multiple African ethnic groups (Ely et al., 2006)
Charting the Ancestry of African Americans (Salas et al., 2005)
– Y Chromosome Lineages in Men of West African Descent (Torres et al., 2012)

For a most recent study (2013) on African American mtDNA see: “Mitochondrial DNA diversity in the African American population“. I will however not discuss that paper as it makes no attempt to correlate American haplogroups with African regions of origin.

Before i proceed some words of caution as usual, perhaps redundant for those who already know but it’s worth repeating as many people still tend to be misinformed about the true meaning of their haplogroups.

  • Haplogroups (Y-DNA, mtDNA) only represent a very limited portion of your total ancestry and often date back from many thousands years ago. Autosomal DNA testing is the only way to find out about your complete ancestry.  See figures below for visualization although generally speaking you will have to go back more generations than just 3 or 4 to find out about your complete African ancestry.

***

mitochondrialdna2

***

dna-pedigree-y-mtdna

  • In fact you will have to trace back to your first African born ancestor carrying your haplogroup in a direct maternal or paternal line to determine the relative contribution of the haplogroup to your ancestry. For African Americans this could take you back at least 7 generations on average to the mid 1700’s when most Africans arrived in the US (see this chart), but in some cases it might also take you back right to the 1600’s! Going with a conservative estimate you could have 128 ancestors around 1750 but in the 1600’s this will have increased exponentially to 1024! (see this link  and also this one for a useful overview of the number of ancestors per generation). Of your dozens or even hundreds of African-born ancestors relocated to the US, your direct maternal/paternal ancestor would be only one, representing no more than 1% of your total ancestry, in most cases (see also “Fictional Family Tree incl. African Born Ancestors“).

***

***

Genealogical-Family-Treeaaa

Click to enlarge

  • A minority of African Americans get tested with non-African haplogroups, a further proof that only autosomal DNA can tell you about your complete ancestry. As the greater part of African Americans will obviously still be predominantly African depite having a non-African haplogroup. Theoretically speaking even a person who scores 99% African in a DNA test could end up with a non-African haplogroup! These haplogroups are usually testimony of minor European geneflow among the African American population. Interracial unions most often involved European American males as measured by European Y-DNA but a considerable % of African Americans also carry European mtDNA. This is unlike most other Afro-Diasporic populations, thereby constituting a distinct feature of AA genetics i suppose. According to several studies it’s in between 8-10% (e.g. Stefflova et al. 2009. or Deason et al. 2012).  Which might imply that generalizations about the nature of European admixture do not always apply as we all have individual family trees. Even if you personally do not have a European maternal haplogroup you could still have a European maternal ancestor through either your paternal family lines or those through your maternal grandfather, your maternal greatgrandfathers etc.

    Lind (fig.3 - Euro contribution)

    Taken from “Elevated male European and female African contributions to the genomes of African American individuals.” (Lind et al., 2007)

 

***

African-American mitochondrial DNA’s often match mtDNA’s found in multiple African ethnic groups

Link to online article

____________________

“Conclusion

[…] In addition, the finding that fewer than 10% of African-American mtDNAs matched mtDNA sequences from a single African region suggests that few African Americans might be able to trace their mtDNA lineages to a particular region of Africa, and even fewer will be able to trace their mtDNA to a single ethnic group. However, no firm conclusions should be made until a much larger database is available. It is clear, however, that when identical mtDNA haplotypes are shared among many ethnic groups from different parts of Africa, it is impossible to determine which single ethnic group was the source of a particular maternal ancestor based on the mtDNA sequence.” (Ely et al., 2006, p.4)

” Many African Americans are interested in learning more about their African roots and are willing to pay to have their mtDNA analyzed in the hope that it will match DNA from a particular African ethnic group. However, as more than half of the mtDNA sequences in the African database are identical to sequences from other ethnic groups, African-American mtDNAs will be much more likely to match sequences from multiple ethnic groups than sequences from a single ethnic group. When this result is coupled with the fact that 40% of African-American mtDNAs did not match any sequence in the database, it is clear that matches to a single African ethnic group will not be the outcome for most African Americans, and even when a match to a single ethnic group is obtained, multiple matches may occur in a larger database.

Furthermore, for the typical African American, the maternal ancestor who was the source of the mtDNA was just one of hundreds of enslaved African ancestors. In fact, it likely that there has been more mixing of African ethnic groups in the Americas than has ever occurred elsewhere. Thus, the ancestors of virtually all contemporary African Americans came from a large number of ethnic groups located throughout the region from Senegal to Angola.”  (Ely et al., 2006, p.9)

____________________

This study might have a somewhat sobering message for those who are emotionally invested in their haplogroup test results. However in the interest of correctly identifying one’s ancestors it should be required to be critical of DNA testing companies and to very carefully scrutinize their claims. Wholeheartedly celebrating a certain ancestral lineage only to be told later it was wrongly assigned or incomplete information was given can be very sobering as well afterall (see also “What Tribe Am I?“).

Basically what’s done in this study is that they compare an extensive database of “3725 mtDNA HVS-I sequences” with the mtDNA sequences of 2 African American samplegroups, one of them consisting of selfidentified Gullah’s. The outcome of this comparison can be seen below (it’s from their corrected paper version which appeared in 2007). For the Gullah samplegroup a perfect match with only 1 single ethnic group was found in 5 out of 74 cases and for the African American sample group it was 4 out of 97, implying that in about 95% of all cases no perfect single ethnic match was found! Rather a matching with multiple ethnic groups, sometimes also being from multiple regions, was a more typical finding. Undoubtedly this outcome is a reflection of widespread ancient population migrations across Africa and the continuous geneflow in between ethnic groups which themselves are also evolving all the time through ethnogenesis.

***

AA mtDNA often match... (Table 3, Number of perfect matches)

(Ely et al., 2007)

____________________

Because only a small fraction of the sub-Saharan African ethnic groups have been sampled, and there are parts of sub-Saharan Africa that are poorly represented in our database (Figure 2), our database cannot be considered a representative subset of the sub-Saharan mtDNA gene pool. Nevertheless, it is clear that a much larger database is needed since 40% of the African-American samples analyzed have no exact match in our database.” (Ely et al., 2006, p.9)

____________________

Above quote points out a major shortcoming of the study which concerns the coverage of their African database. Judging from their overview within the study (Figure 2, Table 7) it seems that especially crucial samples from Ghana and the Igbo were missing. The authors freely acknowledge this inadequacy themselves btw, regrettably the same can not always be said for commercial DNA testing companies…

Some other findings:

____________________

Only one Gullah/Geechee mtDNA and one AFDIL mtDNA matched mtDNAs that are found exclusively in eastern Africa in the sub-Saharan database. This distribution of matches is consistent with the historical information that most North American slaves were originally from western Africa.” [..]

“Surprisingly, five AFDIL mtDNAs matched only mtDNAs from the two Angolan samples that make up 4% of the database. This result is consistent with historical records indicating that a large proportion of the enslaved Africans brought to the Americas came from the West-Central African region of Angola/Congo region, and suggests that ethnic groups in this region of Africa need to be sampled more extensively.” (Ely et al., 2006, p.7)

____________________

***

Charting the Ancestry of African Americans

Link to online article

Charting the ancestry of .. (Fig. 1)

Click to enlarge

____________________

“Our analysis (table 2) indicated, as before, that North Africa, East Africa, southern Africa, and even southeastern Africa (Mozambique) show no significant contribution to the African American mtDNA pool in the United States.” (Salas et al., 2005, p.677)

____________________

As seen on top of this blogpost this study’s main findings are that about 55% of African American mtDNA could be placed in West Africa and about 41% in West Central & Southwest Africa (Angola). This outcome is not very surprising if we compare it with documented slave trade records. Although the proportion of West Africa seems subdued, and the Central African share (incl. Angola) inflated. A splitup of about 75/25 could have been expected based on this graph below. But perhaps it’s because the West Central African haplogroups (mostly from Cameroon, see this link for full overview) fall in between.

However like the authors mention themselves, mtDNA only tells part of the story, even when used in population studies.

____________________

“Of course, these results indicate only the maternal component of the American gene pool. Further research on Y-chromosome and autosomal markers is clearly needed to test whether the contribution of females and males to the African American gene pool has been symmetrical or asymmetrical—research for which, again, large African databases of such marker systems will be needed, such as are now becoming available for mtDNA. With the acceleration of data acquisition now under way, these analyses will no doubt be possible in the near future.” (Salas et al., 2005, pp.678-679)

____________________

Similar to the first study reviewed the authors caution against overestimating the predictive scope of mtDNA in individual DNA testing as opposed to population studies:

____________________

Considerable caution is therefore warranted when dealing with claims in the popular media (such as the acclaimed and prestigious BBC television documentary Motherland: A Genetic Journey, first shown in the United Kingdom in 2003) and those made by genetic ancestry–testing companies about their ability to trace the ancestry of certain American (or, for that matter, European mtDNAs to a particular locale or population within modern-day Africa. Our analyses stand as a warning to unsuspecting members of the public who may be seduced by such promises” (Salas et al., 2005, p.679)

____________________

***

Y Chromosome Lineages in Men of West African Descent

Link to online article

Y Chromosome Lineages (map)

____________________

Abstract: Rice production was a key industry of many colonial South Carolina low country plantations. Accordingly, rice plantations owners within South Carolina often requested enslaved Africans from the so-called ‘‘Grain Coast’’ of western Africa (Senegal to Sierra Leone). Studies on the African origins of the enslaved within other regions of the Americas have been limited. To address the issue of origins of people of African descent within the Americas and understand more about the genetic heterogeneity present within Africa and the African Diaspora, we typed Y chromosome specific markers in 1,319 men consisting of 508 west and central Africans (from 12 populations), 188 Caribbeans (from 2 islands), 532 African Americans (AAs from Washington, DC and Columbia, SC), and 91 European Americans. Principal component and admixture analyses provide support for significant Grain Coast ancestry among African American men in South Carolina. AA men from DC and the Caribbean showed a closer affinity to populations from the Bight of Biafra.

____________________

Even given the inherent limitations of haplogroup assignment these results seem to make historical sense although the analysis perhaps was overtly focussed on establishing an exclusive link between South Carolina with the Grain Coast a.k.a. Upper Guinea. It’s mostly in line with my previous discussion of the Gullah language and South Carolina runaway slaves. There is in fact another study dealing specifically with the maternal haplogroups of the Gullah, but it’s not available online so i couldn’t read it entirely. Its main conclusion seems to be again in support however:

____________________

“Genetic distance estimates indicated a close relationship of Gullahs and Jamaicans with Sierra Leoneans, while African Americans living in Charleston and the West Coast were progressively more distantly related to the Sierra Leoneans “

____________________

For the full abstract of that study see this link: “Mitochondrial DNA (mtDNA) haplotypes reveal maternal population genetic affinities of Sea Island Gullah-speaking African Americans“. 

To return to the Y-DNA study these were the population samples they used (also to be seen in the map above), it seems that especially crucial samples from Ghana & Congo/Angola were missing.

____________________

“population samples: Afro-Caribbean subjects were
recruited from St. Thomas (N =113) and Kingston, Jamaica
(N = 75). African Americans were recruited from Washington, DC
(N = 106) and Columbia, S.C. (N = 426). African American and
Caribbean subjects were recruited for various cancer genetic
studies [36,37,38,39]. West African samples (N =508), included 48
Bini from Edo State, Nigeria; 64 Ibo from Enugu State, Nigeria;
18 Hausa from Jos, Nigeria, 49 Yoruba from Ibadan, Nigeria; 61
Yoruba from Lagos, Nigeria; 28 Urhobo and 21 Itsekiri from the
Warri Delta region of Nigeria; 34 Kru from Freetown, Liberia; 47
Mende and 34 Temne from Freetown, Sierra Leone [38]; 19
Mandinka from Dakar, Senegal; and 85 Bamileke from Yaounde,
Cameroon”

____________________

In below graph their main finding of the South Carolina samples (AASC) being distinct from the other Afro-Diasporic sample groups (JM, ST & AADC) is clearly shown by the way it’s most closely clustering with the Mandenka (MAN) samples from Senegambia.

Y Chromosome Lineages (graph)

Figure 2. Plot of the first two principal components of a Y chromosome genetic distance matrix estimated for 17 populations.

***

A final overview of their findings is fascinating but also a bit puzzling. The African American samples from South Carolina (AASC) are being shown as having shared Y haplotypes with all the African sample groups, the highest number being shared with Nigerians (Yoruba’s, Igbo & Bini), the Upper Guinean groups coming in second place. However the other African Americans from Washington DC (AADC) and the West Indians (JM & ST) hardly share any Y-haplotypes, except amongst each other. Perhaps it’s because of the absence of Ghanaian samples but it’s also peculiar that only a few shared matches are found with the Igbo (IBO) samples while otherwise it’s mostly the Bamileke (BAM) from Cameroon or European Americans (EA). As a concluding observation it’s perhaps telling that the African samplegroups themselves have plentiful matches between each other; the Mandenka are sharing several Y haplotypes with the Yoruba for example.

Y Chromosome Lineages Table - shared haplotypes

Table 4. Matrix of shared Y chromosome haplotypes and RST genetic distances

Advertenties

6 gedachten over “Locating African American haplogroups within Africa

  1. My maternal haplogroup is a subclade of L2c – L2c2a (exactly assigned from NG, FTDNA, and 23). I have moderate matches at HVR1 level, few at HRV1/HRV2 level, and none at full sequence (via FTDNA “Ancestral Origins”). Interestingly, at the HVR1 level, I have matches from Spain and Portugal. Senegal, Ghana and Kenya are included as well. However, at the HVR1/2, my exclusive African match (out of 3 matches) is from Senegal. Studies have shown that L2c occurs in high frequency in the Senegal-Gambia region. This may possibly hint to my most recent African maternal origin, the operative word being “possibly”. My YDNA Haplogroup is E-U174 (E1b1a7a), a subclade of E-M2 (E1b1a). An exact match with 12 markers is from Ghana (Ga). Twelve markers with 1 genetic distance, I have several; but for Ghana, again, I have 3 matches (2 Ewe and 1 Fante). Although this is a low resolution test (12 markers), the exact match from Ghana (Ga ethnic group) in relationship to the three 1 genetic distance matches from Ghana may hint to a YDNA with, at least, a genetic profile similar to the Ga-Ewe-Fante cluster. This is by far inconclusive; yet considering that the Ga and Ewe are very closely related genetically and culturally and live in close proximity to the Fante in Accra says something. Additionally, my highest ethnic estimate via AncestryDNA is Benin/Togo (46%?), where the majority of the Ewe reside. Finally, the autosomal statistics that you referenced seems to be pretty accurate, since the DNA i inherited from Africa, according to all three major testing companies, indicates exactly 80%.

    An intriguing study would be comparing DNA samples from African Americans in Virginia (especially in and around Chesapeake Bay) with those from Nigeria (especially the Igbo) and/or Madagascar.

    As always, this was a most educational and great blog. The Ancestors love you!!!

    Liked by 1 persoon

  2. Thank you for your insightful comment my friend! You seem to have great experience in haplogroup testing. I’ve always been interested in further exploring my direct maternal lineage (L3e4) as it’s inherited from my greatgrandmother who helped to raise my mother when she was a young child. She also used to tell stories to my mother about her own mother, my greatgreat grandmother, so this family line holds a very special meaning to me. My maternal haplogroup is very frequently seen among Cape Verdeans especially from the islands where my family’s from unsurprisingly, also on the mainland it has increased frequency across Upper Guinea, but especially Guinea Bissau & Sierra Leone it seems. However it’s by no means unique to Upper Guinea, in lower frequencies it can also be seen elsewhere.

    About your maternal haplogroup L2c, i just uploaded an overview of haplogroups among 78 Gullah and it seems to be the most frequent among them, interestingly it also shows 1 L3e4 (https://tracingafricanroots.wordpress.com/dna-studies/haplogroups-aa/).

    I absolutely agree that a DNA study comparing AA’s with deep roots from Virginia with Madagascar and the Igbo would be very insightful. I’ve been intrigued about the stories of Madagascar ancestry (as measured by haplogroups as well as DNA matches and minor Southeast Asian %’s) it seems to be quite widespread even outside of Virginia, and i suspect it’s because it’s an early founding effect from the early 1700’s. The Igbo should have left a much greater genetical legacy, proportionally speaking but quite likely again as a founding effect.

    Liked by 1 persoon

  3. I have a good friend (African American) whose mtDNA haplogroups is M32c! I can see M1 showing up here and there but M32c… she so needs to delve into that piece of untold family history.

    Like

  4. I am African American, and I took the FTDNA mtDNA test indicating haplogroup L2b1b. However, I have 0 matches and have found no information on this haplogroup. I was hoping to narrow direct maternal ancestry to specific country/ethnic group. I already knew it was west african but paid for the test to learn more. Any ideas?

    Liked by 1 persoon

    • Hello John, i’m sorry you didn’t receive any matches yet. It seems you’re in the same boat as the 43 out of 97 African Americans who were tested for their mtDNA but received no matches at all, as mentioned in the study above (Ely, 2006). I fully understand why you would want to narrow your direct maternal lineage down to one single ethnic group/country. However as described in this blogpost it’s very tricky and potentially misleading to rely on haplogroups alone. Another outcome from that Ely 2006 study was that only 4 out of 97 African Americans received a mtDNA match linking them to a single place/ethnic group. While 50 out of the same group of 97 African Americans had matches connecting them with multiple places/groups! So really the odds for a “perfect match” seem to be quite slim.

      Of course this could also partially be a database problem but more fundamentally it’s because haplogroups tend to be regionally widespread and not unique to any ethnic group. And at any rate they will only tell you something about a very small part of your total ancestry.

      I’m not aware of any information regarding your specific subclade but perhaps you’ll find this study useful it mentions the frequencies of L2b1 in several West African countries and even among selected ethnic groups (starting from page 654). From the looks of it might be a marker of Senegambian origins, however other scenario’s remain possible i suppose.

      Mitochondrial DNA variation in Mauritania and Mali and their genetic relationship to other Western Africa populations.

      Like

Geef een reactie

Vul je gegevens in of klik op een icoon om in te loggen.

WordPress.com logo

Je reageert onder je WordPress.com account. Log uit / Bijwerken )

Twitter-afbeelding

Je reageert onder je Twitter account. Log uit / Bijwerken )

Facebook foto

Je reageert onder je Facebook account. Log uit / Bijwerken )

Google+ photo

Je reageert onder je Google+ account. Log uit / Bijwerken )

Verbinden met %s