African American Results


1) Introduction

On this page i will attempt to provide some analysis for the African American AncestryDNA results which have kindly been shared with me or else collected by me from public websites. If you want to skip the discussion just scroll to the bottom of the page to see a selection of African American Ethnicity Estimates. I will restrict myself to the African part of these results to be in line with the theme of this blog. In order to enable easy comparison i have scaled the African breakdown to 100% for all, leaving aside any non-African admixture. For more details on my research methodology see the front page of this AncestryDNA section. Follow this link for an overview of all the African American results:

Spreadsheet with African American results

As far as i was able to verify all of these African American results are from persons who have two self-identified African American parents. Louisiana Creoles also being included. For the sake of my research I have made an extra effort to make sure this was the case. Naturally i respect everyone’s right to selfidentify as they wish. Persons with one parent being African-American and the other one being either White or from the West-Indies, Latin America or elsewhere are still mentioned in the sheet but under the heading “Partially African American”. They are however not included in the statistics discussed below.  Again beyond the scope of my research no further implications are intended.

Even when limited in number the samplesize (n=350) of my spreadsheet compares favourable to the ones from published DNA studies, which also feature a regional within-Africa breakdown. Tishkoff et al. (2009) has only 98 African American samples for example, Zakharia et al.(2009) offers analysis for 136 African Americans while Bryc et al. (2010) studies 365 African American genomes. Due to scientific advances being made after these studies were published my survey of AncestryDNA results also benefits from a more wideranging scope of possible within-Africa origins: 9 African regions or ancestral clusters being distinguished (see also AncestryDNA Regions). Even when a most recent study, Montinaro et al.(2015), on first sight seems to include a much more numerous African American sample group (n=2023+56) as well as an impressive number of 33 (sic!) African clusters. However its African dataset is still pretty much the same as used in the referenced 2009/2010 studies, featuring only Mandenka and Yoruba samples to cover the whole of West Africa. Arguably its regional within-Africa analysis therefore might also be considered “outdated” when compared with the one provided by Ancestry.com. See also this press release: AncestryDNA Makes Scientific Breakthrough in West African Ethnicity.

The results in my spreadsheet were collected randomly and as far as i have been informed they are from people with family origins all over the United States (see columns P-U in the spreadsheet for USA state origins). The geographical distribution of my African American samples therefore also seems to compare favourably to the studies referenced above. Some of which were disproportionally making use of African American datasets drawn from only a few places. From the outset it has been one of my research goals to explore any possible correlation between the USA state origins of African Americans and the main African regions mentioned in their AncestryDNA results. In particular for people with “deep roots” (going back at least 2 generations on all lines) from the states: Virginia/Maryland, South Carolina/Georgia and Louisiana. As these states are known to have been the main points of entry for Africans being brought over to the USA. I was limited in this attempt by the availability of such results, regrettably most of all from Virginia. Even though a great number of people did mention partial origins from that state.

To conclude this introduction i like to emphasize these are obviously first of all individual results reflecting unique family trees. In my analysis however i will be focusing on the averages to get a better grip on the underlying patterns. Even when the proverbial exceptions to the rule could in fact be one of the main patterns. Furthermore DNA testing at this stage cannot be expected to be 100% accurate in estimating regional origins within Africa. See this page for more disclaimers, especially on how the country name labeling of the AncestryDNA regions should not always be taken at face value. It is also possible that with more African American testresults available you might see additional or different patterns. Even when the current samplesize (n=350) already seems robust enough to at least indicate a basic regional framework for African Americans as a whole. And in fact the regional ranking patterns from my survey appear to be nearly the same as the ones which can be verified using Ancestry’s much larger client database (see “Genetic Census of America” and section 5). I will now proceed with discussing the main patterns i’m able to pick up on from the data. Of course merely expressing my personal opinions & thoughts and not meant to be conclusive in any way .

_____________________________________________________________________________

2) Summary of Findings

***(click to enlarge)

Stats total African

***

The regional within-Africa origins of African Americans have been the central focus of my research. But the total amounts of African ancestry are of course also a very interesting topic, even if at times controversial. This kind of data often gets presented in a restrictive manner, only highlighting the mathematical mean or average of all the results. That’s why I compiled various other statistical measures for the 350 African American results i’ve collected in the chart above. Follow this link for the source data.

***(click to enlarge)

StatsAA

***

When reviewing the statistics i calculated above based on the data entered in the spreadsheet, it’s good to be aware that averages tend to hide underlying variation. That’s why it’s always advisable to also take into account other measures such as the median and also the minimum & maximum values to get a sense of the range of the scores.

***(click to enlarge)

Nr.1 regions AA

***

The chart above shows how many times each AncestryDNA region was reported as number 1 region with the highest amount in the African breakdown. It should be interpreted carefully because hypothetically if a region is consistently mentioned in second place it will not be shown in the chart. It’s therefore a more pronounced measure to bring forward some of the regional patterns. However it could be useful in hinting towards which African lineages are relatively the least diluted in the African American genepool.

***(click to enlarge)

StatsplusAAa

***

This chart features an additional macro-regional breakdown into “Upper Guinea”, “Lower Guinea” and “Central Africa” which is of my own making. In order to get a strictly “SSA” (Sub Saharan African) comparison i have excluded the region “North Africa” and corrected the other averages accordingly. Making this distinction is admittedly arbitrary and can only produce a rough proxy, given the limitations of AncestryDNA. Still I find it useful because you get to see some patterns more clearly and it conforms with what’s common in slave trade literature. For ethnolinguistical and historical maps from these 3 main regions of provenance see: Upper Guinea, Lower Guinea, Central Africa.

***(click to enlarge)

AAstats vs SC, LA, VA

***

Above chart is showing how my overall sample group of African Americans (n=350) compares with subsets of results of people who mentioned atleast both parents being from either Virginia/Maryland, South Carolina/Georgia or Louisiana. It’s just meant as a preliminary attempt to uncover any variation based on state origins among African Americans. Take note of the minimal samplesize for especially Virginia/Maryland. Also quite a few of the South Carolina/Georgia results were from close relatives (siblings, parents/child). Still it seems likely that some of the already visible trends will also be obtained with larger sample groups and may ultimately be correlated with the documented African origins for these states. For more discussion see section 5.

***(click to enlarge)

Stats Diaspora

***

This last chart shows how the average African breakdown for my African American sample group compares with various other nationalities, incl. two African control groups. It suggests a great deal of overlap and shared regional ancestry within Africa across the Diaspora. Even though the underlying ethnic origins are likely to show more differentiation. Establishing where each African region is relatively more pronounced or instead more subdued might provide insightful clues for the unique ethnogenesis of each nationality being shown as well as for African Americans. For more in depth discussion see: Afro-Diaspora AncestryDNA Results: A Comparison.

_____________________________________________________________________________

3) Observations

Total African Ancestry %’s

Going beyond just the average the first chart i posted above shows the full extent of total African ancestry (29%-100%) among my African American samples. Plus it also mentions that the most frequently observed range of total African ancestry is 80-90%. Despite previous findings there are in fact also African Americans who don’t show any non-African admixture at all in DNA testing. Even though they can still be described as mixed regionally within-Africa. I managed to collect no less than 4 African American results showing 100% African ancestry, which is about 1% of my total sample group (4/350)(edit: after writing this blogpost i have seen two more such results). I verified that these were indeed multigenerational African Americans and not persons with recent ancestry from either the West Indies or Africa itself. Most of the people with the lowest amounts of total African are from Louisiana, but some of these results were also from other states.

***(click to enlarge)

AA - histogram

***

I calculated these data also in order to compare with the recently published statistics for “5,269 self-described African Americans” who tested on 23andme. For these 23andme customers an average proportion of African ancestry of 73,2% was calculated. Which is lower than the 77,8% for my African American sample group and also lower than several other averages from published DNA studies. For example Baharian et al. (2015), has 82,1% for a nationwide sample group (HRS, n=1501) and 84,1% for a sample group drawn exclusively from southern states (SCCS, n=2128). According to 23andme itself this discrepancy might be explained by “heavier sampling from California and New York”, among other reasons, such as survey errors. However on 23andme the most frequently observed African proportion range again seems to be 80-90%, as it was also for my sample group. For more details see Bryc et al. (2014) and below histogram based on 23andme results:

***(click to enlarge)

Distribution SSA percentages for Aframs

Source “The genetic ancestry of African, Latino, and European Americans across the United States”, Bryc et al., 2014, supplement.

***

Multiregional origins instead of one ancestral “tribe”

• Zooming into the African regional breakdown of my sample group the first noteworthy aspect seems to be: a high number of wideranging African regions being mentioned in each single result. On average the first two biggest African regions being mentioned combined make up barely 60% of the total African amount (see column E in the spreadsheet). While the remaining 40% is usually described in 2-4 other main regions. When the Trace Regions are also included this often leads to nearly all of the 9 African AncestryDNA regions being mentioned in a single result.

On first sight the ranking of the main regions seems to shift all the time for each single result. Which would be pointing towards a high degree of individual variation. However when looking at the summarized totals the regional patterns do become more recognizable. All in all it may suggest that the African American genepool hasn’t stabilized yet.  Otherwise you might expect somewhat steadier proportions of each African region being reported in each single result. Instead what seems to be showing up is a great deal of heterogeneity despite African Americans arguably forming an interrelated ethnic group of their own for several centuries. With a relatively small founding population compared with other Afro-diasporic groups because of natural increase already occurring in the mid-1700’s. And additional African geneflow being relatively limited especially  after Abolition of the Slave Trade in 1808. See also:

The near absence of any relatively homogeneous African breakdown (featuring just one single dominant African region above trace level) also seems to argue against the possibility of tracing African roots to only one particular region and even less so only one specific ethnic group. As would be the case for almost any Afro-descendant in the Americas, also African Americans appear to be very mixed on their African side. A melting pot with many different African ingredients if you will. It seems natural to assume that with each new generation of African Americans being born in the 1600’s/1700’s/1800’s their within-Africa origins on an individual basis became ever more complex and numerous. Which would explain the highly varied and multi-regional AncestryDNA results of their currentday descendants.

On the other hand Africans tested on AncestryDNA can also show quite varied breakdowns, hinting at ancient migrations within their continent and geneflow between neighbouring ethnic groups. However the regions reported for them on AncestryDNA will almost always be geographically adjacent and generally few in number. Unlike African Americans whose regional scores run the whole gamut from “Senegal” to “Southeastern Bantuwithin one single breakdown, showing up in all sorts of combinations and usually featuring 4-6 Main Regions, in addition to a host of Trace Regions.

See also:

• Overall these AncestryDNA results are in line with published DNA studies which already established that historically documented origins from West & Central Africa are convincingly confirmed by the genetics of African Americans. For more details:

***(click to enlarge) (keep in mind this chart was made in 2007: updated information might vary slightly; also Inter Colonial slave trade with the West Indies is not included)

***

***

• Getting more specific “Nigeria”, “Cameroon/Congo” and “Ivory Coast/Ghana” are clearly the three most significant regions mentioned in the AncestryDNA breakdown for my African American sample group. Based on average scores and when scaled to 100% each region covering about 20% of total African ancestry. On average the remaining 40% is composed of firstmost “Benin/Togo” followed by  “Mali”, “Senegal” and “Southeastern Bantu”. The “North Africa” and “Pygmy/San” (a.k.a. “South Central Hunter Gatherer”) regions usually representing mere trace amounts, even when possibly offering valuable indications of wider origins from either the Sahel Region or Central Africa.

• Reviewing other statistical measures it appears “Nigeria” might have a slight edge, especially when judging by the frequency it gets ranked #1 in individual results (101/350). Also it seems meaningful that the maximum scores reported for “Nigeria” tend to get higher than for the other regions. It’s the only region for which i have seen original percentages above 50%, the highest being 57% “Nigeria” out of a total of 86% African. While calculated as a ratio of total African it can reach in between 60-70% of total African in a few cases. This might suggest that “Nigerian” DNA markers would be the least diluted within the African American genepool as a whole. The scores for “Cameroon/Congo” however seem to be more consistent judging from its median (19,9% versus 16.9%). Then again unlike “Ivory Coast/Ghana” and “Cameroon/Congo” the socalled “Nigeria” region might be referring more exclusively to ancestry from within that country’s borders and not also in addition from neighbouring countries. Continued discussion in section 4.

• Especially the “Cameroon/Congo” region is defined rather broadly. Which complicates its interpretation for African Americans. This category may very well reflect both Bight of Biafra (Cameroon and southeastern Nigeria) ancestry as well as denote strictly Central African (Congo/Angola) origins. However this distinction cannot be made right now unless further ancestral clues exist. For African Americans both Congolese and Biafran origins could in fact be equally significant. For more details see AncestryDNA regions, and also Igbo results showing high “Cameroon/Congo” scores.

• “Ivory Coast/Ghana” seems to be indicative firstmost of Ghanaian roots. As it is known that unlike Ghana the Ivory Coast played a relatively marginal role in Trans Atlantic Slave Trade. However Ivory Coast/Ghana” might also be referring to origins from Liberia and Sierra Leone. See also the “African Results” page where the AncestryDNA results of an Ivorian, Liberian and Sierra Leonean person can be seen. Overall it seems the average for this region therefore does not only cover the socalled Gold Coast from slave trade statistics but also the Windward Coast and (part of) Sierra Leone.

• The high scores of “Benin/Togo” are most surprising, given that relatively few captives were shipped from this area to the US.  Only about 3% of all documented slavevoyages to the USA left from the Bight of Benin according to the Trans Atlantic Slave Trade database. However on average “Benin/Togo” would be almost 15% of total African ancestry among my sample group and also many high outliers for this region can be observed (maximum being 58% of total African). Generally speaking the consistency of these high scores and also among Anglo Caribbeans seems unaccountable unless you take into consideration that this region is also signalling ancestry found outside of Benin/Togo. Especially among Nigerians and Ghanaians and possibly in particular the Ewe people. In individual cases also other ancestral scenario’s might apply however: such as an early founding effect from the 1600’s and/or early 1700’s when captives from Benin were more prominent among the slave populations in the US and the Anglo-Caribbean. Or rather a relatively late connection by way of the illegal slave trade, taking place after 1808, and known to also have involved the Bight of Benin (e.g. the Clotilde voyage of  1859). For more discussion see:

• “Is “Benin/Togo” really pinpointing origins from within Benin’s borders?

 

Less Upper Guinean ancestry than expected?

***(click to enlarge)  (for more similar maps see this page)

Clipboard01

***

Central African origins (Congo/Angola) seem to be more pronounced than Upper Guinean ancestry, among my sample group as a whole. Combining “Cameroon/Congo”, “Southeastern Bantu” and “Pygmy/San” (or rather “South Central Hunter-Gatherers”) we get a total share of 28%, while combining the average scores of “Senegal” and “Mali” we arrive at a share of “only” 17%. With a slight remaining majority of 55%, as expected, for Lower Guinea, the area in between Ghana and Nigeria. Obviously the delineation of the AncestryDNA Regions is imprecise, especially for “Cameroon/Congo” and “Mali”. As mentioned already “Cameroon/Congo” might also be signalling Bight of Biafra ancestry. While “Mali” could possibily in part also be referring to Gur speaking peoples from Burkina Faso and northern parts of Ghana, Togo and Benin. But still this preliminary outcome might be meaningful as it often seems the reverse ranking is being imagined by the general public and also many historians.

According to most estimates the proportion of Senegambia in the direct slave trade to the USA would be inbetween 20-25%. To which might be added the estimated share of Sierra Leone of 10-12%. Even when excluding the Windward Coast this would already result in a combined Upper Guinean share of direct slave trade arrivals of about 30-37%. Seen from that perspective especially the “Senegal” scores on AncestryDNA might be seen as lower than expected. On average for my sample group they represent a share of about 8% out of the whole African breakdown. Combining with the “Mali” average of 9,3%  still only about half of the documented share of captives from the Upper Guinea would be covered. Although it is likely that some Sierra Leonean ancestry will be described by AncestryDNA as “Ivory Coast/Ghana”. 

The estimated combined proportions of West/Central Africa and Southeast Africa in direct slave trade to the United States range inbetween 25-27% (see chart above and also the Slave Voyages Database, 2010 version). Which actually corresponds quite nicely with the 28% Central African share found among my sample group.

When wanting to correlate slave trade statistics with the ancestral origins of currentday populations it seems therefore one should keep in mind additional factors (see also this page). The ethnic demographics of African captives after arrival in the Americas are a big unknown for the most part, especially when multigenerational timeperiods are being considered. Some regional/ethnic groups from Africa might possibly have been better positioned to have more descendants than other groups. For several reasons such as a favourable gender ratio (more females), cumulative growth because of earlier creolization, greater victimization by domestic slave trade, being disproportionally located in areas with less mortality etc., etc. Either way resulting in relatively more DNA markers inherited from these groups to be identified within the current day African American genepool. Inspite of all the limitations of my AncestryDNA analysis it might be that the data i collected is already reflecting some of these demographic processes.

It is also interesting to contrast this outcome with what has been reported in published DNA studies sofar. Naturally because of different African datasets being used the regional definitions will not always be the same or even intercomparable. But very roughly speaking and applying the same 3-way distinction made by myself when possible:

  • Zakharia et al. (2009) (scaled to 100% African) :
    • 63% Lower Guinea (=“Yoruba”),
    • 19% Upper Guinea (=”Mandenka”)
    • 17% Central Africa (=“Bantu” + “Biaka” + “Mbuti” +”San”)
  • Montinaro et al. (2015) (scaled to 100% African) :
    • 69% Lower Guinea (=“Yoruba”),
    • 8% Upper Guinea  (=”Mandenka”)
    • 23% Central Africa (=“South Africa Bantu” + “Bantu Kenya” + “Herero” + “Sandawe” + “Mbuti” +”San”)

While to repeat my research findings based on 350 African American AncestryDNA results (see also this chart) :

  • FonteFelipe (2015) (scaled to 100% African) :
    • 55% Lower Guinea (=“Ivory Coast/Ghana” + “Benin/Togo”+ “Nigeria”)
    • 17% Upper Guinea (=”Senegal” + “Mali”)
    • 28% Central Africa (=“Cameroon/Congo” + “”Southeastern Bantu” + “Pygmy/San”)

It seems my Upper Guinean results fall most in line with those of Zakharia et al. (2009). While the difference in Lower Guinea and Central Africa might be explained by a greater range of reference populations utilized by AncestryDNA. Arguably providing more distinction. The 8% Upper Guinean (Mandenka) score by Montinaro et al. (2015), seems rather low, but it is actually the same as the 8% “Senegal” among my sample group. Even when not confirmed it is very likely that this socalled “Senegal” region on AncestryDNA is being underpinned by the very same Mandenka HDGP samples as also used by the two other studies. The difference with my total Upper Guinean share being made by the additional “Mali” region on AncestryDNA.

• Looking more closely at the Upper Guinean statistics on AncestryDNA it seems striking how “Senegal” rarely ends up as #1 main region in the ethnicity estimates (5/350). The number of high “Mali” outliers on the other hand is more noticeable with 20 number 1 rankings among my 350 African American samples. And also the maximum scores for “Mali”  tend to get higher than for “Senegal” (52,7% versus 37,2%). Even when the average ratio’s of total African for both regions are quite close (9,3% “Mali” versus 8% “Senegal”). When it comes to exceptionally high scores for just one category it is usually “Nigeria”, “Cameroon/Congo” or “Ivory Coast/Ghana” which get mentioned. Perhaps if “Senegal” and “Mali” had been combined into 1 single Upper Guinean category there would have been more higher outliers. But otherwise it seems to be a sign that these Upper Guinean lineages have been much more diluted than other regional bloodlines from Africa. Many of the higher “Senegal” or “Mali” scores seem to occur for people with deep roots in South Carolina or Louisiana. So there might very well be some variation along USA state origins. More discussion in section 5 and these subpages:

The “Southeastern Bantu” region is least likely to produce high outliers. Sofar it was ranked as #1 main region for only two results among my sample group. Still the average contribution from “Southeastern Bantu” is quite steady. Sometimes even considerable individual scores are being reported even if overall it’s clearly a minor region in most cases. The ancestral implications of this region are potentially very wideranging, going by the descriptions given by Ancestry.com (see also AncestryDNA regions). However it seems likely that primarily ancestral connections with Angola and Congo (interior) are being pinpointed. Going beyond any genetic affinity with the Bakongo which would be reflected already presumably by the “Cameroon/Congo” category. Such interpretation would be based on what’s historically plausible and a comparison with the more pronounced “Southeastern Bantu” results of other nationalities in my spreadsheet, especially from the Hispanic Americas where Angolan connections are well documented.

Strictly Southeast African options and specifically Madagascar ancestry are certainly also possible but based on slave trade statistics chances would be fewer. Captives from Southeast Africa are estimated to have been only 1-2% out of all Africans being brought to the USA. While slave trade with West/Central Africa, incl. parts of northern Angola, would be about 25% for the USA according to the Slave Voyages Database. When combined with additional evidence the Madagascar option may be made more plausible though in individual cases. An updated version of AncestryDNA will hopefully provide more clarity.
_________________________________________________________________________

 4) Additional Nigerian origins beyond reported by “Nigeria”?

***(click to enlarge)

AA vs Africa

***

This section will be speculative and based on conjecture to some degree. But it might be useful for illustrating how it can be worthwhile to critically scrutinize the country name labeling applied on each AncestryDNA region. Actually this goes for the labeling of any ancestral categories used also by other different DNA testing companies. In the chart above a comparison is shown between the average African breakdown for 350 African Americans and two African control groups: 24 Cape Verdeans and 7 Nigerians. Obviously the sample size of especially the Nigerians being very limited. However there are already some illuminating regional patterns to observe:

  1. “Senegal” is only reported at trace level for the Nigerians, while it is clearly the predominant region for Cape Verdeans (see also Cape Verdean Results). This is according to expectation, which should render the 8% “Senegal” score for African Americans more reliable, or atleast add more perspective to it.
  2. “Mali”, “Ivory Coast/Ghana”, “Southeastern Bantu”, “Pygmy/San” (South Central Hunter-Gatherers) and “North Africa” are also all reported at either trace level or with minimal amounts for the 7 Nigerians. Again in line with what you would assume. And adding more solid basis to the scores for these regions as reported for both African Americans and Cape Verdeans.
  3.  But barely half of the breakdown for the 7 Nigerians is assigned to the region “Nigeria”.  While unsurprisingly “Nigeria” is reported at minimal level for Cape Verdeans. The African American average for “Nigeria” being about 40% (20/53) of what’s been reported for actual Nigerians. With some of the higher outliers among my African American results actually scoring up to 70% “Nigeria” out of their total African amount!  In other words more “Nigerian” than actual Nigerians, at least in their African breakdown.
  4. Cameroon/Congo” and especially “Benin/Togo” are significant additional components for the Nigerians. The averages shown for these regions are not that far removed from the African American averages. “Benin/Togo” being more prominent for the Nigerians and “Cameroon/Congo” more significant for the African Americans.

 

***(click to enlarge)

Nigeria2

Source: Ancestry.com

***

AncestryDNA results for 2 Nigerians (both Igbo)

***(click to enlarge)

Igbo (Anambra)

***(click to enlarge)

Igbo On1

***

By now it should be clear that the genetic origins of actual Nigerians are not fully covered by just only the region “Nigeria”. Additional regions are needed to describe their DNA, in particular “Benin/Togo” and “Cameroon/Congo”. This is actually also very well explained by Ancestry.com itself, as can be seen in the screenshot above and additional disclaimers being given on their website (see also AncestryDNA Regions). Based on their more numerous sample size of 67 Nigerians they mention that the “typical native” would be 69% “Nigerian”. So that’s a bit more than the 53% for the 7 Nigerian results i’ve collected. Aside from individual variation this difference is probably also correlated to the ethnic origins of the samples. Ancestry.com doesn’t specify the ethnic origins of their reference populations but among the 7 Nigerian results within my spreadsheet, 5 are Igbo and 2 are Yoruba.  Either way this might be of consequence for the way we should interpret these regions appearing in the results of African Americans (and other Afro-Diasporans).

Afterall it might very well be that similarly also for African Americans the “Nigeria” region may not be reporting the full extent of their genuine Nigerian origins. But rather tends to underestimate them, generally speaking. And the reported Benin/Togo” and “Cameroon/Congo” amounts for African Americans might in part actually have been inherited by way of Nigerian ancestors who already carried these DNA markers in their own genome, just like the two Igbo results shown above.

Just taking a wild swing at the numbers i wouldn’t be surprised if it turns out that about 5% of the 14.6% “Benin/Togo” average score for African Americans is actually hailing from Nigeria (while another 5 % could actually be from Ghana, the remaining 4,6% “Benin/Togo” then being much more in line with what you would expect). Another 5% might have to be deducted from “Cameroon/Congo” for actually referring to southeastern Nigerian origins. Which would imply that the grand total of “genuine” Nigerian ancestry for my African American sample group could be as high as 30% instead of 20%, on average. Of course this is mere speculation at this stage 😉 Only an update by AncestryDNA might clarify this issue, improving their reference panel with additional West African populations.

The Nigerian roots of African Americans may not yet be known in exact full detail or proportion however many valuable indications have been historically documented. And especially southeast Nigerian (Igbo/Biafra) ancestry is a well researched topic among American historians. For more details and references see:

_________________________________________________________________________

 5) Regional variation according to USA state origins?

***(click to enlarge)

1630598_picture1_jpegc1a1f9a3a51259def050dcdad90f68ea

Source: “Exchanging Our Country Marks: The Transformation of African Identities in the Colonial and Antebellum South”, (Michael Gomez, 1998).

***

(click to enlarge)

***

(click to enlarge) (Inter Colonial slave trade with the West Indies is not included)

EltisUSA99_zps95858018

Source: Trans-Atlantic Slave Trade Database (2010 Estimates) (http://www.slavevoyages.org)

***

Distinct regional patterns can be observed in the Trans Atlantic Slave Trade for each main port of entry into the United States.  As can be verified from the charts above. Even when all ports of disembarkation show the same range from Senegal to Mozambique, it is the variation of the regional proportions which may have resulted in localized blendings which stand out from each other. Several historians have written extensively about this topic because it had significant consequences for how nascent African American subcultures developed locally, especially in Virginia, South Carolina and Louisiana. Good grounds perhaps to assume that the within-Africa origins of African Americans could vary as well according to which USA states they are from.

I therefore calculated some additional statistics for the results i have collected for people who were either born or had at least two parents from Virginia/Maryland, South Carolina/Georgia or Louisiana. It was my intention to gather such information especially for people with “deep roots” (going back at least 2 generations on all lines) for those states. However i was limited in this attempt as it turned out that most people of my sample group have mixed USA state origins (133), while for 120 results i wasn’t able to gather any USA state information at all. However for 97 results i was informed of state origins in either South Carolina/Georgia (59), Virginia/Maryland (10) or Louisiana (28). Obviously allowance to be made for possible survey errors ;-). But in several cases i was actually able to verify for myself (after kindly being granted access to view online family trees) that family origins in those states could indeed be traced back for several generations on all lines. Follow these links for the spreadsheets containing their results:

Generally speaking it seems however that few African Americans can trace their origins exclusively to only one USA state when longer time periods are considered. Going back not just 2 or 3 generations but rather 2 or 3 centuries. Around the time when their first African-born ancestors would have been brought over to the United States. This was confirmed to me by a few persons among my sample group who had admirably managed to perform extensive genealogical research into the 1800’s. It is perhaps also better understood when taking into account the high degree of mobility among African Americans throughout US history. Not only in recent times during the Great Migration from the South. But especially because of Domestic Slave Trade, the socalled second Middle Passage into the Deep South.

Check these links for more background info, references and maps:

Given all these interstate migrations and the wideranging regional diversity of Africans already from the start it perhaps seems futile to find distinct African compositions for any USA state. Even for those states with noted profiles like Virginia, Louisiana and South Carolina.  Also i have not yet touched upon the topic of intercolonial slave trade with the West Indies & illegal slave trade after 1808 which would render some of the African origin patterns per state even more complex to disentangle. Especially for Louisiana. Still the data i collected sofar does tentatively seem to point towards a certain degree of regional variation indeed. Even if not to the full extent as suggested by the charts posted above.

***(click to enlarge)

AAstats vs SC, LA, VA

 

***(click to enlarge)

AA - statsplus

 

***(click to enlarge)

Piechart AA

 

***(click to enlarge)

piechart SC

 

***(click to enlarge)

piechart LA

***

It should be emphasized again that the charts i posted above are in part based on minimal sample size and their interpretation is restricted by the limitations of DNA testing itself. Therefore they are not intended for easy generalizations! What i have observed across statelines is still very much individual variation and a high degree of African diversity showing up in each single result. Being from either Louisiana, Virginia or South Carolina does not imply you will automatically have an x amount of African region A and a y amount of African region B.  Neither would any African region be unique or exclusive to any USA state. It’s more complex than that and again also dependent on unique family trees 😉

Still the South Carolina/Georgia data is close to having a fairly representative sample size already. And generally the outcomes seem to reflect localized genepools which on average can show distinct proportions for certain regions. The most noticeable discrepancies seem explainable by way of the documented African origins of each state. Interestingly my own modestly scaled research findings seem to be backed up in most regards by the much greater data resources available to Ancestry.com. More details on the end of this section. But first i will comment on what seem to be the main features of the regional variation i have uncovered sofar in the charts above.

As shown in the first chart the top 3 main African regions observed nationwide are still the same on average for each state. However there are some seemingly telling ranking differences to be noted:

  • For South Carolina/Georgia it is no longer “Nigeria” (15,6%) but rather “Cameroon/Congo” which stands out the most with an average ratio of 22% out of total African.
  • For Louisiana “Nigeria” is less subdued (17,9%) but still it’s “Ivory Coast/Ghana” which appears most significantly with an average (scaled) score of 22,7%.
  • For Virginia, inspite of the minimal samplesize, it seems no coincidence that “Nigeria” is showing up most prominently with an average of 29,2%.

Looking at the 3-way regional breakdown of my own making, it seems only South Carolina/Georgia results stand out for being more Upper Guinean as well as more Central African on average than nationwide. Virginia/Maryland might be expected to mostly conform to the national averages but it seems more surprising that the Louisiana results (sofar) do too. Especially in regards to Louisiana’s Upper Guinean (Senegambia) origins which are always well celebrated. However there are hardly any differences in the Louisiana averages for either “Senegal” or “Mali” compared with the nationwide averages.

The frequency of regions being ranked #1 is perhaps the best indicator of which distinct African lineages may have been preserved the most in each state. For Louisiana it seems especially “Benin/Togo” origins might be more pronounced than the national average. Despite the subdued averages it might still already be meaningful that also the “Senegal” and especially “Mali” outliers are more visible in my Louisiana sample group (n=28) than for African Americans as a whole (n=350).

This also goes for South Carolina/Georgia for which i observed some of the highest “Senegal” and “Mali” scores nationwide. The “Cameroon/Congo” frequency of number 1 region scores looks even more imposing. Given the history of its documented African roots for South Carolina “Cameroon/Congo” might be picking up more so on Congolese origins than on Cameroonian/Biafran ones. Making allowance for individual variation of course. See also these previous blogposts:

I did not (yet) include a piechart for Viginia/Maryland, as its minimal samplesize might result in an overall distorted picture. However it seems telling that out of 10 Virginia/Maryland results sofar “Nigeria” was mentioned 6 times as the biggest region. Perhaps more likely to be based on coincidence but still fascinating also that “Southeastern Bantu” only obtained number 1 rankings (2x) among my entire African American sample group (n=350) in the results of persons from Virginia/Maryland. Both early Madagascar and even earlier Angolan connections for this state being well documented.

 

Confirmation by Ancestry.com’s “Big Data”?

____________________

A Genetic Census of America

Using AncestryDNA results from over a quarter million people, the AncestryDNA science team set out to perform a “genetic census” of the United States: a survey of the U.S. using only DNA. Where did the ancestors of today’s Americans come from? Do Americans in the Midwest hail from similar places of the world as in the Northeast, or as in the South?”

Source: Ancestry Blog

____________________

This section can at best only have been describing the beginnings of a very rough outline of any possible interstate variation, due to limited sampling. In this regard it is however very interesting that similar research has also been performed by Ancestry.com recently. And it seems most of my major findings have  been replicated by Ancestry.com itself, based naturally on a much larger sample size. If you follow this link to their blog you can select each one of the 26 regions used by AncestryDNA and see its average score within each USA state.

My research findings are not completely intercomparable with this “Genetic Census of America”.  The regional averages being given (in decimals) seem to represent the averages for the entire state population regardless of racial/ethnic background, whereas i have been attempting to single out results for people with two African American parents. Also Ancestry’s data seems to be based on the original scores as mentioned in the Ethnicity Estimates whereas i have been scaling the scores in the African breakdown to 100%. The higher averages, shown in dark green, therefore appear to represent firstly a greater overall African American presence in that state. So it’s not surprising that Mississippi turns up as having the highest average for almost all African regions. As afterall it also has the greatest share of African Americans in its overall population (37% according to the 2010 census). This is however not the case for the region “Mali”, “Senegal and “Southeastern Bantu”. If i checked the data correctly these are the only African regions for which not Mississippi but rather South Carolina is being shown with the highest average. I suppose this can serve as some sort of confirmation already that South Carolina has higher than average Upper Guinean ancestry as well as from Central Africa.

By rather reviewing the ranking of all African regions per single state my survey outcomes can be compared more directly with Ancestry’s “Big Data”. It’s quite remarkable that especially the number 1 positions for Virginia, South Carolina and Louisiana are identical with what i’ve found sofar! In fact also the top 3 always consisting of the trio “Nigeria”, “Ivory/Coast/Ghana” and “Cameroon/Congo”, the middle position of “Benin/Togo”, the consistently subdued rankings of  “Mali”, “Senegal”, “Southeastern Bantu” and the trace level mentionings of “North Africa”, “South Central Hunter Gatherers” are all very similar to what i have discussed already above. This also goes for the approximate proportions when scaled to 100%. So in that sense Ancestry’s “Genetic Census” seems to be contributing to the robustness of my own findings.

Regional ranking according to Ancestry’s ‘Genetic Census’

***(percentages in parentheses represent African regions scaled to 100%, see also this sheet for calculations)

Virginia:

  1. “Nigeria”: 0,02301787647 (20,93%)
  2. “Ivory Coast/Ghana”: 0,02195381044 (19,96%)
  3. “Cameroon/Congo”: 0,02183056309 (19.85%)
  4. “Benin/Togo”: 0,01686335024 (15.33%)
  5. “Mali”: 0,007776231427 (7.07%)
  6. “Senegal”: 0,007261001474 (6.60%)
  7. “Southeastern Bantu”: 0,006914652417 (6.29%)
  8. “North Africa”: 0,00245364888 (2.23%)
  9. “South-Central Hunter-Gatherers”: 0,001918334611 (1.74%)

***

South Carolina

  1. “Cameroon/Congo”: 0,02870212366 (20.79%)
  2. “Ivory Coast/Ghana”: 0,02552819495 (18.49%)
  3. “Nigeria”: 0,02496276593 (18.08%)
  4. “Benin/Togo”: 0,0175741489 (12.73%)
  5. “Mali”: 0,0136826057 (9.91%)
  6. “Senegal”: 0,01312859621 (9.51%)
  7. “Southeastern Bantu”: 0,009003610095 (6.52%)
  8. “North Africa”: 0,002766388013 (2.00%)
  9. “South-Central Hunter-Gatherers”: 0,002713272555 (1.97%)

***

Louisiana

  1. “Ivory Coast/Ghana”: 0,02658309405 (19.34%)
  2. “Cameroon/Congo”: 0,02589036324 (18.84%)
  3. “Nigeria”: 0,02572959577 (18.72%)
  4. “Benin/Togo”: 0,01994058628 (14.51%)
  5. “Mali”: 0,01261197325 (9.18%)
  6. “Senegal”: 0,01187605134 (8.64%)
  7. “Southeastern Bantu”: 0,008331560828 (6.06%)
  8. “North Africa”: 0,003908815789 (2.84%)
  9. “South-Central Hunter-Gatherers”: 0,002546339517 (1.85%)

 

African region with the highest average per state

***(click to enlarge)

VANAIJA

***

(click to enlarge)

SCCONGO

***

(click to enlarge)

LAGHA

***

 

______________________________________________________________________________

6) African American Results

As far as I was able to verify all of these screenshots below are from persons with two African American parents. This section is intended to illustrate the individual variation among African Americans firstmost. I will be posting results which feature one region or one aspect in extra pronounced degree. My accompanying comments should be taken as informed speculation on my part they’re not meant to exclude other possibilities or simplify complex family histories. I like to thank again all the persons who kindly agreed to share their results with me!

In addition i have also selected some screenshots taken from public websites. As i found them to be of potentially educational benefit for others. I have asked for prior consent whenever i could but regrettably wasn’t able to do so in all cases. I have naturally taken great care to cut away any name details in order to safeguard everyone’s privacy. Apologies in advance to anyone who recognizes their results and is not comfortable with this blog page featuring them. Please send me a PM and i will remove them right away.

For more information on what type of ethnic origins could possibly be implied by these regional breakdowns, see also:

***

“Average” Scores

These two results below show the closest similarity and most likely also closest genetic affinity to the regional averages i have calculated for my 350 African American samples. Especially with regards to their total African amounts and the ranking of the first four regions. An evenly balanced top 3 being composed of “Nigeria”, “Cameroon/Congo” and “Ivory Coast/Ghana”. With “Benin/Togo” being the most significant of the remaining regions. Despite being “typical” in this sense, one might distinguish various other types of “representative” results as well. Afterall there is a great degree of individual variation to be observed. Actually it took me quite some effort to select these 2 screenshots. Highlighting that statistical averages, even when valuable in themselves, should always be complemented with additional information if you want to acquire more insight.

***

AAaverage1

 

***

AAaverage2

 

***

Regionally balanced

Some of these breakdowns below are showing 7 Main Regions, some even all the 9 African regions available on AncestryDNA when also including the Trace Regions. Other results are featuring a very evenly balanced top 4. Actually most results among my sample group show fairly balanced proportions with about 4-6 regions in the main breakdown on average. These are however some of the more extreme examples. Just to provide some contrast with the other results i will be posting which will represent the highest outliers for just one single region.

***

This first result is also noteworthy for showing “Senegal” in shared first position. A number 1 ranking for “Senegal” seems to be quite rare; among my sample group of 350 people it only occurred 5 times. Also the “Southeastern Bantu” score is quite pronounced.

***

 

AABAL1

***

AABAL4

***

AABAL2

***

AABAL7

***

AABAL10

***

AABAL3

***

AABAL5

***

AABAL8

***

AABAL6

***

100% African

I have sofar seen 6 results for verified African Americans who are of 100% African descent. Not completely impossible to find therefore ;).  It seems some reports in this regard have been exaggarated. The first screenshot is from a person from Florida, the second one from Georgia, and the remaining ones all have origins from South Carolina, two of them also selfidentifying as Gullah/Geechee. Perhaps not by coincidence as these states are also often singled out by DNA studies for having above average African ancestry. The latest one being Baharian et al. (2015).  This is of course a very limited sample, but the high “Mali” scores for the two persons from South Carolina look quite distinctive. A crucial difference with the AncestryDNA results for Africans seems to be the widerranging scope of the regions being mentioned in the breakdown and also a higher number of these exclusively African regions. See also African results.

***

AA100a

***

AA100e

***

AA100c

***

AA100d

***

Eventhough this person’s breakdown is quite diverse, the “Ivory Coast/Ghana” score is still very noticeable and in fact it is the highest i’ve observed among my sample group. Aside from signalling origins from Ghana this region might also be suggestive of origins further west from Liberia and even Sierra Leone. All of these 3 ancestral options being quite likely for South Carolina i suppose.

***

AASC1

***

High “Nigeria”

As discussed above this region might represent the relatively least diluted African lineages among my sample group. Because it is the only region for which i have observed original percentages of above 50%. Plus when calculated as a ratio of total African ancestry, it is the only region to show scores of inbetween 60-70% of total African ancestry. Of course because of individual variation it is also very much possible that other people will score much lower amounts of “Nigeria” or even zero percent 😉

***

AANIG1

***

AANIG2

***

AANIG3

 

***

The next 3 results no longer show original “Nigeria” percentages of above 50%. However because their total African amounts are relatively lower the relative contribution of “Nigeria” is still very pronounced: in between 64-71%. Atleast two of these persons also having significant Virginia connections. An area where southeastern Nigerian ancestry might very well be the most concentrated within the US.

***

AANIG4

 

***

This person with a Virginia background, shows the highest relative proportion of “Nigeria” i’ve seen sofar. It’s a very concentrated regional contribution of 71% (48/67) out of his total African breakdown. He also made an excellent Youtube video which can be seen via this link.

***

AANig8

 

***

AANIG5

***

AANIG7

***

High “Cameroon/Congo”

This region is somewhat ambigious as it might suggest either origins from Congo/Central Africa or from Cameroon/southeast Nigeria. The first result is from someone who’s from South Carolina. Given the strongly documented presence of Congolese people in that state it seems tempting to assume that in this case Central African ancestry is being picked up on.

***

AACON1

***

The same amount of 47% “Cameroon/Congo” but proportionally even more significant because it’s more predominant (47/78=60%) with the total African being 78%.

***

CON1, AACON2

***

This breakdown is from someone from Louisiana. Again Central African origins seem most likely given what has been documented for this state (and also for this person’s family parish). Just like the first result only a minimal trace amount of “Nigeria” is showing up.

***

AACON2

***

Another result from someone with a South Carolina/Georgia connection by way of her parents. The case for Central African origins being implied is made even stronger because of the 9% “Southeastern Bantu” and especially the atypically high 5% “South-Central Hunter-Gathers”. Which in this case would most likely be referring to Pygmy origins, quite possibly already absorbed by Bantu speaking ancestors.

Intriguingly this person’s maternal haplogroup is U6a3b, which can be linked either directly to North Africa or with the Sahel Region. Quite uncommon and therefore providing a valuable ancestral hint. However the score for “North Africa” is less than 1%, highlighting that mtDNA can only give you information about a very limited part of your total ancestry. On the other hand the “Mali” and “Senegal” scores are much more noticeable and might also very well be connected with this person’s maternal lineage.

***

AACON3

***

AACON4

***

This person’s original percentage for “Cameroon/Congo” might not be extraordinarily high. However when calculated as a ratio (31/62) it’s exactly 50% of total African. Among the highest for my sample group.

***

AACON5

***

High “Ivory Coast/Ghana”

Judging from the few African results i’ve seen this category is very predictive of both Ghanaian (Ga) and Ivorian (Akan) origins. However in addition also Liberian and (southern) Sierra Leonean ancestry is being described by this region. Making the distinction at this stage cannot be done yet, unless other ancestral clues exist.

***

AAGHA1

***

AAGHA5

***

AAGHA2

***

AAGHA3

***

AAGHA4

***

 

High “Benin/Togo”

The considerable average score for this region (14-15%) and especially the many high outliers, as shown below, are one of the main surprises from my survey of African American AncestryDNA results. In fact aside from “Nigeria” some of the highest outliers i have observed were for this region. Calculated as a ratio of total African ancestry, it can get as high as 58%. Which implies a relatively “pure” regional lineage. This is rather unexpected given the relatively low documented number of captives arriving from this area. About 3% of documented slavevoyages to the US left from the socalled Bight of Benin. This number is however not including Inter-Colonial trade with the West Indies, nor illegal slavetrade after 1808. Also it’s important to be aware that this region is not always pinpointing origins from strictly within Benin’s or Togo’s borders.  For further indepth discussion see also “Benin/Togo” region.

***

AABEN2

***

AABEN3

***

This result shows the highest relative contribution of “Benin/Togo” for my sample group. Calculated as ratio of total African ancestry it’s 58% (43/74). There were 6 other results who showed a ratio of above 50% “Benin/Togo”. Four people scored more than half of total African ancestry for “Ivory/Ghana”. While for “Cameroon/Congo” and “Mali” there were only 2 results with regional ratio’s above 50%. “Nigeria” seemingly being in a league of its own as it had 18 such results and is the only region to show a relative share of total African ancestry of above 60% and up to 70% in a few cases.

***

AABEN4

***

AABEN5

***

AABEN1

***

High “Mali”

“Mali” is more difficult to interpret than “Senegal” (see also AncestryDNA Regions). Besides the country Mali also Guinea Conakry should be the primary ancestral location hinted at. I imagine the DNA markers associated with this region could perhaps also be found among selected ethnic groups in Sierra Leone, Guinea Bissau or even Senegal itself. However it is possible that at times also ethnic origins are being pinpointed beyond Upper Guinea, and more so from neighbouring areas of Mali to the southeast. Eitherway it seems meaningful that despite similar averages of 8-9% it is “Mali” for which i observed the highest number of outliers and not “Senegal”. Also the maximum scores for “Mali” tend to get higher than for “Senegal”, implying less dilution. Again interpretation at this stage is perhaps too premature. But it might suggest that Upper Guinean origins beyond Senegambia in a strict sense, either from the deep interior or from areas to the south, are relatively more important for African Americans.

***

This result below is quite extraordinary for not only showing the highest “Mali” percentage among my sample group of 350 African Americans, but also the secondhighest score for “Southeastern Bantu”. The person behind the results has family origins from both Texas and Louisiana.

***

AAMAL1a

***

AAMAL2

***

Also a notable score of 2% “North Africa” in this breakdown. Which might very well be correlated with the high “Mali” score. Even if not per se so.

***

AAMAL3

***

AAMAL4

***

These results below are from a person from Maryland, highlighting that high scores for this region are by no means exclusive to any state. Still sofar i have observed the greatest number of “Mali” being ranked number 1 among people with South Carolina & Louisiana origins (11/20).

***

AAMAL5

***

 

High “Senegal”

This region has produced relatively very few high scores among my sample group. Also the maximum original score of 25% “Senegal” is rather subdued and implies a relatively greater degree of dilution when compared with most of the other regions, incl. also “Mali”. It is very striking that sofar the only results i’ve seen with “Senegal” ranked number 1 (5/350) were mostly for people with origins from South/North Carolina, with the highest regional ratio coming from Louisiana, as expected perhaps. Even when the “Senegal” averages for these states are pretty much in line with the nationwide averages, about 8-9%.

The samples being used to underpin this region are most likely Mandenka (see also AncestryDNA Regions). It will be very interesting to see what happens with these “Senegal” scores if also Wolof samples are being utilized. It might be that the ethnic composition of socalled Senegambians being brought to the USA has been misjudged by historians based on incomplete documentation. With the relative contribution of people located more towards the interior and also to the south being underestimated.

***

The first two results are from a father and son with origins from South Carolina. The first breakdown also featuring a very evenly balanced top 4, incl. a rather high “Mali” score. The second screenshot showing the highest original percentage for “Senegal” i have observed among 350 African American results.

***

AASEN2

***

AASEN1

***

Again a result for a person from South Carolina and also featuring an evenly balanced top 4. Highlighting that “Senegal” usually does not reach dominating proportions even when it is mentioned as the region with the highest amount in the breakdown.

***

AASEN3

***

This result however being a notable exception. “Senegal” is being mentioned as the only main region above trace level, and when calculated as ratio of total African ancestry it represents a share of 37% (16/43). By far the highest i’ve observed, the second highest regional ratio being 26,9% of total African. The person behind the results is of well documented Louisiana Creole background on all family lines.

***

AASEN5

***

High “Southeastern Bantu”

This region is very broadly defined, however for most people i imagine Central African origins (Angola, interior Congo) are being referred to (see also AncestryDNA Regions). Even when other ancestral options are also still possible. Only an update of AncestryDNA might bring clarification. It is clearly one of the least significant regions overall, however unlike “South Central Hunter-Gatherers” and “North Africa” it does frequently appear above trace level. And at times even with considerable percentages in between 10-25%. It is very uncommon though for this region to be mentioned as the biggest region in the breakdown. This only happened twice among my sample group of 350 African Americans.

***

AASEB4

***

AASEB3

***

AASEB1

***

AASEB2

***

High “South Central Hunter-Gatherers”

In my spreadsheet i’m referring to this region as “Pygmy/San” as these are most likely the reference populations which were being used to single out the DNA markers associated with this category. It is most likely representing an ancestral component which was inherited from Bantu speaking ancestors who are known to have incorporated geneflow from Pygmy’s (Biaka and Mbuti) but also the San already many centuries or even millennia ago (see this sheet for an overview of Ged-Match results for various Bantu populations). In this sense it might be indicative of wider Central African ancestry even when it is itself almost always shown as Trace Region. In fact for almost a third of my sample group this region is not showing up at all, while the highest percentages i’ve seen, 5% & 4%, only show up for less than 10 results. Even scores of 2% or 3% already being relatively high.

***

This breakdown is quite exceptional and rather puzzling as it has “South Central-Hunter-Gatherers” above trace level, while “Southeastern Bantu” and “Cameroon/Congo” are mere Trace Regions with lower amounts even. It might be due to a quirky recombination or else some other ancestral explanation might be valid. As in fact i have also observed some West African AncestryDNA results featuring this “South-Central Hunter-Gatherers” region (see also African results).

***

AAHG2

***

AAHG1

***

AAHG3

***

High “North Africa”

Overall this region showed the lowest average of all 9 African regions on AncestryDNA for my African American sample group. More than half of all African American results i’ve seen show zero “North Africa”, while only about 30 results have 2% or more. Many of them from Louisiana. The highest outliers scoring 4% “North Africa”, which happened only twice. From what i’ve observed among other nationalities this region seems to strongly correlate with Hispanic or rather Iberian origins. When reviewing the data from Ancestry’s “Genetic Census” this can be confirmed because the highest average for this region is appearing for New Mexico while all the other states being mentioned with higher averages are also known for having numerous Hispanic populations. In this sense any Trans Atlantic Slave Trade connections are out of the question. However there are also other ancestral options involving West Africans already incorporating minor “North African” components in their genome, this goes especially for the Fula people but could also be the case for other ethnic groups from the Sahel Region. For more detailed discussion see also section 4 of this page.

***

This breakdown shows the highest “North African” score i’ve seen. Eventhough it is only mentioned as a Trace Region, the reliability should still be quite high as it’s based on distinct DNA markers when compared with the other African regions. The results are from a person with Louisiana Creole origins. It is very interesting in this regard that there are several ancestral options which have all been documented for this state: Spanish or even Canarian ancestry (Isleños); Fula (a.k.a. “Poulard” or “Peul”) ancestry from Upper Guinea, and even unmitigated Moorish ancestry from Mauritania (the socalled “Nar”).

***

AANA2

***

This breakdown is notable in that the 3% “North Africa”, which is already exceptionally high for African Americans, is showing up for a person of above average African descent. Perhaps making it more likely that this minor component will be associated with the 14% “Senegal” and the 3% “Mali” rather than any distant Iberian/Hispanic ancestry. Also the 4% “South Central Hunter-Gatherers” is quite pronounced btw.

***

AANA1

***

StatsAA

***

Youtube

I always enjoy seeing how other people react to their new DNA test results and how they interpret them. Youtube has many such videos of people discussing their results. Below is just a small selection.

***

***

***

***

***(she’s African American despite the name 🙂 )

***

*** (also check out her follow-up video she made almost two years afterwards)

 

***

***

 

***

***

***

***

***

***

 

 

44 gedachten over “African American Results

  1. Excellent work! Thank you for your immense contribution to genetic genealogy in general and Afro-Descended ancestry in particular. As far as i’m concerned, your seminal research oriented blog carries as much weight (if not more) as extant scientific literature covering this subject matter. I speak from the perspective as a professional research librarian/information specialist of 22 years. This work is critically invaluable, no doubt! Thank you, again, Felipe.

    Liked by 1 persoon

  2. It’s always interesting, thank you for the informations you share with us. About African Americans, I have a few questions to ask:

    First, currently I’m reading a book by Henry Louis Gates: “In Search of Our Roots: How 19 Extraordinary African Americans Reclaimed Their Past” and I search in a lot of websites African Americans who have done DNA tests and I was surprise so many AA have DNA matches with Cameroonian tribes (the Bamileke being the most common) while only few slaves came from this current country, even genealogists and historians are suprise. I know its part of the Bight of Biafra region so they may have been taken them inside and brought to Calabar (Nigeria) for example but their number is still suprising for me. Have you an idea why there’s many Cameroonian roots or maybe it’s easier to detect ?

    The second question I want to ask it’s why DNA results aren’t able to find any Bakongo roots. Even people who have matches with Bantu tribes from Central Africa don’t have Bakongo results like for examples the actor Chris Tucker who has Biaka/Mbenzele (Central African Republic) roots, Oprah Winfrey (Bantu tribe from Zambia) or Mae Jemison who has Kikuyu (Kenya), it’s always far inland. When they discover someone has ancestry from Angola-Congo region, it’s usually Mbundu or they don’t know it. I know it’s only the minority but I find it odd there’s no Bakongo matches while all the historians are agree they were a big part of the American slaves trade.

    Liked by 1 persoon

    • Thank you for commenting! Very observant and with important implications for identifying the African origins of African Americans. Your first question is dealing with an intriguing issue i have been wondering about myself as well for quite a while. I actually have been intending to do a future blogpost about it! I will try to provide more detail by that time.

      Just my 2 cents but it seems to me that these results matching African Americans with Cameroonian ethnic groups are usually based on haplogroup testing (mtDNA and/or Y-DNA), which is inherently limited in scope as it only identifies one single ancestral lineage out of potentially hundreds of others (see also this page). Unlike the autosomal AncestryDNA test which covers a person’s complete genome and therefore attempts to describe someone’s ancestry in its totality. It’s my suspicion that these Cameroonian matches are obtained mostly for two reasons:

      1) widespread sharing of the same haplogroups among ethnic groups in Cameroon/Nigeria (the Bight of Biafra hinterland); receiving a match with a Cameroonian ethnic group therefore doesn’t exclude the possibility of also matching an ethnic group from Nigeria (given the availability of their samples).

      2) overrepresentation of Cameroonian samples in AfricanAncestry.com’s database. Eventhough no actual numbers are mentioned you can verify this for yourself by looking at this overview of their database. Cameroon is represented with no less than 18 ethnic groups. Also Gabon and Guinea Bissau are particularly well represented, at least when counting the ethnic groups. A good number of ethnic groups from Nigeria and Congo is also featured, however not in proportion to their much greater population size and ethnic diversity.

      A third possible reason would be that the involvement of Cameroon in the Trans Atlantic Slave Trade might have been underestimated or obscured because of the main slaveports being located in eastern Nigeria. Even so i’m very doubtful that any genuine Cameroonian ancestral contribution could ever be greater than the one hailing from eastern Nigeria, generally speaking.

      If you check that overview of African Ancestry’s database you’ll see that they also include Bakongo samples from Angola, DRC and Congo Brazzaville. But the number of samples is not mentioned, so it could very well be just a few of them. I’m just guessing here but perhaps that could partially explain why Bakongo matches (based on haplogroup testing) have seemingly been so rare up till now. It would simply reflect a database issue.

      On the other hand it might again indicate that the true extent of inland slave sourcing has been obscured by focusing on the coastal slaveports and disregarding connections with the deep interior. The very frequently used term of “Congo’s” to refer to Central African captives would then indeed have been mostly an umbrella term and not an exact ethnic designation exclusively referring to Bakongo’s. We need more refined DNA testing and especially more sampling among African ethnic groups for any new insights.

      It’s fascinating already that the socalled “Cameroon/Congo” region on AncestryDNA is probably partially based on Bakongo samples (see AncestryDNA regions). As described above it’s also one of the top 3 main regions for the 350 African American results in my survey. It seems to be a very consistent component across the USA but on average especially so for persons from South Carolina, and this state probably has the greatest proportion of documented origins from Central Africa.

      Like

      • Thank you for your answers. Another quick question, can you enlighten me about the number of slaves from Senegambia because the first time I saw the number I think it was 3 or 4 years ago, they represented 14 % of the africans deported to america and now it says they were almost 24 % of the slaves. Besides Louisiana, their influence doesn’t seem so huge, yet there are the second largest group of slaves in Virginia and the Carolinas/Georgia.

        Liked by 1 persoon

      • No problem! I’m guessing you got that 14% estimate from a source which was ultimately referring to Philip D. Curtin’s “The Atlantic Slave Trade A Census”. This historian calculated a total estimated share of Senegambians for the entire USA of 13,3% (p.157). His book has been very important for Slave Trade studies however it has also become outdated in some respects (it was first published in 1969). In the last two decades or so the Trans Atlantic Slave Trade (TAST) a.k.a. Slave Voyages database has become the standard reference for any serious scholarship on slave trade statistics. Judging from that source the number of Senegambians brought into the USA would indeed represent inbetween 20-25% of the total number of captive Africans. There are still some inherent flaws (e.g. inter-colonial slave trade is left out and also not all slave voyages were documented), however for the USA its coverage is very impressive and reliable. From what i’ve read this database might be updated during this year so we might see a slight change still in the total numbers but i don’t expect any radical change of the relative proportions. I have featured some screenshots from this invaluable source on this page, take note it reflects the state of knowledge circa 2010. Also there’s a difference in between graphs which depict estimates and graphs which are based purely on what’s been documented: Ethnic/Regional Origins of African Americans

        Like

  3. One of the major outcomes of my survey btw has been that the genetical contribution of Senegambia (according to AncestryDNA) seems to be less significant than what has been proposed by many historians. Especially when judging by the scores for the socalled “Senegal” region. On average this region “only” makes up 8% of the total (scaled) African breakdown for 350 African Americans. This includes the data i had available for Louisiana. The higher Senegambian share in the slave trade statistics for Lousiana are only valid when leaving out the much more numerous Intercolonial and Domestic slave trade (see also this blogpost).

    It seems very striking that sofar the highest percentage of Senegal i’ve seen for any African American has been “only” 25% (unscaled), which would be an absolute maximum among 350 other African American results. Scores of inbetween 20-25% already being quite exceptional for Senegal. While when going by the slave trade estimate of also about 25% (TAST, 2010) you might actually expect to see such percentages or even higher among many more African Americans. Afterall for regions like Cameroon/Congo, Ivory Coast/Ghana and even Benin/Togo scores of 30-50% are quite common, while for Nigeria i’ve even observed a few scores above 50% (unscaled).

    As i describe in more detail above there might be several explanations for this outcome. Possibly the gender ratio among Senegambians was less favourable for having much offspring when compared with other groups. And consequenty their bloodlines became relatively more diluted and dispersed. Also the more specific origins of Senegambians within the wider area of Upper Guinea might have been misinterpreted sofar. The socalled “Mali” region scores seem to be suggestive of these origins often being located moreso into the interior or to the south of Senegambia proper.

    Like

  4. I received this very insightful comment some time ago by PM. As it might also benefit others i will share it here as well.

    I’ve been thinking of other ways to interpret the Afro-AncestryDNA estimates, as i reflected on your “thinking out of the box’ in settling for conventional means to interpret data.
    So, i gathered the data (results) of members who are AncestryDNA tested and put in the Excel template that you’ve created. This time i looked at the non-zero estimates.
    In other words, i reviewed the regional scores in which AncestryDNA detected DNA similarity on all 40 runs. And it came to me that these are probably most robust or more stabilized estimates for any results. When, you re-examined the data, the highest non-zero clusters were Cameroon/Congo, Senegal, and Benin/Togo. Even though the Senegal set had the “lowest” percentages per individual, 6 out of 9 samples (66.66%) had non-zero ranges.

    I think this could be a very useful way of looking at the data indeed! I have done the same for the 350 African American results i have collected. I suspect it might indicate which African regional lineages are most widely dispersed within the African American genepool and also which ones are (relatively speaking) more rare. This was the outcome:

    Ivory Coast/Ghana: 3 people out of 350 scored 0%
    Cameroon/Congo: 3 people out of 350 scored 0%
    Nigeria: 7 people out of 350 scored 0%
    Benin/Togo: 14 people out of 350 scored 0%
    Senegal: 14 people out of 350 scored 0%
    Mali: 33 people out of 350 scored 0%
    Southeastern Bantu: 42 people out of 350 scored 0%
    South-Central Hunter-Gatherers: 103 people out of 350 scored 0%
    North Africa: 237 people out of 350 scored 0%

    Like

  5. Great blog, I know you wanted to focus on Sub-Saharan African lineages but I am curious to know what happens when North Africa *is* included? There was gene flow into North Africa from Sub-Saharan Africa during the Green Sahara period so cutting them off has always seemed arbitrary to me despite how North Africa is today.

    Like

    • Thanks for your comment Donn! In fact i did keep track of the North African scores for African Americans. If you take a look at this chart with my main findings, you will see it was the region with the lowest average. And like i said in my comment above 237 people out of 350 scored 0% for it. However there were still some noticeable outliers which i posted screenshots of, featured just above the Youtube videos.

      null

      I agree this geneflow from North Africa into Western Africa is very fascinating, especially the possibility of this taking place during the Green Sahara period. Although surely also during later timeperiods this would have occurred. I’m not sure to what extent AncestryDNA is able to pick up on these in part very ancient DNA signals though. You might be interested to see the AncestryDNA results of a Fula person for whom it shows up quite clearly with 16% North Africa, it can be viewed on this page.

      More background info and maps also to be found on this page: Sahel / Interior

      Like

  6. Felipe, thanks for following up with me! I went through your article again and saw where you did indeed account for North Africa, my mistake! Where the samples for “North Africa” including all of North Africa plus Egypt or was ot a small sampling pool?

    Yes very fascinating and not just gene flow from North Africa into West Africa but the *other* way around too, check this out. I can’t see the full article but the abstract looks promising,

    “The presence of sub-Saharan L-type mtDNA sequences in North Africa has traditionally been explained by the recent slave trade. However, gene flow between sub-Saharan and northern African populations would also have been made possible earlier through the greening of the Sahara resulting from Early Holocene climatic improvement. In this article, we examine human dispersals across the Sahara through the analysis of the sub-Saharan mtDNA haplogroup L3e5, which is not only commonly found in the Lake Chad Basin (∼17%), but which also attains nonnegligible frequencies (∼10%) in some Northwestern African populations. Age estimates point to its origin ∼10 ka, probably directly in the Lake Chad Basin, where the clade occurs across linguistic boundaries. The virtual absence of this specific haplogroup in Daza from Northern Chad and all West African populations suggests that its migration took place elsewhere, perhaps through Northern Niger. Interestingly, independent confirmation of Early Holocene contacts between North Africa and the Lake Chad Basin have been provided by craniofacial data from Central Niger, supporting our suggestion that the Early Holocene offered a suitable climatic window for genetic exchanges between North and sub-Saharan Africa. In view of its younger founder age in North Africa, the discontinuous distribution of L3e5 was probably caused by the Middle Holocene re-expansion of the Sahara desert, disrupting the clade’s original continuous spread.”

    –Eliška Podgorná et al.

    Annals of Human Genetics
    Volume 77, Issue 6, pages 513–523, November 2013

    The Genetic Impact of the Lake Chad Basin Population in North Africa as Documented by Mitochondrial Diversity and Internal Variation of the L3e5 Haplogroup

    http://onlinelibrary.wiley.com/doi/10.1111/ahg.12040/abstract

    Liked by 1 persoon

    • No problem Donn! The samples being used for the North African region are indeed very small in number, only 26. Ancestry doesn’t exactly confirm their background, as far as i know. But i highly suspect they are Algerian Mozabite samples taken from the HGDP database. (see also AncestryDNA regions)

      Fascinating paper btw! You might be interested to know that a black Tunisian who tested on both Ancestry and 23andme also is a L3e5. His results can be seen on this page (scroll down for it):

      African AncestryDNA results

      Like

  7. Hi, is it fair to say that african origins of Black Americans are distributed equitably because each regions seems to represent around 20% (except for the Southeast). There is not one region that exceeds 30 or 40% as in the Caribbean with Bight of Biafra and Ghana or South America with Angola.

    Even with your spreadsheet results, I tried to make combinations with the main African regions of deported slaves (although I admit it’s difficult to do) and the distrbution of percentages seems equal for the most part.

    Liked by 1 persoon

    • Hi Geoffrey, i would say “Nigeria”, “Ivory Coast/Ghana” and “Cameroon/Congo” do stand out somewhat when going by the group averages. And you will have to combine “Senegal” with “Mali” into 1 single Upper Guinea region to reach something close to a 20% share. But otherwise i agree that overall the regional proportions are quite balanced. In fact this also goes for much of the Caribbean and other parts of the Americas. It is only when you use a more restricted 3-way breakdown (into Upper Guinea, Lower Guinea and Central/Southeast Africa) that things tend get more pronounced. See below chart which is discussed in greater detail on this page i just published:

      Afro-Diaspora AncestryDNA results: A Comparison

      statsplus

      Like

  8. Hi Felipe:

    Awesome blog…my results were surprising to me and I wanted to get your thoughts. My folks on my moms side were from Georgia and on my dads side from Alabama.

    My results were

    African 93%
    Nigerian 44%
    Ivory Coast/Ghana 17%
    African Southeasern Bantu 13%
    Cameroon/ Congo 8%
    Benin/Togo 6%
    Trace Regions 5%

    America
    Native American 2%

    Asia 1%
    Asia Central <1%
    Asia East <1%

    Europe 4%
    Iberian Penensula 2%
    Ireland 1%
    Europe West 1%

    Liked by 1 persoon

    • Hi Maisha, thank you for your message and congratulations on your results! As i have mentioned throughout my blog AncestryDNA’s socalled “Ethnicty Estimates” can provide very valuable insight but only within a (sketchy) regional framework. You will need additional context/info to pinpoint any specific ethnic details or also combine with other DNA results, especially any African matches you might have.

      Having said that the striking thing about your breakdown is of course the 44% “Nigeria” which represents almost half of your total African ancestry (44/93=47%)! This gives your breakdown a more pronounced regional focus than the usually more fragmented standard. Your relative “Nigeria” score (47%) is far above the “Nigeria” average for African Americans (20,3%). However i did observe similar high scores for the “Nigeria” region among my African American sample group (n=350). You would be among the top 25 of my sample group with the highest “Nigerian” contribution.

      It is somewhat ironic that even a couple of Nigerians who tested on Ancestry showed less “Nigeria” than you did, Instead the “Benin/Togo” and Cameroon/Congo” regions were more important for them (see also section 4 above). This might eventually provide you with a clue as to which ethnic groups within Nigeria might be more likely as being part of your heritage. But this is also very much speculative at this stage. I will feature about a dozen of Nigerian AncestryDNA results shortly on this page:

      Nigerian AncestryDNA results

      A second score which is above average for you (when compared with my African American sample group (n=350) is the 13% “Southeastern Bantu”. As described above this region is somewhat difficult to interpret at this stage and several ancestral scenario’s might apply. However it might very well be that your 8% “Cameroon/Congo” should also be taken into account and together these regions might represent an ancestral connection to Congo/northern Angola.

      Like

  9. Hi Felipe,

    This was a great read. I recently got my results back

    My results
    Africa 76%

    Southeastern Bantu – 33%
    Nigeria – 23%
    Mali 10%
    Africa North 5%
    Senegal 4%
    Africa South-Central Hunter Gatherer <1%

    Europe – 7%

    Middle Eastern – 17%

    I live in California, but my family is primarily from Louisiana with my maternal grandfather's mother being from Mississippi & my paternal grandfather's family being from Texas.

    I've read that Southeastern Bantu migrated from Nigeria & Congo, so I'm guessing my a good portion of that percentage probably comes from Bantu people who stayed in Nigeria? I'm also quite confused by the high Middle Eastern percentage being that no one knows of any Middle Eastern members in my family's history. I assumed that the 5% North Africa may have been correlated with that, but I see according to your research its probably more in line with Mali or Senegal.

    What do you make of these results?

    Thank you.

    Liked by 1 persoon

    • Hello Eddie, thanks a lot for your comment! To be frank for an African American you show highly unusual results. Your Southeastern Bantu score of 33% is the second highest i have observed sofar. Actually among my original survey of 350 AA results, the greatest amount of SE Bantu was “only” 25% and only two persons had this region as the biggest one within their African breakdown. Since then i have literally seen hundreds more of AA AncestryDNA results and again only a very small number have SE Bantu as number 1 region and typically the scores are rather subdued, with the higher ones only reaching the lower 20% range. However i did see one AA result with an astonishing score of 48% SE Bantu. So i suppose it’s fair to say a SE Bantu score of above 30% is very atypical although not impossible for an AA.

      Now it gets more trickier to establish any likely origin hiding behind this socalled SE Bantu region, which is spread out across a very great part of Africa. As i have described above on this page, generally speaking for AA’s and other Afro-Diasporans i think that despite the labeling SE Bantu is most likely to be indicative of Southwestern Bantu origins, especially from Angola, eventhough origins from Southeastern Africa, especially Mozambique & Madagascar still also remain a possibility.

      Your case is extremely intriguing however because in addition you also score 17% Middle Eastern and 5% North Africa. These kind of amounts are truly exceptional for African Americans, who tend to not show any noticeable percentages above tracelevel for these regions. And especially not double digit ones like your 17% Middle Eastern. To be more precise the combination of high Middle Eastern + high Southeastern Bantu + above trace level North Africa is exactly what you would expect for Northeast Africans. See for example the Eritrean and Somali results featured on this page (scroll down for it):

      African AncestryDNA results

      So without any further clues i would say that your profile seems to be suggestive of a high degree of Northeast African ancestry. I would advise you to further investigate the possibility of having recent non-AA ancestry. To be clear I do not mean to lead you into only one research avenue as there might still be other logical ancestral explanations for your results 😉 However it may turn out i would really appreciate it if you can keep me up to date when you find out anything new. I would also love to see a screenshot of your results! If you’re okay with it you could send me (FonteFelipe) a sharing invite to view your ethnicity estimates. This link tells you how it’s done:

      https://support.ancestry.com/s/article/Sharing-my-Full-AncestryDNA-results-1460088592896-2580

      Like

      • Wow, this was very helpful. I have been told quite a few times I look East African – so I guess I shouldn’t be surprised. According to my grandmother her grandfather (my great great grandfather) was half Turkish, which is where the Middle Eastern could come from. However, the percentage seems to be too high since he was only half and was so long ago.

        Like

      • Hi Eddie, thanks a lot for sending the link! Very interesting about your half Turkish gg grandfather. But i agree that his genetic contribution (inbetween 3-6% i suppose) would be too small to account for your overall results. Really it seems to suggest a more recent connection. Have you tried searching your DNA matches yet for birth location? You might type in places like Ethiopia, Somalia, Sudan, Kenya etc. To see if any close DNA cousins from East Africa might pop up. You could also try Turkey and other Middle Eastern countries. Let me know if this gives you any further clues.

        Like

  10. This is a great overview. I was born in Florida, with my birth mother’s family coming from Georgia; no information on my birth father.

    My results were:
    91% African
    -32% Cameroon Congo
    -18% Benin Togo
    -13% Senegal
    -12% Bantu
    -10% Ivory Coast Ghana
    -3% Mali
    -2% Nigeria
    -Less than 1% Hunter Gatherers

    Now for the rest:

    Less than 1% Asia (trace region)
    -Asia South

    7% Europe (all trace regions)
    -2% Europe East
    -1% Europe West
    -1% Ireland
    -Less than 1% Iberrean,Scandinavia, and Great Britain

    Less than 1% Pacific Islander
    -Polynesia

    –Two observations on these results. I am very surprised how low the Nigeria percentage is compared to the other videos I’ve watched. I do have 8 of the 9 regions in my results, of course, but 2% seems quite low. Is there anything to be “read” from my results?

    –I’m surprised that all the rest are trace regions, including Europe, and that there is less than 1% Great Britain as I see that quite often as the European contribution.

    What do you think? Are my results pretty standard or unusual at all to you? Any insight? Thanks~

    Like

    • Hi Sarah, your results are an unique reflection of who you are so in that way they are always going to be special 😉 But yeah otherwise i’d say your breakdown is pretty much in line with the variation i observed among other AA results. I have not really read a great deal about it but I have a feeling that many AA’s from Florida tend to have family origins from either GA or SC. Both recent ones, such as you mention, but also dating back from several generations ago to the 1800’s and the 1700’s. It’s still under preparation but i will eventually create a separate blogpage based on results from those states. I think your own results will probably be an even better fit for their averages.

      Your Nigeria score is indeed low but actually i have observed this for many other AA results. It’s just part of individual variation. Then again you could still have additional Nigerian origins hiding under your socalled “Cameroon/Congo” and ” Benin/Togo” scores. Afterall actual Nigerians when tested by AncestryDNA also have those regions in their results. See also section 4 of this page.

      As for your European breakdown, i haven’t really systematically kept track of it when doing my survey. But i did pay notice. From what i’ve seen it’s indeed Great Britain, as well as Ireland, Scandinavia and West Europe which appear most frequently as main regions for AA’s. Which is as expected. However when it comes to the trace regions it’s usually more random and i would say much less reliable or better said not to be taken literally. If you click on each of these regions as well as the description for trace region it self it already says that AncestryDNA is less confident about classifying these segments of DNA and the range of estimates will include zero%. If you’re looking for more clarification about your European ancestry i would advise also testing with 23andme as I find that they usually do a much better job with the smaller percentages.

      Like

  11. One of the videos talked about people who believed there to be Native American ancestors but there was nothing in their DNA that showed up. Maybe this could be from african slaves running away and joining Native Americans, marrying other run-away-slaves, and having children identifying as Native American?
    From someone not at all American 🙂 Maybe I misunderstood.

    Like

  12. Great blog! This inspired me to find out my DNA results as well. I have a Ghanaian father (Asante, a subgroup of the Akan) and an African American mother. These are the results I got:

    Africa 95%
    -Ivory Coast/Ghana 60%
    -Cameroon/Congo 13%
    -Benin/Togo 13%

    Trace
    -Africa Southeastern Bantu 4%
    -Senegal 3%
    -Mali <1%
    -Nigeria <1%

    Europe 5%
    -Great Britain

    My mother is African American (on both sides) and this is what she got:

    Africa 92%
    -Cameroon/Congo 29%
    -Ivory Coast/Ghana 21%
    -Benin/Togo 17%
    -Africa Southeastern Bantu 11%
    -Senegal 10%

    Trace
    -Nigeria 3%
    -Mali 1%

    Europe 8%
    -Great Britain 7%
    -Iberian Peninsula <1%

    Liked by 1 persoon

    • Hi Rachel, thanks a lot for sharing those fascinating results! A regional score (for just 1 single region) above 50% is very exceptional for African Americans from what i’ve seen. However given your background it makes perfect sense that you would show 60% “Ivory Coast/Ghana”. Actually you might also be partially Ghanaian by way of your mother. Given that you would have inherited at most 50% “Ivory Coast/Ghana” from your Ghanaian father (even when most likely he also will score some Trace regions). The 21% Ivory Coast Ghana your mother scored is very close to what i found to be average for the 350 AFrican Americans i surveyed. To cover all grounds it might still also include ancestral connections with either Liberia or Sierra Leone. But i think there is a high likelyhood that there might also be Akan lineage included. That 21% Ivory Coast/Ghana for your mother will afterall be the genetical sum inheritance of several ancestors who lived mostly in the 1700’s (unless your mother also happens to have a recent Ghanaian connection😉

      Like

  13. Hello- I absolutely love your blog. I never tested on Ancestry, I tested with Family Tree and uploaded to GedMatch and DNAland. Could you help me make out the following .

    FamilyTreeDna – 93% African/ 7%European = { West Africa 83%, East Central Africa 5%, South Central Africa 4%}

    DNAland – 93% African /5.8% West Eurasion/ 1.4% Ambiguous = {Lower Niger Valley 38% Yoruba and Esan, Mende/Akan 21% Sierra Leone and Senegal River Valley 17% Mandenka in Senegal and Gambia in Western Gambia, East African 11% Bantu, Kikuyu, Luhya, Luo and Masai in (Webuye, Kinyawa and 4 other sites) Kenya; Datog in Tanzania and Luhya in (Kenya) Webuye, Mbuti 2.1% Mbuti (Pygmy) in Congo, 3.6 Ambiguous African.

    West Eurasion 5.8% , Ashkenazi 2.1%, Ashkenazi Jew in Poland and Ashkenazi Jew from East Europe especially Lithuania (expat in Baltimore MD), Sardinian 1.1% Sardinian in (Sardinia) Italy, Ambiguous West Eurasian 2.6% West Eurasian is a very general category containing Arab/Egyptian, Ashkenazi/Levantine, Central Asian, Northeast European, South Asian, South European, Northwest European, Southwestern European and Central Indoeuropean

    My family is from Alabama, Florida and Georgia. Oral history, family from the Caribbean islands of St. Croix … also haplo is L2a1

    Any light you could shed, I’d appreciate it.
    (attempting to post, its not uploading, I apologize if it post multiple times)

    Liked by 1 persoon

    • Hello Sarah, thank you for the comment, glad you like the blog! To be frank i do not know a great deal about the FamilyTreeDna test. So i may not be the best person to ask. I do know that regional labels such as “East Central Africa” should always be carefully scrutinized for Afro-Diasporans. First of all by finding out which kind of sample groups are being used, secondly by comparing with results of people with confirmed African background and lastly by counterchecking with documented history. This is what i have been trying to do for the AncestryDNA test.

      I have seen quite a few DNAland results, but overall i wasn’t really convinced by their analysis. I find it confusing and potentially misleading in many cases. Also for people with confirmed and non-complex background they often get it very wrong. Not even to be explained by ancient migrations or genetic overlap. In these aspects i find AncestryDNA easier to interpret, although surely it also has its own limitations.

      What i do like about DNAland is that they provide exact information about the sample groups they are using to compare your DNA with. It is crucial though to be aware that with any kind of DNA testing their database of reference populations will by default be limited in scope. And this sample input directly determines the way your results output will look like.

      For example your socalled Lower Niger Valley score is based on the genetic similarity they found between your DNA and their Yoruba and Esan samples from Nigeria. However this doesn’t mean that you actually share recent (within last 500 years or so) ancestry with these specific ethnic groups! If DNA land was instead using only Igbo samples from Nigeria it might very well have been the case that your socalled Lower Niger Valley score would still have been about the same. Also because DNAland – unlike AncestryDNA – doesn’t have any samples available from Ghana, Togo, Benin and Cameroon, it means that any part of your DNA which is to be linked with these countries will instead be described by DNAland as mostly Lower Niger Valley and probably also some socalled Mende/Akan. Again only a genetic similarity is being picked upon, in the absence of better fitting samples!

      From what i have seen most African Americans score high amounts of this “Lower Niger Valley” category, which should however be interpreted very broadly! If you check the map provided by DNAland it shows that West Africa is divided in three parts with this socalled Lower Niger Valley comprising the greater part, extending beyond Nigeria’s borders!. AncestryDNA definitely has an edge here because besides socalled “Nigeria” it also has on offer the “Cameroon/Congo”, “Benin/Togo” and “Ivory Coast/Ghana” regions.

      Interestingly your Senegal River Valley score is quite elevated compared with what i have seen for other AA’s. I actually do find this category on DNAland quite useful as i have counterchecked it with the results for Cape Verdeans who are known to have primary Senegambian origins. Your DNA is being compared with most likely the same Mandenka samples as being used by AncestryDNA. Eventhough you could actually have other ethnic origins besides Mandenka, i would consider this score as strongly suggestive of genuine Upper Guinean roots.

      Your socalled “East African” score is also quite noticeable. I have seen comparable scores for other AA’s as well. However i would take this regional labeling with a HUGE grain of salt. This particular outcome was obtained by comparing your DNA with Bantu samples from Kenya and Tanzania. However if DNAland had instead used Bantu samples from Angola and Mozambique it might very well have been a better match. As this would actually correspond with plentiful documented history and cultural retention while any significant East African connection with the Afro-Diaspora in the Americas is lacking all together.

      Like

  14. Thank you for this explanation. About a week ago, I ordered the AncestryDNA test. Once I receive the results back I will upload them to your blog.I think between my oral history, the various dnas test and my haplogroup, I should have even a clearer understanding of the actual countries. I cant wait to share my AncestryDNA results with you.

    Liked by 1 persoon

  15. Hi Fonte Felipe,

    don’t you think the Ivory Coast/Ghana category is more so due to the Sierra Leone/Liberia connection rather than Ghana for Americans ?

    I saw more Americans with Mende or Kru (Liberia) dna results rather than Akan dna match. Even I include the early Akan slaves from Barbados in South Carolina, their influence doesn’t seem big like the Mende, Igbo or Kongo influence. Akan people are usually more associated with Caribbeans.

    I feel the IC/G category is a bit overestimated because it can regroup origins from Ghana, Ivory Coast, Sierra Leone or Liberia which is around 39 % of the slaves origin in the Carolinas/Georgia for example while other category can only indicated two countries.

    Like

    • I agree that the socalled “Ivory Coast/Ghana” region as reported for Aframs is due to not only Ghanaian ancesty but also Liberian and Sierra Leonean ancestry. While any genuine Ivorian ancestry will be least likely. Just going by the Slavevoyages Database estimates you might indeed assume that Liberia/Sierra Leone would be somewhat more significant than Ghana. Especially for South Carolina.

      Source: slavevoyages.org

      But as you also rightly mention there is also the influence of Slave Trade by way of the West Indies which is likely to have been more so in favour of the Gold Coast. Another aspect to take into consideration is the time period these captives arrived. Those West Indian captives as well as early arrivals in Virginia (where Gold Coast trumps WW Coast + Sierra Leone) would have had a headstart in having offspring. Many of their descendants eventually getting scattered across the Deep South due to Domestic Slave Trade. Personally i have a hunch that the breakdown for Virginia (eventhough less in numbers) might be more representative for the average AA than the South Carolina breakdown.

      Source: www.slavevoyages.org

      All in all i suspect that there’s not a dramatic difference between the likelyhood of “Ivory/Coast/Ghana” denoting either Liberian/Sierra Leonean origins rather than Ghanaian ones for Aframs. I think the odds pretty much even out. The best way to get any more certainty is finding any locally born matches from these countries. Even when actually they will only represent one single familyline, which might imply a DNA contribution of about 1% on average. Someone who for example scores 21% “Ivory Coast/Ghana” might very well have 7 ancestors each from both Ghana, Liberia and Sierra Leone all included in that 21%.

      Interestingly I have had the opportunity to look through the DNA matches of two Liberians. They both had about 25 pages of DNA matches (each page having 50 matches). And i was amazed by the number of not only African American matches but also matches across the Caribbean and in particular from Guyana. Of course when receiving a Sierra Leonean or Liberian match you will have to be very careful and check whether they do not have any partial Krio ancestry. Because in that case the ancestral connection might actually be reversed! Both Liberia and Sierra Leone having absorbed freed ex-slaves from the US and the West Indies. See also:

      Americo Liberians
      Krio from Sierra Leone

      Like

  16. I love the work that you are doing!
    Both of my parents are African Americans with Georgia roots. I had the ancestry DNA testing done by Ancestry.com

    My results:
    Africa = 93%
    Ivory Coast/Ghana 25%
    Nigeria 22%
    Cameroon/Congo 13%
    African Southeastern Bantu 12%
    Benin/ Togo 10%
    Mali 6%

    Trace Regions:
    Senegal 3%
    Africa North 1%
    Africa South-Central Hunter-Gatherers <1%

    European = 7%
    Finland/Northwest Russia 3%
    Ireland 1%
    Great Britain 1%
    Europe West <1%
    Iberian Peninsula <1%

    I will be having both of my parents tested soon.

    Like

  17. Dear Sir;

    I can’t begin to thank you enough for the work that have and continue to do. The information and data that you have provided has been far more useful to me as the ‘resident genealogist’ for my family than what I have found on AncestryDNA as it relates to understanding our African origins.

    Below is the result of my AncestryDNA test for your use in your studies. I am African American. My maternal line is from South Carolina, I have found my maternal lineage back to the early 1800’s and have a high confidence level that most of this line were brought to and lived in South Carolina until many migrated north to New York, New Jersey and Pennsylvania.

    My paternal line is from Trinidad, W.I. and Barbados. I have no other information indicating that anyone from this side of my family lived on other islands before migrating to the United States.

    Thank you again for your work and I hope my data is helpful to you.

    AFRICA — 92%
    Cameroon/Congo — 37%
    Nigeria — 17%
    Ivory Coast/Ghana — 14%
    Mali — 8%
    Senegal — 8%
    Benin/Togo — 6%
    African Southeastern Bantu – 1%
    Africa North — 1%

    EUROPE — 5%
    Finland/Northwest Russia – 1%
    Great Britain — <1%
    Iberian Peninsula — <1%
    Scandinavia — <1%
    Italy/Greece — <1%
    Europe West — 0%

    AMERICA — 1%
    Native American — 1%

    WEST ASIA — 2%
    Caucasus — 2%

    Best Regards,
    AL

    Liked by 1 persoon

    • Dear Ambrozine,

      I am very gratified and grateful to hear that my blog has been useful to you! I also greatly appreciate your willingness to share your results. However for the sake of my research i can only use results which are representative for just one country/island (going back 3 generations). So in that way regrettably i cannot make use of your data. However your results remain very fascinating!

      Given your family background you might be interested to know that i intend to publish the following blogposts shortly (in order of planned publication date):

      Tutorial on how to systematically find African DNA matches on Ancestry.com
      AncestryDNA Results from the West Indies (updated version incl. new data from Trinidad & Barbados)
      – AncestryDNA Results from South Carolina & Georgia (might take some months still to finish)

      Best of luck in your ancestral quest!

      Like

  18. Fonte Felipe- Its me, Sarah again. As promised, I told you I would share my Ancestry DNA results once I received them. Here are the results:

    African 92%
    Cameroon/Congo- 34%
    Senegal- 20%
    Ivory Coast/Ghana- 17%
    Benin-Togo- 8%
    Nigeria- 6%

    Low Confidence Regions
    Mali- 3%
    Africa- South Central- 2%
    Africa North <1%
    African Southeastern <1%

    My other 7% was a myriad of countries in Europe which ALL came up low confidence regions, so I did not bother to post.

    I'm a little confused again, as most results from Gedmatch were Yoruba, Bamoun, Hausa, Igbo, Kongo…which I assumed based on this my Nigeria score would come out higher than 6%. Once I think I understand, then some new countries get thrown into the mix. Any feedback you have will be helpful.

    Again, thank you for your great work, as you can see you are helping alot of people. I think I speak for the majority that, we appreciate your time and insight.

    Like

    • Hey Sarah,

      Thanks a lot for your message! It’s great to know that my blog can be helpful. I am intending to blog very shortly about a tutorial on how to systematically detect African maches on Ancestry. So keep an eye out on that! As i believe that African DNA matches can deliver some of the much wanted specification people are seeking. Even when these individual matches will only cover a *single family line* out of potentially hundreds. So they won’t provide answers to all your questions about your *entire ancestry*.

      Really interesting to see your AncestryDNA breakdown. Your previous results seem to be in line on quite a few aspects. Especially the total amount of African (92/93%) as well as your significant Upper Guinean roots being detected both by DNA land (17% Senegal River Valley) and on Ancestry (20% “Senegal”). These kind of scores are not very common for AA’s from what i have seen. The maximum Senegal score on Ancestry in my survey is still “only” 25%. So your score is really indicative of having an above average connection with the Upper Guinea region (Senegal to Sierra Leone). When you receive any African DNA matches from this area it will help you to pinpoint it even more closely.

      Your 17% “Ivory Coast/Ghana” score on Ancestry also aligns well with the 21% Mende/Akan on DNAland. Again any African DNA matches will clarify (to some extent) how to interpret these regional scores which could link you to ethnic groups from not only Ghana but also Sierra Leone and Liberia (and also any combination of these groups).

      Your socalled “Nigeria” score on Ancestry is indeed rather low even when in fact many African Americans might score similar amounts. Keep in mind though that the labeling of ancestral categories is always to some degree arbitrary or imprecise. Actually you could still have a greater degree of Nigerian lineage but it might be “hiding” under your socalled “Benin/Togo” and “Cameroon/Congo” scores. Because Nigerians are so genetically diverse AncestryDNA needs these additional (and neighbouring) regions to describe Nigerian DNA. Still this outcome is an improvement on the rather misleading blanket category of socalled Lower Niger Valley on DNA Land. For more details see:

      Nigerian AncestryDNA results

      Gedmatch only uses a limited number of African ethnic groups to compare your DNA with. This database restriction is in fact inherent to any DNA testing company. For example a few years ago it was very common to only use Yoruba samples which sort of functioned as a standin/proxy for generic West African ancestry. However many people were misled by this circumstance in believing they had genuine Yoruba origins when in fact only a mere genetic similarity is being reported (in absence of better fitting samples). That’s why i always caution against taking these ethnic predictions from admixture analysis too literally. Your Bamoun and Kongo *similarity* on Gedmatch is however very consistent with your prominent Cameroon/Congo score on Ancestry, as both of these groups are in fact from Cameroon and Congo. Also your family origins being from AL/FL/GA might be supportive of a greater Central African contribution as it is known that historically the Congolese and (northern) Angolans had a major presence in South Carolina where much of your earliest family origins within the US might be from if you do family tree research into the 1800’s or even 1700’s. See also:

      Ethnic Origins of South Carolina Runaway Slaves
      “Cameroon/Congo” = moreso Angola/Congo for Diasporans?

      The one thing that seemed conflicting with your previous results is the mere <1% Southeastern Bantu on Ancestry versus the 11% East African Bantu on DNA LAnd and 5% East Central Africa on FTDNA. Personally i think Ancestry's analysis will be much more reliable in this case. As i already mentioned earlier a prominent Central African score actually corresponds with plentiful documented history and cultural retention while any significant East African connection with the Afro-Diaspora in the Americas is lacking all together.

      Like

  19. Thank you for this thorough breakdown. It is so exciting. I was very pleased with my results, of course it would have been nice to be 50% or more since my overall percentage of Africa is large, but instead my majority was spread between 5 countries. However, through your explanation and research that I’ve done, I find that just because it is spread in various countries, that could account for migration patters and everything else. Anyway, thank you for your passion and the research that you provide, you are bringing clarity to many who wish to know more of their history!

    Liked by 1 persoon

  20. My parents are born in England the next hundred years they were in slavery in Jamaica straight from Africa I believe I’m third generation jamaican here are my results unsurprising as I found my dad’s side were jamaican Maroons which I believed kept my african heritage high as my mom is from Creole descendants according to records I’m 95% african to see my journey of 300 years have a look at my blog. Your blog is amazing and I will be also trying g to analyse the details you have gathered.

    Liked by 1 persoon

    • Thanks for your comment! Your own blog is very fascinating! I had a look at your post where you describe your AncestryDNA results. Very interesting! Given your Jamaican background you might be interested to know that i have collected 100 Jamaican AncestryDNA results right now and i intend to blog about my survey findings shortly. Here ’s a link to my online spreadsheet which has all their results. I also noticed your blog post about your Ghanaian (Ga) DNA match. You might want to try out my tutorial on how to systematically detect your African DNA matches on ancestry. Here’s the link for it:
      How to find those elusive African DNA matches on Ancestry

      Liked by 1 persoon

  21. Interesting blog. I just received my results. My mom is from Louisiana and my father is from Trinidad. Below are my results:

    African 67%

    *Nigeria 25%
    *Mali 16%
    *Benin/Togo 8%
    *African North 3%
    *African South-Central 3%
    *Senegal 6%
    *African Southeaster 3%
    *Ivory Coast/Ghana 2%
    *Cameroon/Congo 1%

    Asian <1%
    *Asian South <1%

    Europe 31%

    *Europe West 17%
    *Iberian Peninsula 6%
    *Ireland 4%
    *Italy/Greece 2%
    *Great Britain 1%
    *Finland/Northwest Russia <1%

    America <1%
    *Native American <1%

    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