How to find those elusive African DNA matches on Ancestry

Zoom in Gambia


Ethnicity summary of my most likely Gambian DNA match on

Gambia Match

Because of the new Compare tool I have now (November 2018) also learnt that his main region is 81% “Senegal” (pre-update). I cannot see the remaining part of his breakdown though.


A couple of months ago I found my very first mainland West African DNA cousin on Judging from his name and ethnicity preview he is a Mandinga from Gambia. I was exhilarated!  I had never been able to find a mainland West African match before. Even when I took my first DNA test with 23andme already in 2010! In other words I had to wait seven years for it! Inspite that I haven’t had any meaningful contact yet this genetic connection is still very valuable to me. As I am of Cape Verdean descent this finding seems very appropriate and in line with my primary Upper Guinean roots (see Top 20 Ethnic Roots for Cape Verdeans).

Many people seeking to trace back their African ancestors are very eager to be connected to a specific place and person within Africa. They hope to achieve this by finding a DNA match from the continent. But just like me they are having a hard time to do so. In the beginning this was mostly because only very few Africans were included in the customer databases of companies such as 23andme or Fortunately this has been changing lately because a quickly growing number of Africans or rather African migrants and their children are taking a DNA test. This increases the likelihood of receiving African DNA matches. However many people still find it difficult and/or tedious to sort out their DNA matches. In this blog post I will therefore describe a method which enables a systematic, comprehensive and time saving detection of your “100% African” DNA matches on

If you continue reading you will find:

  • Some considerations on how to interpret your African DNA matches
  • A step by step tutorial on how to sort out your African DNA matches


Notes of Caution

It is probably no exaggeration to say that for many Afro-Diasporans getting connected with an African DNA cousin represents nothing less than a dream come true. It is considered a highly prized outcome. As a genealogical research reward in itself but often also on an emotional or even spiritual level. Understandably this sometimes leads to wishful thinking and tunnel vision whereby DNA results are not critically assessed. Without wanting to rain on anybody’s parade I will list some considerations which might be helpful to improve the interpretation of receiving African DNA matches. Naturally this section is not meant to be exhaustive.

  • Don’t jump to conclusions. The first thing to take into consideration is establishing whether your African match is indeed genuine or “identical by descent” (IBD) and not just a socalled false positive or random “identical by state” (IBS) match. Getting your parents tested or any other relatives of older generations will increase your ability to be more certain about any given reported DNA match. Also the size of the shared DNA segment will be indicative. It is well advised to carefully read the Predicted Relationship Info given by and especially the confidence score they will assign to your African DNA match.
  • Don’t be misled by the myth of “single tribe origin”! This is a line of thought which seems to be psychologically comforting for many people. It is however not based on historical reality. Practically all Afro-Diasporans will have dozens if not hundreds of relocated African-born ancestors. Depending on your background these ancestors would have been born mostly in the 1700’s but in some cases also in the early 1800’s, 1600’s or even 1500’s. Statistically speaking it is nearly impossible for all those people to have been from just 1 or even just a handful of ethnic groups. Instead, on average an Afro-Diasporan will have various ethnic origins from several places in between Senegal and Mozambique. The particular mix and proportions will vary per individual but not so the fact that a Diasporan’s DNA is basically a melting pot of many different ethnic lineages. It will never reflect just one single ancestral “tribe”. Still your best bet to confirm one particular ethnic lineage (out of many others) is by finding an African DNA cousin with verifiable background.
  • The MRCA (most recent common ancestor) shared between you and your African DNA cousin will not per se be of the same ethnic background as your African DNA match. Without any 100% complete and accurate paper trail in place (or oral traditions to the same effect) you should not be surprised that your African match might also be ethnically mixed further down the line. If not recently than perhaps several generations ago and even beyond family recollection. The extent of inter-ethnic unions taking place within Africa itself is often ignored. There are almost always several possible ancestral scenario’s to consider when you get “matched” with someone. None of them to be ruled out in advance. Careful interpretation is therefore essential.
  • Due to dilution of shared DNA segments across the generations there might be a higher likelihood of African MRCA’s from the early 1800’s and late 1700’s among your DNA matches. While there might be a built-in bias against African ancestry which is to be traced back to the early 1700’s, 1600’s or even 1500’s. After all due to Ancestry’s matching threshold DNA matches with smaller shared segments will not be shown. I am actually in favour of such a conservative approach as I believe that it’s best to avoid IBS matches and the false hope they may evoke. It is of course always a blessing to connect with any family line certified in a robust manner. But for those Afro-Diasporans who happen to have relatively more distant African roots this circumstance might very well lead to a distorted overview of their African heritage. Especially when the regional origins of their African ancestors born in the 1700’s/1800’s were markedly different from their African ancestors born in the 1500’s/1600’s.
  • Certain African countries will tend to be overrepresented in Ancestry’s customer database. Notably English speaking countries such as Nigeria and Ghana. While Portuguese or French speaking countries such as Angola and Benin are underrepresented for the most part. This is mostly due to these countries’ migrant presence in the US/UK (see this graph for biggest groups in USA in 2015). Although by now I actually have personally seen African Ancestry profiles from almost all countries on the continent. So overall the coverage in Ancestry’s customer database is still already quite impressive. But just not so proportionally speaking. This situation creates more chances of being matched with for example your Nigerian lineage but doesn’t potentially invalidate that you might still also have Beninese or Angolan ancestry in addition.
  • Eventhough the wait might seem to take forever Big Data will eventually also be available for African DNA. Whether provided by paying customers of African migrant descent or through academic sampling within Africa itself. Because of upcoming scientific advances your outlook on your African origins might very well be modified a few years from now. Instead of receiving just one single African DNA match (if you’re lucky) you will then be guaranteed to get many pages containing perhaps hundreds of African DNA matches! Comparable to the situation of people of European descent right now. This could actually create initial confusion as you are bound to get DNA matches from a myriad of African countries. From personal observation on 23andme’s former CoA (Countries of Ancestry) tool Europeans typically received DNA matches from allover their continent (to their great surprise). For the most part not due to any genuine links within a genealogical timeframe but rather due to IBS and very ancient shared origins from thousands of years ago. Hopefully promising new features such as the Genetic Communities will be further developed to aid in solidly analyzing any of the arising patterns.
  • Look at the bigger picture!  Admixture analysis (based on your autosomal DNA) is often misunderstood or even mocked. Unjustly so as I have found AncestryDNA’s Ethnicity Estimates to be very insightful as long as you know how to correctly interpret them and are aware of inherent limitations. My assessment is based on the many AncestryDNA results of native Africans I have seen. Which were usually in alignment (broadly) with their verifiable background (see this overview). Also my survey of Afro-Diaspora results was largely a confirmation of historically documented African origins for each nationality.
  • Your ethnicity results will always be relevant to put things in perspective. If only to be able to (roughly) distinguish between major sources of ancestral origins versus minor lineage. Finding one single African DNA match can of course be very valuable but it will still represent only one isolated connection with Africa out of potentially hundreds others. Individual DNA matches – no matter how insightful – will not be able to fill in all the missing pieces of your entire ancestry. Therefore don’t put all your eggs in just one basket. If you want your research to be all encompassing you will want to rely on the complementarity of admixture analysis and DNA matches. The truth has many angles. Explore them all!

For more detailed discussion follow these links:



Up till recently it was extremely difficult to find African DNA matches as there were simply very few native Africans available to get matched with in Ancestry’s customer database. It is a true blessing that this is rapidly changing as more and more Africans are starting to get DNA tested themselves. This presents a unique opportunity to learn more about your African roots (at least along a few selected family lines) and connect with your African DNA cousins.

For some people the following tutorial might be redundant as they were already aware of advanced filtering on Excel or because they have devised other effective ways to detect African DNA matches. This blog post is however aimed at people who are currently missing out on African DNA matches   because they simply don’t know where to start. Or because they feel over-awed about scanning through all their DNA matches or unsure on what exactly to pay attention to.

So how does it work and what do you need? 

  1. Install the DNAGedcom Client
  2. Some basic Excel skills (for first-timers: google is your friend!)
  3. Follow this step by step tutorial

And basically you are all set! Of course no guarantees given that this will automatically result in finding African DNA matches 😉 This method is merely meant to ensure African matches do not remain undetected. However many people will find that Ancestry’s customer database is still lacking in Africans to be matched with. Don’t despair! Just regularly check your DNA matches according to this filtering method and eventually your efforts will be rewarded.

Just as a p.s. the screenshots used for this tutorial are obviously a reflection of my personal system settings. The display might be slightly different depending on your own settings. For privacy reasons I have cut out the names of my matches.



This tutorial has been modified several times since I first published it in 2017. In order to ensure a continued or even improved detection of African DNA matches. Mostly triggered by updates on Ancestry but also through helpful tips from the comment section. Scroll all the way down for a log of all the changes made.

As an alternative you can also watch this demonstration video I made for Rootstech 2021:



Do a fully automated scan of your DNA matches by using DNAGedcom.

***(click to enlarge) 

DNAGedcom 2

***(click to enlarge) 

DNAGedcom 4


  • Open the DNAGedcom Client and click on the “Gather” tab, displayed above
  • Choose Ancestry
  • Type in your Ancestry login name and password
  • Click on “Login”
  • Make sure that you only tick the box next to “Gather Ethnicity”. Leave everything else unticked. The other options are not needed for this purpose.
  • Select the desired profile and click “Gather DNA Data”
  • Downloading the DNA matches will take a while but should still be pretty fast. Personally I must have scanned more than a 100 profiles shared with me by now. And the scanning is almost always finished within an hour.
  • For any troubleshooting consult the DNAGedcom webpage or join its Facebook group


Open the file of your DNA matches when the scan is completed. You will be able to find the file in the export folder designated in Options on the first view of the DNAGedcom Client. The original file will be in CSV format (Comma Separated Values). You will need to save the file in Excel (Workbook) format to proceed with the next steps.

  • Go to “File”
  • Select “Save As” and choose your preferred Excel format
  • With my system settings the columns are shown separately right away after saving the file in Excel format.  If for some reason this should not be the case for you:
  • Apply “Text to Columns” to your DNA matches data. See step 4 in the first version of this tutorial.
  • For any troubleshooting just google: “convert CSV into Excel” or “Text to Columns”

***(click to enlarge)

Stap 2



Change the display of your DNA matches data. This will be done by only leaving in the most relevant columns. The other columns will be hidden from view. Because you are not deleting these columns you can always decide to unhide them at a later stage when required.

  • First check if the scanning has indeed been fully complete. Scroll down to the last row and check if all rows are filled in with data from columns A to X. If not repeat step 1.
  • Select all columns you like to be hidden from view
  • Right-click and select “hide”

***(click to enlarge)

Stap 3


The purpose of this step is to keep things manageable and obtain a less data crowded overview. It will be up to your own preference and research purposes to decide which columns you like to keep in and which ones to leave out. I personally have found it convenient to:

  • Only display column D (admin), column H (shared cM), and columns V, W and X (ethnicregions; ethnictraceregions; matchurl).
  • I changed the column width when needed.
  • Also I changed the display format for column H (shared cM) from “general” into “number” (right-click entire column; select “format cells”; choose category “number”).
  • The end result looks like this (column D/admin is blank because of privacy reasons):

***(click to enlarge)

Stap 4



Initially your matches appear to be sorted on the amount of shared DNA as measured in centimorgans in column H (shared cM). It is however advisable to sort your DNA matches according to Ethnic Regions (column V). This is very helpful for your first manual browsing through your DNA matches. It will enable you to zoom in on DNA matches with a particular regional profile. Because they are sorted on primary region they are already grouped together.

For example as I am of Cape Verdean descent it makes it very easy to single out my most likely Cape Verdean DNA matches (I literally have hundreds of them!). Due to their distinctive profiles Cape Verdeans will almost always have “Senegal” mentioned as biggest region in their African breakdown. For other parts of the Afro-Diaspora it is however more complex. Because their African breakdowns tend to be far more diverse and not consistent at all when it comes to the primary region. But let’s say you yourself happen to have a prominent “Africa Southeastern Bantu” or “Mali” score then it might be a rewarding avenue of research to get in contact with your DNA matches who likewise have these not so commonly appearing regions in first place.

In fact this sorting on ethnic region can potentially also be useful to research any of your non-African lineage, such as Asian, Native American or European. You might also want to use text filtering to zoom into one particular region. Do keep in mind though that there are always several ways you could be connected to people of mixed background. For example just because you happen to have a DNA match who is primarily Asian doesn’t per se mean you both have an Asian ancestor in common. It could also be the other way around and your MRCA actually is connected with the minor non-Asian origins of your DNA match. Plausibility rather than wishful thinking will be the best guideline as always.

  • Go to Tab “Data”
  • Select the entire worksheet. Either by pressing Ctrl+A or by clicking the select all button in the upper left corner
  • Click on Filter, which should result in each column getting its own filtering menu
  • Click on the filter button for column V (ethnicregions)
  • Choose “Sort A to Z” or if you prefer the other way around “Sort Z to A”

***(click to enlarge)

Stap 4


It should be noted that the labeling of the ethnic regions is sometimes different from the labeling on AncestryDNA. Scanning with DNAGedcom will alter the regional labeling in most cases (AncestryDNA Helper maintains the original labeling). Perhaps a bit confusing at first but once you’re aware of it the labeling still makes sense.

UPDATE  15-12-2018:

The overview below is also including the new regions which have been added as of the update in September 2018! Because not all of Ancestry’s customers have actually updated their results the scanning will currently result in an overview including both updated and unrevised ethnic regions. The appearance of brackets […] also being indicative in this regard.

  • “AfricaBantu” = “Africa Southeastern Bantu” (only appearing for profiles which have not been updated since September 2018)
  • “AfricaN” = “Northern Africa” (“Africa North” before the update)
  • “AfricaSanPygmy” = “Africa South-Central Hunter-Gatherers”
  • “America” = “Native American” (only appearing for profiles which have not been updated since September 2018)
  • “AmericaN” = “Native American—North, Central, South”
  • “AmericaS” = “Native American—Andean”
  • “AngloSaxon” = “England, Wales & Northwestern Europe” (“Great Britain” before the update)
  • “AsiaC” = “Asia Central” (only appearing for profiles which have not been updated since September 2018)
  • “AsiaE” = “Asia East” (only appearing for profiles which have not been updated since September 2018)
  • “AsiaNE” = “Central and Northern Asia”
  • “AsiaS” = “Southern Asia” (“Asia South” before the update)
  • “Balochistan” = “Balochistan”
  • “Baltic” = “Baltic States”
  • “Basque” = “Basque”
  • “Burusho” = “Burusho”
  • “Celtic” = “Ireland and Scotland” (“Ireland/Scotland/Wales” before the update)
  • “Dai” = “Southeast Asia—Dai (Tai)”
  • “EuropeE” = “Europe East” (only appearing for profiles which have not been updated since September 2018)
  • “EuropeIb” = “Iberian Peninsula” (only appearing for profiles which have not been updated since September 2018)
  • “EuropeJe” = “European Jewish”
  • “EuropeN” = “Scandinavia” (only appearing for profiles which have not been updated since September 2018)
  • “EuropeS” = “Europe South” (only appearing for profiles which have not been updated since September 2018)
  • “EuropeW” = “Europe West” (only appearing for profiles which have not been updated since September 2018)
  • “Germany” = “Germanic Europe”
  • “GreeceAlbania” = “Greece and the Balkans”
  • “Gujarati” = “Western and Central India”
  • “Korea” = “Korea and Northern China”
  • “Luhya” = “Eastern Africa”
  • “NearEast” = “Middle East”
  • “Slavic” = “Eastern Europe and Russia”
  • “TurkeyArmenia”: = “Turkey and the Caucasus”
  • “Vietnam” =  “Southeast Asia—Vietnam”
  • “UralVolga” = “Finland” (“Finland/Northwest Russia” before the update)


Apply advanced filter criteria in order to leave in only “100% African” profiles among your DNA matches. These advanced filter criteria are defined in such a way that only DNA matches with exclusively African AncestryDNA regions will remain in your overview. This is probably going to be the most important step to see if you have any native African DNA cousins. However this advanced filtering method can actually also be used to zoom into DNA matches who are “100% Asian”, “100% European” etc. (see filters 4 & 5 in my google spreadsheet). This is all to be done at your own discretion.

Below instructions are based on my trial & error experience. I have learnt the hard way that you really need to be very precise if you want to make these advanced filter criteria work. So be patient and don’t despair if it doesn’t work at once. Then again depending on your system settings/Excel version it might very well be that there are several roads leading to Wakanda 😉

  • Enter the advanced filter criteria into your Excel sheet
    • Copy the advanced filter criteria which I have already prepared from this google spreadsheet. They can be found under the heading “Strictly “100% African” filter (A8:AQ9).
    • You can also devise these criteria yourself (use this tutorial)
    • Or just copy them manually from the screenshots below [keep in mind that the criteria have been expanded because of the update in September 2018! The new regions are however not included in the screenshots on this page]
  • Go to Tab “Data” in your own Excel sheet
  • Make sure to enter the criteria horizontally and not vertically
  • Also there should be at least one blank column in between the criteria and the last column of your DNA matches data (at least this is the only way how it worked for me)
  • If your last column is X (matchurl) then for example rightclick on cell AA1 and paste the copied criteria from my googlesheet. (A8:AQ9). When needed you can also do one row at the time.

***(click to enlarge)

Stap 5a

***(click to enlarge)

Stap 5b***

Apply the advanced filter criteria on your DNA matches data

  • Click on Advanced filter
  • Define the range of your DNA matches data (“list range”)
  • If the filter buttons from Step 4 are still in place and you first click on any given cell within the range this will be done automatically
  • Make sure that all of your matches are included in the range by checking the last row number which should equal the total amount of your matches (minus 1, in my case 5,846)

***(click to enlarge)

Stap 5c


  • Define the range of your advanced filter criteria (“criteria range”)
  • The easiest way to do this is to
    • first click on the upperleft cell (in my case AA1)
    • keep the SHIFT key pressed
    • scroll all the way right
    • and then click on the bottom-right cell (in my case AT2)
    • or else just type it in manually

***(click to enlarge)

Stap 5d


  • For the finishing touch just click on Enter or OK and voilá only your “100% African” DNA matches will be left over (if you have any). In my case my most likely Gambian DNA cousin shows up as the only profile among my 5,846 matches who is exclusively African (genetically speaking).
  • Obviously “100% African” DNA profiles could also very well belong to a Afro-Diasporan rather than be a native African. In STEP 7 I will list some tips on how to make the distinction. But just like for this step there are no guarantees that this will enable you to single out each and every native African profile. In the next step I will first discuss the matter of native African DNA cousins who aren’t “100% African”.

***(click to enlarge)

Stap 5e



Apply less stringent advanced filter criteria to also include native Africans who are not “100% African”. Afterall many native Africans are “mixed” according to AncestryDNA’s format because they also have additional regions appearing such as Middle East, Asia, Europe or even the Pacific. These native (atleast for more than a thousand years or so) Africans would be North & East Africans, the Malagasy from Madagascar, and also the Fula and related people from across Sahellian West Africa. This complicates finding adequate advanced filter criteria because naturally you wouldn’t want to miss out on any possible DNA matches from these areas. Even when especially North & East African matches will for the most part be quite unusual and uncommon for Trans Atlantic Afro-Diasporans.

In my googlesheet I have included additional sets of advanced filter criteria which might function as a workaround. Naturally this less stringent filtering method does also increase the number of DNA matches who are actually Afro-Diasporan. In the next step I will discuss some tips on how to make the distinction. But first:

  • Undo the filter from the previous step by simply clicking on Filter.
  • Make sure that all your (sorted) DNA matches have reappeared again
  • Enter the less stringent advanced filter criteria into your Excel sheet.
    • Copy the advanced filter criteria which I have already prepared from this google spreadsheet. They can be found under the heading “More inclusive African filter” (A14:AB15).
    • Actually in my googlesheet I have also defined a third set of criteria which should ensure that only matches without any European regions are left over: to be used at your own discretion.
    • You might also want to tweak around with these criteria yourself (use this tutorial). Please leave a comment if you think you have found any improvements!
    • Or just copy them manually from the first screenshot below [keep in mind that the criteria have been expanded as of the update in September 2018! The new regions are not included in the screenshots on this page]
  • Follow all the remaining instructions from STEP 5

***(click to enlarge)

Stap 6a


In my case 12 matches are appearing when applying this filter. But going by their ethnic regions as well as profile details I could verify that no additional native African matches were among them. I have also used these less stringent criteria on several other profiles I have access to. And for some of these people I did manage to find additional native African matches (in particular Fula but also Kenyans). The second screenshot is showing the DNA matches of a woman who is half African American and half Senegalese Fula. You will notice many of her matches even have European trace regions (low confidence) aside from Middle East scores. I would consider this to be merely a consequence of genetic overlap between North Africa/West Asia and Southern Europe. And not any robust indication of genuine minor European lineage. Column D with the most likely background details was filled in by myself.

***(click to enlarge)

Stap 9b



Determine if your filtered DNA matches are native African or Afro-Diasporan. Of course getting in contact with your DNA match will provide the best answers. Otherwise basically you will want to go by every possible clue available. Obviously this list is not meant to be exhaustive and the results will also reflect your own research skills 😉

  • Check if your match has a public family tree and look for the birth places. This information will not always be conclusive though due to adoption, migration etc. Also from my experience this information is regrettably not filled in by the greater majority of native Africans on Ancestry (as well as Afro-Diasporans actually).
  • Pay attention to the profile name. If it seems “African” google for it, even if it’s just a nickname! You might also want to search for the name on or on Facebook. Often you will be able to zoom into most likely nationality or even ethnic group. Although many African names are in fact widespread across several countries. Also be careful because sometimes Afro-Diasporans adopt African names (due to conversion to Islam, marriage or as an Afro-pride statement).
  • Check the profile details of your DNA match (by clicking on his or her name). Usually the member profile is left blank but sometimes spoken languages, research interests etc. are mentioned, which could provide a useful hint.
  • Click on “shared matches” to see if you and your match have any mutual DNA matches. This will usually not be the case for native Africans. While especially for people of your own nationality the likelihood will naturally be much greater.
  • Do a check on plausibility for the ethnic regions mentioned for your “100% African” matches. Native Africans tested on AncestryDNA will usually also show several regions in their breakdowns. However most of the time results of native Africans can easily be distinguished for having a much narrower regional scope on average than Afro-Diasporans, even if the latter happen to be “100% African”! The regions reported for native Africans will almost always be:
    • geographically adjacent
    • generally fewer in number
    • and often with only one single predominant region
  • Even if your “100% African” DNA match turns out to be an Afro-Diasporan it might still be very helpful to contact them as they might yet somehow clarify parts of your African lineage. Otherwise it will simply be interesting to find out how you both are related.

***(click to enlarge)


***(click to enlarge)

AA-HT-JAM - 100


Finishing Remarks

I have applied this advanced filtering method not only for myself but also already for over a hundred other people who have shared their profiles with me. I am therefore confident that this method works! Up till recently I actually used a much less thorough method based on simply browsing through my list of matches on Ancestry’s website and paying close attention especially to the profile names which appeared to be African. And then checking if the preview of their ethnicity estimates would indeed be “100% African”. That’s how I discovered my most likely Gambian DNA cousin. Another method I used was performing a search on birth location by typing in various African country names.

Eventhough these latter methods do also deliver results I have found that there are added benefits to this filtering method:

  • It is more systematic and comprehensive. Because of the automated scan performed by either DNAGedcom or AncestryDNA Helper and the automated filtering performed by Excel. In principle it should show you all “100% African” DNA matches (if you have them). Including those profiles without African names or family tree! Which from my experience is actually the case for a majority of native Africans on In other words if you are merely browsing through your matches based on name or birth location you are potentially missing out on native African DNA matches!
  • Furthermore it is timesaving. I have heard some people open the profiles of each and every match in their list to check the preview ethnicity estimates for “100% African” DNA matches. Again this can also be effective. However it is also tedious and especially if you have over a 100 pages of DNA matches it can take up a LOT of your time! Naturally such a manual proces will also be prone to inadvertently overlooking native African DNA matches.

All of this is not to say this advanced filtering method will be flawless 😉

  • Not all native Africans are going to be “100% African” genetically speaking. And Afro-Diasporans can of course also be “100% African”. I already dealed with this matter in STEP 6 & 7. However it could also be that one of your DNA matches happens to have one parent who is native African and one parent who is from somewhere else. These DNA matches could also provide you with valuable clues as long as you can ascertain that your MRCA is indeed along the native African parent’s side.
  • Some of your 100% African matches might actually not be 100% African at all! In fact based on my experience in scanning other people’s matches a greater majority of filtered matches may turn out to be false positives! This will involve a greater effort on your part to sort out the native Africans from Afro-Diasporans (see STEP 7). If not by some glitch on part of either DNAGedcom or Ancestry this outcome will usually be due to an incomplete summary of the ethnicity estimates for your matches. In most cases these filtered matches actually will have some non-African admixture as well. These would be people who opted out of the so-called full “Ethnicity Profile Display”. Seemingly a default setting and applied to most of your DNA matches. But in some cases it might happen that someone chooses to un-check this box and then “your DNA matches will only see the portion of your ethnicity estimate that they share in common with you“. (see DNA settings on
    • Click to enlarge
  • Even if this method will provide you with the contact details of your African DNA matches, don’t be surprised if the response rate will generally speaking not be very high. This is a general complaint on and actually due to several reasons and not per se a sign of disinterest. Instead of being judgmental or disappointed just lower your expectation level and be thankful for every reply message that you do receive from your newly found cousins!
  • Registering with GED-Match will enable you to “fish in a bigger pond”. As the database of this free of charge website will include DNA tested Africans from other companies as well (such as 23andme and FTDNA). You can upload your raw data from by following these instructions:
  • Best of luck in your ancestral quest!


Please Leave a Comment!

I wrote this tutorial because I was not able to find one aimed specifically at Afro-Diasporans.  And I strongly feel this kind of information can potentially be very beneficial for people seeking to connect with their African DNA cousins.  To be honest it was my first time writing a tutorial and therefore I found it quite challenging. If you have read this blog post all the way down to this section I would really appreciate any kind of feedback! Anyone who knows how to improve on this tutorial or has any additional tips/methods please share and let me and the readers of this blog know so we can all mutually learn! Also if you have any questions feel free to ask them (do keep in mind that I am no IT genius so I won’t be able to answer all of your technically related questions 😉 ). Did this filtering method result in any new African DNA matches for you that you weren’t aware of before? If so how many? I am really excited about this method so I would love to hear how it works out for other people!



This tutorial has been modified several times since I first published it in 2017. In order to ensure a continued or even improved detection of African DNA matches. Mostly triggered by updates on Ancestry but also through helpful tips from the comment section.

UPDATE  02-03-2018:

This method for detecting African matches revolves around scanning and then filtering all of your DNA matches. When I first published this blogpost I was only aware of AncestryDNA Helper as a suitable tool for scanning Ancestry’s DNA matches. However in the meanwhile it has been brought to my attention by a helpful commenter that the DNAGedcom Client is a very efficient tool for scanning DNA matches on Ancestry. For many if not most people this tool will probably be much faster and overall less bothersome than AncestryDNA Helper.

I have therefore replaced the screenshots in my tutorial to reflect the process while using DNAGedcom instead of AncestryDNA Helper. From my personal experience with DNAGedcom I can indeed confirm the scanning is much faster. And furthermore it runs very smoothly in the background without interfering with your possibly other PC/laptop activities. It should be pointed out though that AncestryDNA Helper is a free service while DNAGedcom requires a monthly subscription fee of $5,-. This monthly subscription can however be canceled immediately so you might choose to pay it only once. In any case I would advise everyone to do your own research to find out if this tool might be appropriate. These sites might be useful to decide:

The steps of my tutorial essentially remain the same however there are some minor display changes when using DNAGedcom instead of AncestryDNA Helper. If you like to use AncestryDNA Helper instead of DNAGedcom please follow this link to the old version of this blog post:

Instructions/screenshots when using AncestryDNA Helper instead of DNAGedcom


UPDATE  15-12-2018:

AncestryDNA’s ethnicity estimates have been updated as of September 2018. Because Ancestry now provides 43 global regions instead of 26 global regions in the previous version this has impacted my filtering methodology. I have expanded my advanced filter criteria in my online spreadsheet to reflect the new situation (only for DNAGedcom). I have however not changed the screenshots nor the main text further below! All references to Ancestry’s ethnicity estimates are therefore pertaining to AncestryDNA version 2 which was current between September 2013 and September 2018. Unless specified otherwise.

In my opinion the new AncestryDNA version regrettably has been a downgrade rather than providing any meaningful improvement in regards to the African breakdown. This means it has become more difficult to distinguish African DNA matches based solely on their regional admixture scores. Another complicating factor is that Ancestry has discontinued the usage of trace regions a.k.a. “low confidence” regions. For my scanning/filtering method this implies many finer distinctions can no longer be made. At least not for any profiles which have been updated. For more details on the update see:


UPDATE  15-12-2018:

As of October 2018 Ancestry has enabled an Ethnicity Comparison feature for all customers and not just USA-based smartphone users. With this very useful tool you can actually verify the exact regional percentages of your DNA matches! Unless they have disabled the complete ethnicity profile display (see above). This greatly enhances your chances of verifying your possibly African DNA matches!

Compare ethnicity results between Cape Verdean and her Senegalese match

***(click to enlarge)

CV-SEN compare

This overview is obtained after clicking on the green “Compare” button appearing on the page of a most likely Senegalese DNA match for one of my Cape Verdean cousins. Aside from the very plausible regional combination of almost exclusively “Senegal” + “Mali” also the surname of this match is very indicative of a Senegalese background. Take note that no European regions are mentioned for the Senegalese DNA match (on the right). While the total is almost 100% (76% “Senegal” + 23% “Mali”). One percent missing because that trace region “Southeastern Bantu” is not being reported for my Cape Verdean cousin. Although it is being mentioned in the preview of the profile page of the Senegalese match.


UPDATE  29-03-2019:

For those who do not have Excel installed you might want to try out this alternative filtering software called OpenOffice. It has very graciously been suggested by a helpful commenter who also provides instructions on how to use it. I have not tried it out myself yet. So I cannot offer any assistance or guidance. Nor can I guarantee its well functioning.  As always do your own research to find out if this tool might be appropriate. See this link for more details.


UPDATE  31-10-2019:

The alternative option of scanning by way of AncestryDNA Helper is sadly no longer available. Due to Ancestry’s change in the display of DNA matches. Which is why I have removed this possibility from the main text. Ancestry is also (gradually) rolling out a new update of their Ethnicity Estimates (see this link). This will likely impact the filtering criteria. But for now I do not have a complete oversight. The current filtering criteria should still work though. Possibly only leading to more matches to manually browse through after filtering.


UPDATE  08-02-2020:

The screenshots for Step 1 (scanning with DNAGedcom) have been replaced. DNAGedcom regularly updates its software. Also the display is changed quite frequently. The current screenshots are taken from client version


UPDATE  25-02-2021

I created a demonstration video for Rootstech 2021. Which can still be seen by way of Youtube:


71 thoughts on “How to find those elusive African DNA matches on Ancestry

  1. I got lost at Step 8. Plus, I don’t have access to the google sheet. Still, this is obviously an excellent tool if one can whiz through it.

    Liked by 2 people

    • Thanks a lot for your feedback, much appreciated! I have changed the access settings for the googlesheet, it should work now. Please tell me if it is still not giving you access.

      Step 8 is the most crucial step, i got stuck here myself also at first haha. Please let me know where exactly you can’t proceed. Is it when applying the advanced filter criteria and defining the range?

      I’m keeping my fingers crossed that this method will produce new African matches for you!

      Liked by 2 people

  2. Okay, to make it easily for me, I just filtered by “text” -> “begins with” (e.g., “Mali,” “Senegal,” “Nigeria,” or “Ivory” etc. This still produced the desired results. The search criteria returned all matches with the top ranked targeted regional cluster. Although it doesn’t filter out the admixtures, I can get an overall view of my matches that have “Mali” or “Senegal” ranked first in the primary regional cluster. I was still able to easily spot my 100% African genome matches. THANKS FOR SHARING!

    Liked by 1 person

    • You’re very welcome! I find that sorting your matches on ethnic regions (STEP 7) is very helpful indeed and furthermore it can also be used for other research purposes.

      How many native African matches on are you currently aware of and what is their background?


      • Thanks for giving me access to your DNA matches! I applied both advanced filters and some really fascinating findings appeared. This is very helpful as I intend to do a follow-up blog analyzing the DNA matches for people across the Diaspora. I verified the background of each of your filtered matches according to STEP 10, but only the ones with a * can be confirmed through their family tree or family name.

        FILTER 1 – STRICTLY “100% AFRICAN”
        Total number: 10x (out of 5,426)
        Native Africans (most likely): 2x
        GHANA*: 1x

        MADAGASCAR (possibly): 2x

        Contrasting these native African matches with your entire ancestral breakdown seems useful. Your African breakdown is consisting mostly of a predominant socalled “Benin/Togo” score of 40% as well as 25% “Cameroon/Congo”. Yet it appears you have no DNA cousins sofar from these places. While your Ghanaian match most likely corresponds with your “Ivory Coast/Ghana” score of 7% and your Nigerian match is possibly/perhaps derived from your Nigeria score of barely 1%! Even when the labeling of the AncestryDNA regions should of course not be taken too literally and especially Nigerians are often described by additional regions, incl. Benin/Togo and Cameroon/Congo. Still i suppose this outcome would be suggesting that a greater part of your African ancestry is not yet reflected in your DNA matches.

        What i found even more surprising were your two possibly Madagascar matches (one of them seems pretty certain going by the name the other less so). Given that you show no traces of either Asian or Polynesian DNA. Also your “Southeastern Bantu” is rather minimal at 1%. Then again you could also be related to them by way of shared “Cameroon/Congo” DNA, which is spread out across southern Africa to a greater extent than indicated by

        Furthermore just browsing through your entire list of matches i spotted two persons who are most likely of half Ethiopian and half African American descent. I suppose the most plausible way you are related to them would be through their AA parent. Intriguingly i also came across a profile with a seemingly Nigerian/Igbo name. The ethnicity preview also looks in line with a Nigerian background except for one European traceregion (Great Britain). Of course it might be possible for a few Nigerians to have some distant European ancestry reflecting their individual familyhistories. This goes to show one of the limitations of the “strictly 100% African” filter.

        I also found it very interesting to see some of your cross-Diaspora matches from the West Indies and Latin America. Going by their Portuguese names and ethnicity breakdown i think it’s very likely you also have two Cape Verdean as well as two possibly Brazilian matches. However one of these possibly Brazilian matches could in fact also be a socalled mestiço (mixed) Angolan! This is the ethnicity breakdown, you will notice an absence of any Native American and a predominance of Central African/Bantu origins:

        Main regions: Cameroon/Congo, Africa Southeastern Bantu, Nigeria
        Traceregions: Iberian Peninsula, Mali, Ivory Coast/Ghana, Italy/Greece, Africa North, Ireland, Senegal


  3. Congratulations on your new find Felipe!

    Out of curiosity, what was the length of your shared segment?

    Thanks again for your continued interest, your blog means a lot to an increasing many


    • Thanks a lot for your comment Ben, i truly appreciate it!

      The shared segment is 7,3 cM, with a confidence score of “Moderate”. Estimated likelyhood of it being IBD is said to be in between 15%-50%. Also perhaps tellingly none of my Cape Verdean DNA cousins are shown as shared matches. So eventhough i am still very much thrilled about this finding i do realize it may not be as clearcut as it may seem at first.

      Just to apply some of the considerations i discuss in this blog on myself :

      – There definitely seems to be a possibility this match will turn out to be IBS or perhaps also IBP (Identical by Population). In the latter case i could still be genetically connected to my match but just in a much wider timeframe.

      – Judging from historical demography the greater part of my (mainland) African ancestors lived in the 1500’s or 1600’s before they ended up in Cape Verde. Of course i might have a few mainland ancestors as wel from the 1700’s, maybe even 1 from the early 1800’s but they would most likely make up a small share as most Cape Verdeans had been free from slavery atleast since the census of 1731 when the enslaved portion of the population was a minority of 17%. I suspect most African segments in the Cape Verdean genepool are therefore too diluted and fragmented (due to recombination) for IBD matching.

      – I have been assuming my match is from Gambia and a Mandinga based on his name. However the mentioning of “Nigeria” as one of his main regions does already suggest he might be ethnically mixed. Possibly some distant Hausa lineage? Even if this only turns out to be a misreading on Ancestry’s part the fact remains that the Mandinga of Gambia are known to have absorbed many other ethnic groups within their ranks ever since they migrated from Mali. In particular the Banhun (a.k.a. Bainuk) and the Jola (a.k.a. Diola) who had been residing in Senegambia prior to the arrival of the Mandinga. Both of these ethnic groups were also heavily present in Cape Verde during the slave trade era so my MRCA with my Gambian match could very well also have been either Jola or Banhun instead of a Mandinga.

      – Another historical possibility is that the direction of geneflow is actually the other way around! With my possibly Gambian match having distant Cape Verdean lineage rather than me having a distant Gambian ancestor. The Gambia river used to be a very important trading area for Caper Verdean merchants in the 1500’s and many actually settled here. Culturally speaking they formed distinctive socalled Luso-African communities up till the late 1700’s. After that time no historical reports are known and they are assumed to have just assimilated and merged within the general population.Nonetheless genetically their traces might yet still exist!

      – Although i know already of a few persons from Guinea Bissau who have tested with still their customer database will be biased against Cape Verdeans receiving matches from that country. While the number of DNA tested Gambians and Senegalese also isn’t that large i suppose given their greater migrant presence in the US/UK the odds might be greater to receive matches from those countries.

      – My African breakdown as reported by AncestryDNA is a predominant combination of “Senegal” and “Mali” (together about 2/3 of my African DNA). This is in line with other Cape Verdean profiles and historical evidence pointing towards Upper Guinea being the overwhelming source of African ancestry for Cape Verdeans. In this way it’s not at all surprising that i received a match from Gambia. Arguably the Mandinga may even have been among the most numerous and definitly the most culturally influential among the many groups which are known to have been present in Cape Verde. Still based on historical plausibility i am pretty much convinced that within my African DNA are included many more ethnic contributions from allover Upper Guinea, from Senegal to Sierra Leone, And quite likely also beyond to a minor degree (Angola and/or Mozambique?)


  4. Sadly, I can’t get past the scanning part. For some odd reason it doesn’t load all of the information. I rescan and sadly, there is still chunks of data missing. I am going to be brave enough to try one more time…lol.

    Liked by 1 person

    • Thanks for the feedback! Scanning the matches can be troublesome sometimes. I completed the full scan of your matches and also filtered them according to my tutorial. Out of 4,749 DNA matches 9 profiles are “100% African”. At least genetically speaking. Unfortunately i don’t think anyone of them appears to be a native African. Only 1 profile might be half Nigerian and half African American judging from the name and familytree. the others are either African American or West Indian.

      Interestingly i did however find 4 profiles who are most likely Cape Verdean! Now you might actually also be related to them via an European MRCA but it could also be linked to your 10% “Senegal”! If you should ever find a DNA match from Senegal or neighbouring countries i would be very interested to hear about it!


  5. Im kinda in the same situation a Ameera,Its taking me weeks to scan and not all of the data is scanning. Plus i had 102 pages of matches to go through that has now moved up to 105, I feel like im in a never ending chase. I don’t think i’ll be able to get this done, the only way would be to go through it manually.So far Ive only been able to find about 2 Pure blooded africans all nigerian.I found other Africans but they were not pure blood but also looking like they have a recent Nigerian relative that mixed with and afro american. People have been contacting me that have gotten matches with me(non pure bloods), but we cannot find any link and i just end up giving up.One Woman i ended up matching on ancestry,Then i matched her on Ged-match also.Not only that her mother took the test and came up as match also on Ancestry and Gedmatch to me. But non of our families share ONE last name or relative in our family tree. And her’s is very large.She doesn’t even live anywhere near me and her family history most of it is far away from mine.She’s also insanely mixed as far as lineages. Im a bit dissapointed, it makes me wonder how accurate these match ups are.Also a generous of my close matches have alot of Madagascar combos in their Dna (I.E Polynesia,Melanesia,Asia South or some form of asia all at once)But i don’t have an ounce detected of any of them on ancestry, just some slight noise in Asia central and west asia. But all my Asia ranges are ZERO.Also the “Search by birth locations” Seems to be unreliable, because i type in african countries and get people that said they were born in Alabama, but ancestry says they were born in ghana, or ethiopia,or angola. Come to find out they just have it in their lineages,they wernt actually directly from their. Is it even normal to have this many matches and growing? cause im abit discouraged right now. Ive decided to give 23andme a try to see what they come with and get 2 estimates. from my experience most people get simular results between ancestry and 23 and me.with only a max of 3% difference.

    Liked by 1 person

    • Thanks for the feedback Vanessa! I have been hearing this from several people now that the scanning of the matches is causing problems. I have finished scanning your 105 pages of matches and it took me a while but still no more than about 4 hours i think (i will get to my findings later on). The only time i had to really take almost a day out for scanning was for a person who had over 400 pages of matches (=20,000!) So i’m really curious why for some people it would take several days and still the scan is not completed even. I suppose it might have to do with internetspeed when i do a speedtest i get: DOWNLOAD 15.03 Mbps // UPLOAD 30.17 Mbps. What is your internetspeed?

      But perhaps it’s also caused by an additional factor. According to AncestryDNA Helper :

      The speed of a scan is highly dependent on the number of matches you have, the speed of your computer, the speed of your Internet connection and the activity going on with the website at the time. The faster your computer and Internet speeds, the faster your scan will be.

      Once the extension starts scanning, it will open a new tab in the Chrome browser for each match, so you will see flashing or find it difficult to use the browser or anything else on the computer for other things. It is best to leave it running for however long it takes. This feature was implemented due to earlier versions that were blocked by memory issues in Chrome. While scanning, the AncestryDNA web pages might error out or hang occasionally. The extension makes every attempt to auto-refresh and these time-outs are far less frequent then they used to be.

      If the scan fails to automatically refresh and appears to be stuck for several minutes, you will need to press the F5 key or click the Reload button to get it started again. You can Cancel and restart the scan at any time with no loss of data. If you get the Chrome “Aw, Snap” screen, click Reload, then go to your DNA home page and click the Resume Scan button.


    • Hi Vanessa, I completed the full scan of your matches and also filtered them according to my tutorial. Out of 5,236 DNA matches only 3 profiles are “100% African”. At least genetically speaking. This would explain why you have found it so hard to find any African matches!

      I have a strong feeling this is correlated with your family US state origins. I found it very interesting to see that Ancestry gave you 3 genetic communities, two of them exactly corresponding with what you shared with me earlier: North Carolina & Maryland! The third one “Deep South” is most likely due to ancestral connections created by Domestic Slave Trade originating in the Upper South. In other words the people in this GC will (most likely) be connected to you not because of a MRCA from the Deep South but rather a MRCA from NC/MD who was relocated. Either way the fact that your family hails from the Upper South and also the fact that you seem to have very few 100% African DNA matches would suggest your family presence in the US is relatively speaking very longstanding and to be traced back atleast to the mid 1700’s if not (on some lines) to the late 1600’s!

      From the three 100% African matches, one would be Nigerian judging from name and ethnicity breakdown. The shared segment size is quite impressive (16,4cM)! The second match might be from Ghana (Ewe) and is perhaps to be linked with your primary “Benin/Togo”score! However i have no certainty about this. It might turn out he is actually from somewhere in the Americas. The third 100% African match could possibly be from Guyana (based on some googling) however also otherwise this match is most likely either West Indian or AA and not native African.

      Applying my more inclusive filter i found two additional profiles which might belong to Africans! One of them seems very likely to be a Cape Verdean, and the second one could be a person of Fula descent (but probably only partially so judging from his regional breakdown). So this might possibly link into your 5% Senegal but of course also other ancestral options could be valid. Either way this finding might shed more light on your Upper Guinean origins.


    • To hark back at some of the issues you mentioned:

      – Finding out how you are exactly related to any of your DNA matches is usually going to be difficult indeed. Even with family trees in place and discounting the possibility of false positives or IBS matches. For useful tips on African American genealogy i would recommend checking out this great overview of blogs:
      *African-American Genealogy Blogs, DNA Blogs, and Genealogy Podcasts

      And also this blogpost could be inspirational:
      *African-American Genealogy: Unearthing Your Family’s Past, From the Present to the Civil War

      – Any Madagascar connection you might have is likely to be very diluted as most Madagascar captives arrived in the early 1700’s. Due to socalled genetic “wash out” it could be that the Southeast Asian DNA trace amounts are no longer detectable for yourself but still are for some of your DNA cousins. Also of course it could be that you inherited African DNA from a Madagascar MRCA! Which could be either “Southeastern Bantu but also Cameroon/Congo”!
      See also:
      * Southeast Africa (maps, charts etc)

      – The “search by birth locations” approach is indeed not always very productive. That’s why i devised this tutorial instead for finding native African matches. From from my experience most Africans who get DNA tested do not fill in any public family tree so you will not be able to find them by searching on birth location. Furthermore of course the reliability or rather quality of the information contained in any given family tree is dependent on the person who created the family tree.

      – Lately the customer database of Ancestry has been increasing very rapidly and that’s why also the number of your DNA matches is growing. Not all of these matches are going to be helpful but like i said in this blog post the more people (incl. Africans) get tested the higher the odds of eventually finding insightful connections!


  6. Fonte Felipe, thanks for your great contribution! I’m sure it will help many people.

    Something important I have to share, I think should help many of the people who’ve had trouble with the scanning using AncestryDNA Helper.

    A much, much easier way to download files of your Ancestry matches is to use the DNAGedcom Client program. DNAgedcom is another site like Gedmatch that’s volunteer-run. The DNAgedcom Client program will download all of your matches from any of the three main companies, Ancestry, 23andMe, or FTDNA. It’s very easy to set up, and it rarely gets hung up in scanning, as AncestryDNA Helper does.

    It produces a very nice CSV file with all the columns already there, so you won’t need to convert text to columns. Not only does it give you the Matches file with all their ethnic ancestry estimates, but also, two other wonderful files. The Trees or Ancestors file lists EVERY ancestor of ALL your matches who have a public tree, including the dates and places of their birth and death. If you are lucky enough to have African matches who have trees, you should be able to spot them with this file, which has many uses. Finally, it creates the ICW file, which lists all of the Shared Matches among all of your matches. Very interesting information!

    All of your filtering could get started much more quickly by starting with the DNAgedcom Client. I use sorting instead of filtering, myself, and find that it works just as well.

    Also, I constantly use color-coding, so the interesting ethnic ancestry will jump out at me. This is done through Conditional Formatting. You choose “Highlight Cells Rules,” and then select “Text that Contains”. Choose a word or phrase that you want to spot easily, like “Bantu,” and choose a Custom cell color of your liking. This is extremely helpful! You can choose to apply a rule to only the columns you want, and multiple rules can apply to those columns, one for each word or phrase of interest.

    Please be aware, though, that the DNAGedcom Client changes the names of the ethnic ancestry categories to make them more abbreviated and easy to sort. It’s still easy to tell which are which.

    I’m very interested in Malagasy ancestry. I’m not sure that I have any (that would be great, but I have to be realistic), but a number of my friends, clients, and their matches, whose origins are primarily European, are showing strong signs of trace Malagasy ancestry. They have various combinations of SE Bantu and South-Central Hunter-Gatherers, with Melanesian, Polynesian, South and East Asian, etc. These are mainly folks with deep roots in early colonial Virginia.

    I wish everyone much joy and success in this exciting endeavor!

    Liked by 1 person

    • Hello Bonnie, thanks so much for your wonderful suggestions! This is exactly the kind of constructive feedback i was hoping for! Many people have indeed been complaining that the scanning process through AncestryDNA Helper has been troublesome. I had heard about DNAGedcom before but never tried it out myself. I have done so now in the meanwhile and it went exactly as you said!

      The set-up was straightforward and easy. I have 75 pages of matches and a full scan took me just under 60 minutes!!! Also scanning for family tree details and shared matches took me under 1 hour! I have since then proceeded with scanning several other profiles i have access to,and again no hick-ups, just smooth and fast scanning. So this definitely seems like a good alternative for AncestryDNA Helper! As an extra bonus i understand you can also use this tool for downloading your matches on 23andme although i haven’t tried this out yet. You will have to be willing to pay a small subscriber’s fee ($5) for this tool unlike AncestryDNA Helper which is for free. But in my opinion it’s totally worth it!

      I have already received positive feedback & recommendations from other people who are currently using DNAGedcom. Naturally everyone has to decide for themselves whether they like to try this. Performance might vary for some people. I would advise everyone to do their own research therefore to find out if this tool might be appropriate.

      Just speaking for myself i think it’s a very helpful tool and I will make some needed adjustments in my tutorial. Again many thanks Bonnie, truly appreciate this!

      p.s. You might be interested to know that i intend to publish a new blog post featuring the DNA matches of about a dozen Africans across the continent who tested on Ancestry. I will provide a summary and some comparative analysis of their DNA matches. Among the Africans featured will also be a person from Madagascar who has no less than 63 pages of matches! Many of them also being European Americans! I actually have been coming across European American matches for all African profiles i have analyzed sofar, most of them with some minor African ancestry, showing up as mere Trace regions. But intriguingly also several of these matches were “100% “European”genetically speaking, atleast according to Ancestry’s estimates.


      • Hi Felipe,

        I wanted to thank you for all of the great information you blog about. I have learned so much reading your posts! This particular one has allowed me to find a lot more African matches than I could dream of, and so far, two of them are emailing with me and interested in working together to research more. Unfortunately, my matches haven’t done their family trees, and as an African-American, I don’t have too much info on my ancestors to work with. One of my matches is Fula and the other is 100% African from Jamaica, and has several African regions. Do you have any suggestions for where to take it from there using Gedmatch, or other tools that can bring us closer to finding out who the common ancestor is?

        Liked by 1 person

        • Hi Aurelia, really glad that you have found my blog to be useful! Also exciting that you have managed to find an African match through this filtering method!

          From what I’ve seen many Africans on Ancestry tend to not have public family trees. And if they do it’s usually not going back that many generations. Not beyond the 1900’s at least. One also has to keep in mind that traditionally speaking many African cultures keep track only of the direct paternal and/or direct maternal line. All other family lines, starting for example with one’s paternal grandmother or maternal grandfather and all their predecessors are often not recorded going back more than 3 generations.

          Of course for most Afro-Diasporans there will also be a great deal of missing information going beyond the 1900’s. So generally speaking the chances of finding a mutual ancestor by way of comparing family trees are quite slim. Nonetheless I do think that several advanced genetic genealogy tools/techniques might be fruitful in learning more about which possible family line the common ancestor with your African match is to be associated with. So for example finding out if this Fula match is on your maternal grandmother’s side or your paternal grandfather’s side is already very valuable information.

          Depending on the amount of shared DNA with your match you can already get an idea of how solid the match could be. However by testing close relatives, preferably your parents or older generations, it makes it much more easier to establish if your match is indeed a socalled IBD match (Identical By Descent). And also on which family line your common ancestor may be placed.

          Although I have no personal experience with it triangulation and DNA Painter may also give you more insight. This will be very tricky given scanty information and will also require a lot of patience. But it might still be worthwhile for eventually zooming in closer to your West African origins along a certain family line or even actually identifying a West African ancestor! For more details see:

          Benefits of Triangulation (Segmentology)
          Using DNA Painter to reconstruct ancestral DNA (Ultimate Family Historians)
          Pinpointing the Origin of Family’s Igbo Ancestry with DNA


  7. Can I use Google Sheets instead of Excel for the processing? If not, will one of the free Excel substitutes work (Libre Office, for example)? Thanks!


    • Hi Michael, I do not think Google Sheets has a filtering functionality so it will not be suitable. As for the other Excel substitutes i have not used them myself. Please let me know if they turn out to be helpful!


      • Thanks much for the reply! I appreciate it. I don’t own Excel, however I found that one can subscribe to the Office suite for $6.99 a month. So I’m off and running to work with your method. I’m excited. 😊


        • You’re very welcome! I’m hoping you’ll find some African DNA matches. If you should get stuck at any of the steps i will gladly assist.


      • Hello Fonte Felipe and Michael Douglas-Lyr. This is a late response, but I have used Open Office with some success. It is a free word processing application which comes bundled with a spreadsheet program called OpenOffice Calc. Mac OS and Win PC versions are available.

        The program may be downloaded at this link:

        Here’s how you can manually run filters (skip to step 5 if you just need run a filter):
        1. Open OpenOffice Calc
        2. Go to File–>Open and navigate to the .csv file (your list of matches) you want open.
        3. Calc may ask you about import options. Choose ‘Unicode(UTF-8)’ for the Character Set and ‘Comma’ for the Separator options.
        4. When Calc opens you should see your data organized by column. (testid, matchid, name, etc)
        5. To filter the data, in the top menu bar, go to Data–>Standard Filter.
        6. Hit the little arrow in the box below ‘Field Name’ and from the roll down choose ‘ethnicregions’.
        7. Now Calc will filter according to what is contained in the ethnicregions column.
        8. The ‘Condition’ for filtering should/can be ‘Does not contain’ in the roll down list.
        9. In the ‘Value’ box, manually type “EuropeN”.
        10. Hit the OK button. All matches that contain ‘EuropeN’ in their admixture will be deleted.

        So essentially, you are telling Calc that only under the condition that EuropeN is not a part of that person’s make up should you retain that match. If the person doesn’t have Sweden in their makeup, retain them. If they don’t have any EuropeS retain them. etc. etc. You keep filtering until you are essentially left with people who only have ancestral genetic ties to the continent.

        Copy the remaining matches and past them in a new spreadsheet by right clicking to the right of the “Sheet1” label and choosing ‘Insert Sheet’. Past the filtered entries into that sheet. Run another filter on the new sheet (Let’s say EuropeS this time). Copy those entries into a new sheet. Rinse and repeat until you have whittled your entries down to African-only matches. Fonte has provided the list of names for exclusion in this article.

        I hope this helps!


        Liked by 1 person

        • Hi Brent, thanks so much for providing this alternative! Much appreciated!
          I have not tried it out myself yet. But your instructions on how to use it look straightforward and easy to follow. It should be helpful indeed for people who do not have Excel installed.


    • Hi Jamelle,

      Have you tried out this tutorial i describe in this blog post yet? If you came across any difficulties please be specific and let me know at which step you get stuck. Generally I would recommend using DNAGedcom rather than AncestryDNA helper for the scanning part. Because for many people AncestryDNA Helper doesn’t seem to work or just is too troublesome.


  8. Wow. I just tried it and it worked! I’m sitting here looking at these matches like, wow. I understand that I need to go through a few measures to confirm that they are true matches, but having these names to start that process with is SO HELPFUL!! THANK YOU!!

    Liked by 1 person

  9. I’ve only been in contact with one Nigerian cousin on ancestry and since then I’ve found another cousin. He has by far the highest Nigerian percentage I’ve ever seen.
    His results are:
    93% Nigerian
    3% Cameroon/Congo
    2% Africa South-Central Hunter Gatherers

    I’m not sure what to take away from his results. Not to mention what tribe he could be.

    Liked by 1 person

    • A good starting point would be with this person’s name or surname in terms of their ethnic group if you’ve not yet been able to contact them/get a reply back.


    • That’s a very high score indeed! I agree with Damon, very often you can get valuable information from a person’s name and surname. If it’s given of course. I tend to just google it or else also do a search on FB to see which locations/states are mentioned for people who carry the same surname. The website called is also a great source for finding out which countries are associated with particular surnames.


  10. Hello. I recently found another possible African cousin, but I’m not sure what tribe she belongs to.(I have a pretty good idea).
    I won’t reveal her first name so it remains confidential but her last name is “balde”.
    Her updated results are:
    66% Mali
    28% Senegal
    5% Northern Africa
    1% Middle Eastern
    When I searched her full name on Facebook, I saw mostly women of whom birthplaces were Guinea Conakry.

    I do know the Fulani originated in Senegal, so that region gave me confidence also that they often always have North African dna.

    What do you think?

    Liked by 1 person

    • I think Balde could indeed be a Fula surname. They do not seem to have too many different ones. From what I’ve seen it’s only the spelling which can be different sometimes. Depending on if the person is from an English-, French- or Portguese speaking country. Guinea Conakry is home to a very large population of Fula people. So no surprises there. Also the breakdown looks in line, as it seems to be the updated version.

      Facebook is indeed a good way of checking where a certain surname is most current. Another way to check is


      • Thanks. I’ll check facebook. Also, recently on Ancestry I found two arabian dna matches, not sure if it’s in line with my senegal/mali score or if it’s unrelated noise?
        One of them has 93% Middle East with a splash of Eastern Africa/Cameroon/Congo. She also had 3% Africa SC HG which im not sure if thats common for arabians.
        The other is mainly middle eastern with caucasus/asia south.

        Do most black americans have middle eastern/north african matches?

        Liked by 1 person

        • Very nice! You’re on a roll! How much shared DNA with these Arab matches? I have seen North African matches reported at times for African Americans but quite rarely so. Arab matches for AA’s I have not seen yet. I will shortly publish my findings on the DNA matches I found for 50 Cape Verdeans. They tend to have about 4 North African matches on average, each person. And also Fula matches are reported for them very frequently. Generally speaking a different historical context will apply. But it still might be useful for you too as I will also discuss methodology and African surnames.

          Liked by 1 person

          • My shared dna for my Fula match:
            7.8 centimorgans across 1 segment.

            For my arabian matches, one of them is higher than one of my Igbo matches at 6.5 centimorgans across 1 segment.
            The other is at 6.1 centimorgans also across 1 segment.

            Just a side-note last night, I finally connected with a woman from Ghana who is Akan(Ashanti). So that explains IvoryCoast/Ghana, but im not sure why I haven’t found any actual (IBD) Liberians/SierraLeonans. Do you think some of my Senegambian/Guinean dna is described as Ivc/Ghana?

            Liked by 1 person

            • Prior to the update “Ivory Coast/Ghana” was reported in considerable amounts or even as primary region for Guineans. Also further north into Senegambia I have seen it being reported in double digit amounts in a few cases.


              • Wow. Last thing, me and my Fulani match together share the migration Virginia & Southern States African Americans.
                I remember reading on one of your blogs here that according to stats, only about 3,000+ slaves from sierra leone in Virginia. Which is less than the 4,000+ slaves from IvoryCoast. But may I say that your report on this was very accurate. Most virginia slaves had a combination as follows: Biafra, Senegambia, Ghana/West-Central Africa. So I guess it’s likely that my Ghana score is describing guinean ancestry.

                Liked by 1 person

  11. Fonte,

    I think I’ve identified my last African match. The Masendu fella turned out to be a Shona from Mashonaland in Zimbabwe. This is my second Zimbabwe match, the other being from a related people on the other side of the country, a Kalanga from Matabeland. Then I have Zambian Nsenga from nearby. I’ve been kind of surprised to find all of these southern Africans. In the old update, all of these had predominate Cameroon/Congo with significant Southeast African Bantu minorities, which I found interesting. It might show that these are more recent arrivals to their land, though the Kalanga are said to be ancient, but maybe they were genetically assimilated while keeping their language (before the Ndebele moved in in the 1800’s, anyway).

    I was really confident that my Cameroon, Congo and Southern Bantu region would be weighted towards the west (Cameroon, Congo, Angola), but so far what I’m finding is southern Bantu peoples, who are much more underrepresented in the diaspora. Then again, given how far back these matches are, maybe it’s just that these groups moved very recently from their former homelands.

    I’d love to be able to find more, but it looks like the tool has to be fixed first.

    Liked by 1 person

    • Hey Damon, thanks for your help in earlier post. Prior to the update, did you have a high Cameroon/Congo score? I want to try and find some eastern african matches.

      What I have noticed now, is when I compare my results with ancestry users who haven’t updated their results, Cameroon/Congo and Southeastern Bantu is combined for me but for them they are separate regions still. I have 52% Cameroon/Congo in the new region after the update; which is quite a jump from 14%.

      Maybe you and I could be cousins, what is your name on Ancestry?


      • No, not particularly. I had 14% Cameroon/Cong in the old update + 2% African Southeast Bantu. The regions were indeed combined with the new update. Mine did rise – to 21% – but that’s still not particularly high for an African American.

        The reason for some people still not having them combined is that they’ve either not been on to accept the update, or they’ve chosen not to.


  12. Same situation as well. Before update I had 14% Cameroon/Congo with a trace of (2%) Africa SE Bantu.
    Now I have 0% Eastern Africa
    Inflated Cameroon/Congo

    I believe I have much higher upper Guinean ancestry, but it’s lost due to Ancestry oversampling Mali. I have 6 Igbo matches and 3 Upper Guinean matches. As well as 1 Ghanaian match, 2 Arab matches in total that I’ve found.

    I was so excited to find a Fula match, since I was only finding mandinka. I’ve tried to find many more but ancestry’s surname search is kind of faulty.


    • I believe Mali would technically be Upper Guinean, at least interior Upper Guinean. I also lost my “Senegal (3%)” to “Mali.” Most of the rest of my matches West African are also Igbo (I lost my 1% Nigerian, though), one Ibibio/Efik Nigerian, and two Ewe cousins. I have a few far-flung cousins (Ethiopia, Madagascar, maybe North African), and then three Cameroon, Congo & Southern Bantu cousins, I’ve yet to find any cousins from the western end (Sierrea Leone, Liberia, Senegambia, Mali) of West Africa, though I suspect they are there and simply haven’t tested or they haven’t yet matched me with them. I get a full 20% of my DNA on MyHeritage being measured as “Sierra Leone,” which I think

      Ancestry has to do better by Nigeria, though. Over a third of my matches – a group larger than any other individual region – are Igbo or Igbo-adjacent, yet before and after the update I basically have no “Nigeria” region.


  13. Do you have any known Middle East matches?
    Even though MyHeritage has Sierra Leone, it has no senegambia region. It instead groups all of west Africa into one region, while Nigeria gets its own region.

    Any northeast African or Congolese dna shows up as Kenyan/Masai.

    Lastly, how did you go about finding your ewe, madagascar and Central African matches?


    • I used one of the methods in this post, which I’ve now been told is no longer of use since Ancestry changed their regions.

      No, I’ve found no Middle Eastern matches, which is not a surprise as I never had any Middle Eastern or North Africans regions. As for MyHeritage, I’ve found no African matches. But that is also not surprising as it seems customers seem to be very European and North American-based, even more so than Ancestry.


      • I used one of the methods in this post, which I’ve now been told is no longer of use since Ancestry changed their regions.”

        The scanning and filtering method described in this post can still be used also after the update. There is only a temporary hick-up right now when scanning the profiles of customers who have either accepted the update or who were updated right away because they are new customers. The addition of several new regions (mostly from Europe & Asia) has not yet been properly accounted for by DNAGedcom. They are already aware of the issue and working on adjusting their scanning tool.


  14. Wow. Then I wonder why I have Arabian matches, since I too have no North African or middle eastern dna on ancestry.


  15. Felipe, do you think my Arab matches and Fulani match is a confirmation that I have deep Fulani roots? Or could it be something else?


    • Well I would say your Fula match of 7.8 cM is certainly indicative although not per se a final confirmation. Having both of your parents tested will give you more certainty that this match is indeed a socalled IBD match and not a false positive.

      But even if it is also showing up for one of your parents I suppose that theoretically your mutual recent common ancestor (MRCA) with this Fula match could possibly still be of a different ethnic background. Such as Mande, Wolof or Bambara, all of these people are known to have intermixed with the Fula across the generations. Usually along the maternal line, with their descendants eventually being considered fully Fula.

      Your Arab matches have a very minimal shared DNA amount (< 7cM). Which greatly increases the odds of them being false positives or generic population matches (MRCA will be from ancient times rather than from within a genealogical timeframe). Given that you do not show any Middle Eastern or North African DNA yourself I suppose the connection with these matches could be by way of their minor African admixture. A chromosome browser would be ideal to make the call on the ethnicity of the shared DNA segment. But such a tool is unfortunately not available on Ancestry.


      • Thank You. Also, do you think Benin/Togo(updated) can also describe Mali/Senegal too?

        I wish I kept my previous results because most of my Nigerian/Congo/Northeast Africa dna is described as Cameroon, Congo, and Southern Bantu Peoples.
        Mali and Senegal have been deflated greatly.

        I really really hope Ancestry fixes this.😔


        • I can’t say for sure but by coincidence these are the updated results I came across for a most likely Gambian person:



  16. I try to look at my new maches daily, and I’ve come across two more what I believe to be African matches, though neither has yet replied to me. As usual, I want to through out some surnames and see if anyone knows what country and people they could be from. The first surname is “Mur.” Her regions included 44% Benin/Togo, 33% Cameroon/Congo/Bantu, 12% Mali, and 11% Ivory Coast/Ghana. The surname is unusual for coastal West Africans, I believe, so I’m guessing it’s somewhere up north/further in the interior?

    The second surname is “Ekwuocha.” She is 53% Benin/Togo and 39% Cameroon/Congo/Bantu, and 8% Nigerian. This surname looks more coastal West African, and just looks Nigerian. What I can’t figure out is if she’s Igbo, Yoruba or something else.

    What I’m most excited about is that “Mur” – if she’s of direct African descent. – is my closest African match, yet, by far at 15 centimorgans. I have cousins I converse with that don’t match me that closely. So finding her out could really help me trace a small part of my African origins.



  17. I took a shot at trying to pin down some african relatives but I used google sheets instead of Excel. When I uploaded the spreadsheet into google sheets the “ethnic region” columns doesn’t show any information. What can I do to get around this?


    • Which software did you use for downloading your matches? DNAGedcom or AncestryDNA Helper? Because of the recent update on Ancestry which included an increase of especially European & Asian regions the scanning may not have been complete in the last two months. I would suggest to scan your matches again (preferably with DNAGedcom), as it seems atleast DNAGedcom has now adjusted their settings.


    • Thanks a lot for your comment, Hannah!

      I actually intend to provide an extra service on this blog in regards to finding & interpreting African matches on Ancestry. So stay tuned!


  18. Hello Fonte Felipe. Thank you for sharing your thoughts and research with the world! I have learned a great deal from reading your posts!

    You mentioned being careful about concluding that a distant relative match is IBD “Identical by Descent” instead of IBS “Identical by State” (IBC chance). Many of my matches are below 7cms. I have been using a feature on that will list other matches that you share in common with the current person you are being compared against.

    If using the results of your parents was not an option, but some of these shared matches were 4-6th removed cousins, would this method still qualify as a form of triangulation?

    I would be interested in your thoughts on the matter. Thank you!

    Liked by 1 person

    • Hi Brent, you’re very welcome! I likewise would like to thank you for your excellent comments! This is an interesting question indeed. Because Ancestry does not provide any segment details or a chromosome browser it unnecessarily complicates proper identification of how exactly you and your matches “match”. However this shared matches feature on Ancestry certainly can be of potentially informational value!

      I have scanned for African matches on Ancestry for many people now. And actually it does not happen that often that a shared match is being shown. This is because of the restrictive settings which if i’m not mistaken specify that only matches with shared DNA >20cM will be shown. But generally speaking such close matches between Africans and Afro-descendants in the Americas are rather uncommon, from what I have seen.

      However whenever a shared match IS being shown I always take notice of it as I believe it may very well provide additional insight on which family line the African match is to be placed. This does require some basic family tree building effort though, both for your self and your shared matches.

      For some useful & inspiring research methods see these blog posts:


  19. My most recent African cousin match in 23andme’s database is a man from Ivory Coast.

    On his paternal side both his g’parents were from west Ivory Coast very closely bordering guinea.(I’m guessing he’s malinké on his paternal side).

    On his maternal side his g’parents were from north Ivory Coast. One from Korhogo, and the from Noumousso in the northern part of the country. (Probably senufo?)

    The other ancestral birthplace he has is Mali as well as an ancestral surname: ‘Coulibaly’, which is a Bambara surname.

    Liked by 1 person

    • Nice! How did you find this match? Just by browsing your new matches on surname? I wish there was a way to also scan the matches on 23andme on ethnicity so that you can filter them on “100% African” profiles as well.


  20. I was just scanning through the matches and I found him and figured he was African.(His name was a giveaway).

    His name is Moustapha Sanogo, which is pretty popular in Ivory Coast as well as Mali.

    Sub-Saharan Africa 99.6%
    Ghanaian, Liberian and Sierra Leonean 76.6
    Nigerian 7.5%
    Senegambian & Guinean 4.7%
    Congolese 0.7%
    Broadly West African 9.5%
    Broadly Sub-Saharan Africa 0.7%

    European 0.2%
    Broadly European 0.1%

    Unassigned 0.2%

    Liked by 1 person

  21. Hello Fonte Felipe,
    trying to use your tools with the new features from Ancestry DNA and finding the ethnic trace regions no longer are present in the download file.
    Will there be any future updates


    • Hi Cecilia, good to know you are trying out this tutorial! The ethnic trace regions have been discontinued by Ancestry since their update in September 2018. Regrettably so as they were very useful for my filtering method. The DNAGedcom software has also had several updates this year. To keep up with the latest developments you might want to check out this FB group:


  22. Hello Fonte Felipe, thanks for this site. It has given me a new place to learn more about my DNA results. In a previous post, Jen mentioned her high Mali DNA of 22% I too have been trying to find more info on my Mali DNA. I am African American with several generations of my family from South Carolina, so I was surprised to find my Mali DNA at 40% (Ancestry DNA). I often look for matches and I still have not found any that has this high Mali percentage. I also have 3% Indigenous Americas- North, which my family says comes from my great grandmother (my mom’s father’s mother). I’d love some tips on how to learn more about my DNA, especially the highest percentage – the Mali one. Thanks!

    Liked by 1 person

    • Hi Valerie, thanks for commenting!

      That’s quite a high “Mali” score indeed! However given your South Carolina origins not really surprising. Because I have seen such primary “Mali” scores for other people with SC state origins as well. Usually from the Lowcountry. It is a valuable indication for your West African lineage. However you do have to keep in mind it will have been inherited through multiple family lines. And ultimately it might trace back to dozens of individual African-born ancestors from different places or countries even. Because the country labeling is not to be taken too literally. As in fact Sierra Leoneans and Liberians also tend to receive high “Mali” scores whenever they test with Ancestry. This also goes for people from other neighbouring countries of Mali.

      So really this score may imply several things at the same time. However given that we know that South Carolina has strong historical connections and also valuable cultural retention from the so-called Rice Coast it makes it very likely that this score is pinpointing Upper Guinean lineage for you. Which in itself is already very informational!

      For more specification really you need to look into your African DNA matches. Have you tried my tutorial yet? In fact it can also be used to single out DNA matches which may connect you with your Native American admixture.

      In case you want me to have a look at your matches and/or also analyze your DNA results you might want to consider my African DNA Matches Service:


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s