The GEDmatch Dodecad Project (Explained For Beginners)

The Dodecad project on GEDmatch was one of the first admixture tools on the website. The other projects are based on the original Dodecad software.

It’s important to remember that the admixture projects aimed to research ancient populations. You may get some insight into modern ethnicity, but it largely depends on how your heritage aligns with the DNA samples used in the analysis.

This article reviews the Dodecad calculators and lists out the modern populations that may best align with the project. That should help you assess which (if any) of the Dodecad calculators are useful for you.

Origins Of The Dodecad Project

From the late 2000s, a small number of bloggers were researching and writing about anthropology and genetics outside the traditional university environments.

Their independent analysis was made possible by international genome projects that put a wide variety of DNA samples into the public domain.

Some of the genetic analysis tools used by university departments also became available as open-source software. Open-source means that they are available to download and use for free (as long as proper credit is given).

One popular application allowed a DNA sample to be compared to hundreds (or thousands) of reference samples. The idea was that you carefully choose reference samples that represent a single ethnicity or geographic region.

The software would go to work on the target sample and eventually spit out a percentage breakdown of similarity with the reference samples. In other words, the admixture percentages.

Enter Dienekes

In 2010, a Greek blogger turned his attention to using open-source tools to look at ancient ethnicity patterns. This blogger writes under the psuedonym of Dienekes Pontikos.

Dienekes felt that the big academic genome studies were overly focused on west European DNA. So he set out to study other European populations.

The Original Dodecad Project

Dienekes already had a blog that focused on anthropology and population migration. He put out a call for readers to send him copies of their raw DNA files from 23andMe. This was his original goal:

To build samples of individuals for regions of the world (e.g. Greeks, Finns, Albanians, Southern Italians, etc.) currently under-represented in publicly available datasets.

Dodecad blog

Notice that he didn’t want English, Celtic, French and German samples. He felt these were already well-researched.

Nor was he looking for African samples, despite those also (still) being highly under-represented. Dienekes tried to keep a tight focus within his original project. He would expand later into other regions.

Dienekes also developed his own versions of the analytics tools. In a crucial move (for GEDmatch), he made the desktop software available as an open-source download. He called this the DIYDodecad.

Spawning More Projects

Dieneke’s open-source offering allowed other anthropology bloggers to use and amend his analytics software for their own interests.

Several projects quickly appeared on other anthropology blogs. These had names like MDLP, Eurogenes, and Harappa. If those monikers sound familiar, you’ll have seen them in the list of GEDmatch admixture projects.

But before they appeared on GEDmatch, they were run as part of individual blogs. This caught the eye of the renowned Nature research journal. They covered the phenomenon in late 2010 in an article titled “The Rise Of the Genome Blogger”.

I’ve mentioned that these were independent bloggers. Each tweaked the software in ways they felt produced “better” results. Some collaborated, but there was outright hostility amongst others. When you’re evaluating the different projects, it’s worth remembering that some creators felt that rival projects were very misleading! And vice-versa.

GEDmatch Dodecad – An Inauspicious Start

Around the same time as Dienekes had launched his new Dodecad blog, the GEDmatch website was launched by two co-founders.

GEDmatch had a small team of voluntary programmers. At some point, they got interested in the open-source software that Dienekes had put online for download.

Another blogger, Zack Ajmal, had already co-opted the software for his own project. Ajmal worked with the GEDmatch developers to create a web version of the desktop software. In other words, they produced a modified version that could run on the GEDmatch website.

The GEDmatch team launched a new Admixture section as part of their free tools. They loaded up several projects, including a version of Dodecad.

The problem is that they didn’t provide the appropriate credit to the original developer. Dienekes was understandably annoyed when he spotted an altered version of his work on the website.

Thankfully, the GEDmatch guys got in touch with him and addressed the attribution issues. They had briefly removed the admixture tools, but now it was full steam ahead on the website.

The Dodecad Calculators

The next sections will go thorugh the different tools within the project itself. These are the various “calculators”.

But quickly – what is a “calculator”? It’s a specific version of the software that tries to produce the best results for a set of populations. So, if your ancestors were all from Olde England, you’ll probably steer towards different calculators to those used by people of Asian descent.

Dodecad V3

The original Dodecad calculator was aimed at what Dienekes called “Eurasia”. What’s that, you ask?

Always bear in mind that these labels were assigned by the project creators. They may not correspond to academic or lay usage.

So, I’ll take the list straight from the source:

  • People of Greeks, Cypriots, and Turkish descent
  • People from the Anatolia, Balkans, and the Caucasus
  • Armenians, Iranians, Assyrians, and Arabs
  • Jews from Italy, the Balkans, or Anatolia
  • Non-Indo-European speakers from Europe (e.g., Finns, Hungarians, Basques)
  • Scandinavians and Icelanders
  • Italians

Dienekes tweaked the calculator a few times, and the last version is up on GEDmatch as Dodecad V3. If your heritage is in the list above, you may want to take this calculator for a spin.

However, Dienekes created other calculators as he worked to improve his findings. Read on.

Dodecad World9

The next Dodecad calculator on GEDmatch has the misleading name of “World9”. That may sound like it represents all populations of the world. But spot the “9” at the end.

This was an amendment to a project (not on GEDmatch) that targeted seven of his “Eurasia” populations.

World9 was an attempt to add indigenous Americans (Amerindians) and Australians (Australasians). In other words, an extra two populations from other parts of the world. Hence…World9.

This is a link to the blog page.

It’s important to take note of Dienekes cautious note about Americans researching their native lines. As a general rule, these projects aren’t great for people with mixed heritage.

Dodecad K12b and K7b

More and more people were sending their DNA results to Dienekes. As he continued to expand his work, he produced the K12 and K7 calculators.

The numbers refer to how many regions or communities are considered within the calculations. One has 12 regions and the other has seven.

And why “b”? Well, these are the second iteration of the calculators.

Both K12 and K7 are based on the same reference data set. It’s just that K12 breaks them down into more specific regions.

Don’t be mistaken that more is better. The higher the number of regions, the fewer samples are in each region. And therefore, the confidence levels (accuracy) go down.

These are the twelve regions that Dienekes identified and labeled:

  • Caucasus                            
  • Siberian               
  • Atlantic_Med   
  • North_European             
  • East_Asian         
  • South_Asian     
  • Southwest_Asian            
  • Southeast_Asian             
  • Northwest_African        
  • East_African      
  • Sub_Saharan
  • Gedrosia (explained in the next section)

Remember that these labels (as in every project) are assigned by the project creator. For example, North_European may not mean what you think it means. If you want to dig deeper, then this spreadsheet matches the reference samples to the project’s labels.

Dodecad Gedrosia

Never heard of Gedrosia? It’s a historic region of Asia that was invaded by Alexander the Great. You’ll find a map at this link.

Remember, the Dodecad project isn’t a peer-reviewed academic study. It’s a project developed by one individual pursuing personal areas of interest. And if you dive into Dienekes’ pursuits, you may find a rather Hellenic perspective of the world. And that’s fine. Other projects have different slants.

If you decide to take a closer look at the Gedrosia breakdown, you may have similar questions as someone posed on the Dodecad blog.

Do you have any theories as to why the Gedrosia component is relatively high in the Irish sample compared to other European populations?

Blog comment

And this was the answer from the project creator:

My guess is that Central Asia has something to do with it, and there is a latent element that is beyond our reach using modern populations, because Central Asia has been much changed due to the arrival of Turkic peoples.

Dienekes reply

So, now we’re getting into speculative anthropology. Which is pretty much what these projects were invented for. I chuckled when I first read the answer, as it seemed to be the perfect piece of hand-waving. Those latent elements are always useful when an anomaly cannot be explained. But to be fair, it’s a clear acknowledgment of the limitations of using modern DNA samples to predict ancient heritage.

Alternative Gedrosia Calculators

The Dodecad projects date back to 2011/2012, and haven’t been more recently updated on GEDmatch. However, there have been some major finds of ancient human remains since that date.

The GedrosiaDNA project has calculators that were uploaded in 2016/2017. These calculators target heritage from the Gedrosian region, and uses a selection of more recent archaic DNA discoveries. We have a full article on the GEDmatch Gedrosia project here.

Dodecad Africa9

Dienekes expanded the Dodecad project to include a calculator for African heritage.

Below are the nine populations in the breakdown. These labels are defined by the project creator, and may not represent modern usage.

  • Biaka    
  • Mbuti   
  • San
  • North West Africa           
  • East Africa          
  • Southern Africa
  • West Africa        
  • South West Asia              
  • Europe

The reference samples were gathered from public genome projects and supplemented with volunteer DNA samples sent to the project blog. You can see how the project samples fall within the populations in this spreadsheet.

Dienekes has a very specific warning about who this project is and isn’t useful for.

It should be used only by Africans and African-West Eurasian admixed individuals. It is not meant for people with additional admixture (e.g., South/East Asian or Native American).

To be clear, this means that your heritage should preferably be entirely African. If you are of mixed heritage, the non-African side should be “West Eurasian”.

So, what the heck is West Eurasian?

It’s not always easy to tell whether the population labels in these projects refer to the academic consensus on these terms. Bear in mind that some of Dienekes views on anthropology and population migration go against the mainstream.

West Eurasian broadly covers northern and western Europe and extends through the Balkans and the Ottoman Empire. If you want to dive into it, you can use the search box on the blog.

If you’re looking for alternatives, I have a roundup of the best GEDmatch calculators for African heritage.

Dodecad Vs Eurogenes

The creators of the Dodecad and Eurogenes projects had a prickly relationship. The genome bloggers weren’t constrained by academic niceties. Their arguments sometimes descended into flame wars that spilled out of their blogs and onto other anthropology forums.

You can get a feel for this from Dienekes’ rather provocatively titled article: Eurogenes is upset.

Dienekes describes Eurogenes as rude, ungrateful, and mean-spirited. Eurogenes (Davidski) makes a few choice comments underneath the article, including calling Dienekes a prat.

The entire exchange is embarrassing. This particular dispute revolves around which of Ukrainians or Poles are more Western European (I’m paraphrasing). That’s a quite specific issue.

But both men have a more serious dispute about the credibility of each others’ calculators. The issue revolves around whether the calculators can be accurate for people who weren’t part of the original project. Yes, that’s you and me.

If you want to read the arguments and rebuttals, here’s Eurogenes and here’s Dienekes.

If you don’t have the time or the energy, I’ll summarize here. Both guys agreed there was a problem. Both guys say the issue is fixed in their own projects but is present in the other. Hope that explains everything.

But if you want a pictorial representation, then the double-spiderman meme never fails.

How To Use The GEDmatch Dodecad Calculators

Do you need a step-by-step guide to using the Dodecad calculators?

We have a separate tutorial that covers interpreting your admixture display using the GEDmatch Oracle utility and spreadsheets. The guide specifically uses the Dodecad project as an example, although it applies to all other projects.

Other GEDmatch Admixture Projects

We have a detailed run through of several other GEDmatch projects:

Alternative Ways To Research Ethnicity

If you’re interested in researching your recent ethnic heritage, then you should look at alternatives to the GEDmatch projects. If you’ve tested with Ancestry, then we have an article on interpreting Ancestry ethnicity estimates.

But aside from that, you could check out the new genetic groups on MyHeritage. This shows geographic regions based on your DNA matches. I found it remarkably accurate for my Irish lines.

Although you can transfer your DNA for free to MyHeritage, the ethnicity features require an unlock fee of about 30 bucks.

More Articles And Tutorials?

.

Margaret O'Brien
Latest posts by Margaret O'Brien (see all)