GEDmatch.com is about to get a face lift. Many have complained that it is not intuitive and does not offer enough help for a newcomer. The new site hopes to correct some of that; I am helping with the beta test now; it sure is prettier!
Why use GEDmatch and what is it? Well it’s a free 3rd party site of tools where you can upload your DNA test results and compare them with those from people who have tested at other companies. Plus there are many analysis tools, some available nowhere else. These include GEDCOM (family tree) comparisons, including yours to your DNA matches.
Recently I gave a talk for the Family History Fanatics at their Winter of DNA conference where I demonstrated how to use GEDmatch without worrying about segments. An approach that gets a lot out of the site without having to be a geek like me. The slides are here: https://slides.com/kittycooper/gedmatch-basics-2021
Here is a step by step of that very basic approach as outlined in the syllabus for my FHF presentation
1. Create a User ID and Upload a DNA raw data file, aka a “kit”
Click the image above to go to the video from Andrew Lee, of FHF, which shows you how to create a user id and upload your DNA raw data to GEDmatch.
Or read this blog post of mine which also has links to my other GEDmatch posts
2. Once Logged in, your Dashboard Page is not Intuitive
The tools are in the right hand column and the resources you have uploaded are listed on the left. You can upload more than one DNA kit and multiple GEDCOMs. There is a help section on the top right.
This post from Jim Bartlett is a useful explanation of the home page, as is my basics post listed above.
And here is what my beta test Dashboard looks like, an improvement? I think so.
3. Try out the Different Ancestry Composition Calculators
These are called Admixture at GEDmatch. They are fun and have some things that might be useful but, like everyone else’s, they are hardly a certainty yet. Click the image below to view the slides from my admix presentation to learn how to use these tools.
4. Upload a pedigree only family tree
It is good to upload a privatized family tree of about 10 generations of just ancestors. In the past when I had a more complete GEDCOM, people would think their match with me was to an in-law they found by comparing our trees. To avoid that problem I suggest you only include blood relatives. To export a partial GEDCOM from your family tree try this advice: https://www.wikitree.com/wiki/Help:Splitting_a_GEDCOM …Note that a GEDCOM is a text file so you can always use a plain text editor to privatize it yourself if your software does not do that. Just change the first names to LIVING and the birth dates to just the year.
Plus if you have your family tree at WIKItree (a free online collaborative tree), you can add your DNA kit number to your profile there; then your one to many listing will link to that tree (provided it is “public”). Click here for my post on WIKItree’s DNA features and here for more on WIKItree and GEDmatch.
5. Make Some Tag Groups
My recommendation is to use the same colors for your tag groups that you used for your dots at Ancestry or in your Leeds Chart. I suggest making one group for each (great) grandparent line and groups for each locality for as yet unknown matches. Click here for my basic tag group blog post and here for my slightly more advanced post.
6. The Main Tool to Find Relatives at GEDmatch is the One-to-Many Comparison
There are two different one-to-many functions in the free tools. The one marked Beta has the most functionality with the ability to sort by column and display tag groups. The other will tell you in English the kit source and will highlight a low number of shared SNPs for comparisons in pink or red when there are too few for a confident comparison.
Click here for old post that explains the column headings for the Beta version. This is the One-to-Many that uses your colors to indicate the kits that you have added to your personal tag groups that are matching the kit being looked at.
7. When you find a good new match, look to see who else they match
Note that the definition of a good match varies by population group (any endogamy?) but a good start for third cousins or better would be to share > 90 total cMs with more than one segment and at least one segment larger than 20 cM. Use this online calculator to see the possible relationships for a specific cM amount
Compare your kit and theirs with the People who match both, or 1 of 2 kits tool to see who else they match at GEDmatch
If you have set up some tag groups, you can also click on a match’s kit number from the one-to-many to see their own one-to-many match list, with your tag groups on, to see which groups of your known relatives they match.
If the match has a GEDCOM, you can look at it for common names and locations or better yet use the 2 GEDCOMs Comparison tool to compare your gedcom to theirs. Click here or my blog post about using GEDCOMs at GEDmatch.
If you have Tier 1 (a paid account), there is a tool to find the common ancestors with your DNA matches who have uploaded GEDCOMs, see
Another thing to do with Tier 1 is to combine your kits from different testing companies, click here for that blog post. The problem is that each company is testing slightly different SNP sets which can cause errors in matching when not enough SNPs are in common between the two kits. If you combine kits from different companies, you will get better results since more SNPs are included.
Now all those steps are pretty easy and do not require delving into segments. However once you get the hang of the GEDmatch site you may want to try using segments. There are lots of posts here plus I recommend Jim Bartlett’s blog called Segment-oloy – https://segmentology.org/ or some of the later slides in my presentation, starting at slide 36