Getting Started with GEDmatch

Posted on by

Many of us ask Matches to upload to GEDmatch. Some do. Most are bewildered by what they see – I sure was, when I started! It’s particularly daunting with AncestryDNA Matches who generally don’t have any prior experience with DNA segments. Well, just a little noodling around can go a long way. This blog post will suggest some easy steps for anyone who has just uploaded to GEDmatch.

First the login. Your GEDmatch page is anchored on the email you used to sign up and a password you provided. If you forget your password, just enter your email and click on forgot password.

When your GEDmatch home page opens up – stop for a moment and look around.

– Messages from the GEDmatch Admin are at the top.

– Look at each of the big boxes:

– Information – your profile info.

– File Uploads – links to upload raw DNA data files on the left and links to upload a GEDCOM (Tree info) on the right (it pays to do both).

– Learn More – Several links to learn more about GEDmatch – not just yet, but soon, click on each link and look it over.

– Analyze Your Data – Just read the title of each of the Utilities for now; I’ll come back to some of them later.

– Your DNA Resources – a list of your GEDmatch kits (you can upload more kits). Note the EDIT or DELETE link.

– Your GEDCOM Resources – the GEDCOMs (Trees) you’ve uploaded.

– Tier 1 Utilities – some advanced utilities for user who pay a subscription (see bottom of your home page); see how much you’ll use GEDmatch before subscribing.

 

JUMP IN – Get your feet wet – take a utility on a trial run and see what it’s like! Start with the ‘One-to-one’compare – just click on that line. Enter your GEDmatch kit number in the first box; and someone else’s kit number in box 2. Usually someone has asked you to upload, and they should have given you their kit number. Then hit enter, or click on the Submit button. You’ll get a table of the DNA shared segment(s) with Chromosome, Start and End Locations, cM, SNP – this is the physical information about the DNA you share with a Match – see my blogpost here for more info. Under the chart are some other data, including “Estimated number of generations to MRCA” – please take this number with a grain of salt! It’s a calculated number. 1.0 means parent/child. As the number gets larger, it’s actually more of an average than anything specific. Don’t put a lot of stock into anything over 4.

Now – use your browser back arrow to get back to the Comparison Entry Form – it should have the two kit numbers still there. Select the button for Graphics and Positions to get a colorful display of your 22 chromosomes compared to your Match. Read the legend at the top with particular attention to the red (no match) and yellow (half match) – when the yellow (and green) is long enough, the utility will show a shared segment with a blue bar. And you’ll see the same table info you saw before. This view helps put the whole DNA matching thing in perspective. For most of your cousin Matches, the colorful chromosome bars will be alternating red and yellow (maybe a little green). This indicates that we match on a lot of our DNA, but just not in long stretches. Generally, you and a cousin Match will have only one stretch of matching yellow (and green) that is long enough to call a Match. As you read through this blog, you’ll learn that when the shared segments of different Matches overlap and match each other, you have a Triangulated Group, which can be very helpful.

 

NEXT STEP – Now that you’d tried the basic one-to-one compare, you’re ready to try the ‘One-to-many’ matches. Go back to your home page and click on that utility. Now just enter your GEDmatch kit number and click on Display Results. Please read the explanatory material at the top – all of it is important. Then scroll down to see your closest 2,000 Matches, arranged by closeness. For each one you’ll see the Kit Number; Autosomal and X-DNA data; each Match’s name/alias, and their email. Maybe you’ll recognize some of the top Matches from your testing company… Now click on the hyperlinked A in the Autosomal/Details column – you’ll see the one-to-one comparison page come up with your two kit numbers already filled in – just click on Submit to “see” the shared segment(s), as described above.

Put on your genealogist hat and email any of your Matches and share Trees and info to discover how you are related. To fully use the DNA data, read my blogposts about Triangulation. It takes a while to get up to speed on the DNA analysis, so I highly recommend using your genealogy hat for a while and get to know your cousins….

If you are trying to relate cM values to cousinship, there is a wide range of possibilities. Check out the August 2017 chart at ISOGG here.

 

ADMIXTURE – Try a test run. On your home page, click on “Admixture (heritage)”. Select, say, Eurogenes, and click on Continue. Enter your kit number in the box; and click on Continue. Look over your results. Go back and try different parameters – each one will give different results. Such is the nature of admixture analysis – different utilities have different reference populations and algorithms. Don’t take any of them as gospel. Have a little fun and try the “Archaic DNA matches” to see how close you are to some ancient people, like Clovis man. Or try the “Are your parents related?” utility.

 

Comments to improve this post are welcomed.

 

Permission is granted to anyone who wants to include a link to this blogpost in their message to Matches just starting with GEDmatch.

 

[21B] Segment-ology: Getting Started with GEDmatch; by Jim Bartlett 20170919

16 thoughts on “Getting Started with GEDmatch”

  1. Hi Jim,
    I have been using the new Gedmatch Genesis overlap tool recently. It compares the number of common positions that Gedmatch can use to match two kits. Different companies test our DNA in different physical positions on the chromosomes so only the common positions can be used in matching two kits from two different companies. Even kits tested with the same company at different times may have a different spread of positions where a match test can take place.

    The reason I am writing is that the tool suggests that certain combinations of kits from different companies or different vintages of kits from the same company will yield a higher level of false positives simply because they have a reduced overlap on common positions to test for a snp match. It seems an excellent number of common positions is around 600,000 but this drops off say by comparing a new 23 and me kit with an older ancestry kit to only 150,000 common positions . At this low number I believe the mismatch probability goes up considerably.

    The colour coding on the results suggest that over 300,000 common position are necessary for the match process to perform well anything less may be a problem.

    I did some segmentation work and two kit match work recently and all the resulting overlaps had only 150,000 common positions and all were 23 and me kits.

    So my questions are;
    1: Are you aware that we may have a problem from incompatible numbers of common positions?
    if so is there a safe limit, like 300,000 to be more certain of a match,
    2: This level of detail is invisible on Gedmatch but is it still an issue there or are we dealing with Genesis still being in Beta and they will work it all out .

    All help as always gratefully received !
    Cheers Steve

    Like

    Reply ↓

  2. Jim, you’ve given us a lot to think about in your BLOG – Thank you.
    I’ve used the GEDMATCH tier 1 tools to triangulate where AncestryDNA matches are on my tree. I’ve trying to understand their new TG tool, but notice it only pulls in a much smaller number of kits to form their TG. I think I could do many more if I understood the way they create their TG label.(i.e. c52). I get that the “C” is chromosome 3, but how is the “52” created? If I knew this, I could sort on this field in my spreadsheet of all AncestryDNA matches, Thanks again for your help,
    Terry

    Like

    Reply ↓

    • Terrance, It is my understanding that the TG lable is not for us – it’s a programming artifact to indicate where to place the box on the page…
      I sort my spreadsheet by chromosome and Segment Start location. I then compare adjacent (overlapping) segments to form the Triangulations.

      Like

      Reply ↓

  4. Pingback: DNA Painter – a new tool | Anne's Family History

  5. Hi Jim,
    I really enjoyed your article! Do you know or can you tell where I can find answers to my Gedmatch problems:
    I am a GEDmatch user and have a problem of not being able to locate someone on the One-to-Many Match List who was listed previously and they don’t come up doing the “people who match 1 or both of 2 kits”. However, they do come up doing the “1:1 compare” option. Also, this 1:1 compare indicates our MRCA of 3.7, however, in the “All Match” list my lowest MRCA is 4.2. I was not able to find answers to these questions on the GEDmatch site. Please help. Is there a telephone number that I can call?
    Thanks so much!

    Tom Lackner

    Like

    Reply ↓

    • Tom – the website email is GEDmatch@gmail.com . I volunteer there and answer some of the emails for them. Each of the utilities has it’s own algorithm and default values. Each set of values going in gives a slightly different output. The people who match 1 or both… utility will only list the top Matches, not all of them; and the lowest ones will drop off as closer ones come on the list. Using the Tier 1 utility for 1 to Many you can list the top 10,000 if you want. On the regular 1 to Many you can see more Matches by sorting the list on various columns. The Gen estimate is just that, a relative estimate, and beyond about 2C it drifts away. a 3.7 or 4.2 could be a 4C, 5C, 6C or greater – don’t put too much stock in that estimate. Mostly use GEDmatch to find Matches, and -most importantly- compare them to each other to see if several overlapping segments are on the same chromosome side. There is a LOT of info about using GEDmatch on the left side of your home page. Jim

      Like

      Reply ↓

  8. Jim, I tell people to start with the One-to-Many comparison in order to see a list of their own matches. Then for each match, they can click on the “L” to see a list of the match’s matches, or click on the “A” to see in detail how they match that person. I also tell them the beginner’s tutorial in the “Learn More” section is very useful. I point out that unlike on Ancestry, on GEDmatch anyone can see anyone else’s matches, and if they know someone’s e-mail address, they can look up their GEDmatch number.

    It took me a long time, but just recently I realized that when I am looking at any match list, if I want to do a one-on-one compare, instead of going back out to the home page, I can just randomly click on any “A” and get to the one-on-one input page.

    Like

    Reply ↓

    • Julie, My post was focused on folks who have just uploaded, and cannot do one-to-many yet. I’ve already given them my kit #, so they can get started. Since I use a master spreadsheet, the A is my go-to click. However, I still say noodle around – each person develops their own process. Thanks for adding your tips. Jim

      Like

      Reply ↓

  9. I enjoy and learn from every posting of yours. And I encourage as many of my connections to join Gedmatch and your blog. It helps make a seemingly complicated process more doable. I am still in the early stages of learning but I am also convinced it will solve several of my stone walls for me once I get started in the right direction. thank you Jim for helping so many. Pat J

    Like

    Reply ↓

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.