Cluster Your DNA Matches | GEDmatch TUTORIAL Genetic Genealogy

[@FamilyHistoryFanatics]

👨‍🏫 Learn 3 ways to cluster your DNA research 👉🏼 kzbin.info/www/bejne/nqHdgp2nfNKCm9E

[@cathyc6725]

Definitely a must watch for anyone using the clustering tools on either Gedmatch or any other site. I can't wait for the next video on this!

[@FamilyHistoryFanatics]

Thanks so much. Glad you enjoyed it. Please share it with others to further show your support.

[@kerinstinear8662]

Fabulous tutorial. Thank you.! . You may have just helped further my research on a long standing mystery. See you when I come out of the rabbit hole!

[@FamilyHistoryFanatics]

Awesome. Keep me posted.

[@peterwhitehead7271]

Purchased these tools Andy and they are very good. I’ve now phased and my raw data is now a super kit

[@maxtrenck8511]

Wow! I’m beginning my dive into the DNA study. I was unaware this existed. Thanks for the great presentation!

[@FamilyHistoryFanatics]

Glad it was helpful!

[@mariacapaldi5062]

I so appreciate that RootsTech does the "Live Streaming" I have watched passed years and haven't been able to attend the Event. I am hopeful of attending at some point in my life, fingers crossed. I am still debating about the Virtual Pass..

[@mariacapaldi5062]

Oops it switched over, this is a comment to Devon about the RootsTech Conference , sorry

[@FamilyHistoryFanatics]

No worries. (Can you go give Devon some love on that video? She often thinks few folks like her videos.)

[@paulbaltzer4745]

Great video on Auto-Clustering tool.

[@FamilyHistoryFanatics]

Thanks, make sure you use the principles to your advantage.

[@tjklassen]

Thank you for bringing this feature to my attention. I especially appreciate your comments related to endogenous ancestors - which is a perspective that I must continually deal with (100% endogenous) I'm curious to see how it reacts to my matches.

[@FamilyHistoryFanatics]

This is a tool to definitely play around with.

[@wesleyjohnston2667]

Going to the 1to1 really matters -- I set my lower cM to 20, and the kit at the top of the largest cluster (By Cluster Size) shows 40.7 cM with me, but clicking on the 40.7 cM reveals "No shared DNA segments found". Dropping the 1to1 threshold from 7 cM to 3 cM reveals a lot of small shared DNA. I would like to have parameters not just for total cM but also for largest segment cM.

[@FamilyHistoryFanatics]

Excellent suggestion, if you haven't emailed GEDmatch, I would do that.

[@bluesaris]

Myheritage now offers a similar tool. I was wondering, how is it grouping people in a cluster. What is the determining factor. To be in a cluster do you have to match with more then "x" amount of other people in the cluster etc?

[@FamilyHistoryFanatics]

Yes, the limits are usually set to somewhere around 30 cM to 400 cM.

[@patricksweetman3285]

@@FamilyHistoryFanatics It looks like it is clustering on shared segments rather than on total shared cM.

[@MathCabin]

Hey, thanks for the video. I have 16 clusters on myheritage. The min threshold is 15 cM. My friend has 33 of them with min threshold 20 cM. So myheritage knows the best threshold for a user and each report varies? Does it also mean these 16 clusters I have are my ggg-grandparents? I know there are 32 in total, but I know there is always room for more clusters. Any idea?

[@jamespearson8341]

Thank you, Andrew

[@FamilyHistoryFanatics]

You're welcome

[@Iwalanijoy50]

I tried clustering for the first time and the search came back with not enough kits to perform a cluster. How do I overcome this issue. I played the upper and lower thresholds but still got nothing.

[@daveb1576]

I'm trying to obtain a cluster diagram that has a particular GEDmatch kit in it. The kit is a 10.8 cM match to the cluster kit person. I've tried several thresholds but fail to find the 10.8 cM in the clusters. What is the trick for forcing a particular match into the cluster? Thank you.

[@FamilyHistoryFanatics]

Manual clustering (using shared match lists), it is time consuming but the only way to get the info you want.

[@gnarfgnarf4004]

You need to explain how to interpret the clusters.

[@FamilyHistoryFanatics]

That's a great follow-up video idea. I've added it to my list.

[@anna-karins1176]

Great video Is there a possibility to get the HTML output to a spreadsheet for example Excel?

[@FamilyHistoryFanatics]

I would try and copy the whole page and paste it in Excel. If that doesn't work, you would have to get into the coding of the HTML with a third party program.

[@howardforer8794]

the gedmatch auto cluster tool no longer lets you enter a lower and upper threshold the largest lower # is 35 this doesn't work well for me due to endogamy can you please offer any suggestions? - thanks

[@FamilyHistoryFanatics]

Endogamy is a pain (as you already know). I don't have it in my family and I haven't worked with any clients that have it so I am very unfamiliar with how to deal with it. I would suggest looking up Leah Larkin. She has some info on her website about endogamy. thednageek.com/blog/

[@teetee9315]

When you use the compact segment tool; does all the color lines on the chromosome indicate a common ancestor? If so, is this similar to triangulation?

[@FamilyHistoryFanatics]

The colors are associated with specific kits that match the main kit. This is the first step of triangulation, but it is not complete triangulation.

[@LSBurchette]

Will there be a new video with the Hybrid Auto Segment?

[@FamilyHistoryFanatics]

Eventually, I have lots of things to catch up on.

[@sandywallace1154]

Is there any way to use the clustering tool when you belong to one of the Gedmatch forums? I'd like to run it against my matches within a specific group.

[@FamilyHistoryFanatics]

You would have to put all of the members into a tag group

[@maryriser7836]

How do I print the cluster map out?

[@FamilyHistoryFanatics]

You can't (other than doing a "Print Screen"),

[@thomasmanson1119]

Andy, it looks to me that clustering will have the effect of losing matches that stood out previously because of the amount of match. For instance, my wife’s top match in the one to many was a relative that shared 130 cMs, and the second on her list was an 85. Then it dropped down t0 the 50s and 60s and kept dropping slowly (until I lost interest) LOL. So the clustering might miss a lot of potential due to the thresholds. Can you address that?

[@FamilyHistoryFanatics]

Clustering doesn't consult close matches. It clusters your more distant matches. It's a tool that can help you visualize how they are related. So, start by identifying your close matches. Then leverage your more distant ones. It's just a tool in the tool box.

[@junegenis]

I am Ashkenazi and when I used the defaults I got one huge cluster with 483 matches. Following your suggestion I change the thresholds to 20 and 30 which produced something more reasonable but which had some strange things as well. The first four cluster ranged from 2 per side to 4. The next 9 x 9 but it included both green and purple (lavender?) boxes. There were other clusters with mixed colors as well and colored boxes that seemed to be floating off on there own but might have really been another mixed color cluster I suppose. Then I got a huge cluster as you predicted. I thought that would be the last one but scanning further I got another bunch of small clusters after it. Some of these were also mixed colors. What can you make of this and what do you suggest I try next?

[@FamilyHistoryFanatics]

Start trying to figure out who the common ancestors of each of the clusters is.

[@patporte2134]

i tried to save the graph your way, but all i saved iar the directions ! i see nothing else on the html (full screen), no arrows to move around..... what am i missing ?

[@paulbaltzer4745]

The "save as..." function may not work with a Chrome browser. Try to use "Save as type" of "Webpage, Complete".......or use a different browser like Microsoft Edge to "save as." Then your saved HTML file will open in either browser in the future.

[@alanheadrick7997]

I have not found any cousins so far that are related to each other. So my guess then is this tool would not work in my case?

[@FamilyHistoryFanatics]

I wish I could answer you definitively, but I can't without seeing more information.

[@mariareyes-qj2od]

what is the triangle in the single cluster means?

[@FamilyHistoryFanatics]

It means that those two people and you triangulate.

[@MxGardiner]

Hello, I have been watching your videos, trying to find the biological parents of my great grandmother. I'm only missing two of my 16 second great grandparents. I'm still having a hard time determining who they are.

[@FamilyHistoryFanatics]

Best of luck to you. I hope that DNA can be helpful in your research.

[@peterb445]

Hello can you please explain how to save clusters when there is no right click option to save on my version of Gedmatch please (I am in Australia) thanks

[@paulbaltzer4745]

The "save as..." function may not work with a Chrome browser. Try to use "Save as type" of "Webpage, Complete".......or use a different browser like Microsoft Edge to "save as." Then your saved HTML file will open in either browser in the future.

[@FamilyHistoryFanatics]

No right click? Are you using a windows or an apple system?

[@a.d.chambers5706]

What is the web address for genesis gedmatch?

[@FamilyHistoryFanatics]

www.gedmatch.com/

[@citizenprayer5644]

I appreciate the video and the information, but your recording is a bit slow in the frames vs audio :(

[@FamilyHistoryFanatics]

Thanks for the feedback

[@cathyc6725]

I'm not exactly sure I fully understand what Cluster by average cm to Reference kit means. How doe sit differ from order by kit cms with the reference kit? I see they produce very different results, but I'm not sure I understand the reason.

[@FamilyHistoryFanatics]

The average looks at all the kits in a cluster and averages the amount of cM. Then it orders clusters by this average number. Order by kit cM looks at the cluster that has the highest individual kit cM, the next cluster will be the next highest, etc. So if you had a bunch of matches with 30cM and 1 with 200 cM, then that cluster would be at the top.

[@videoman77365]

what about CRI Genetic

[@FamilyHistoryFanatics]

They don't add anything to the genealogical research, so I don't recommend them.

[@bayyinahzhaxx7620]

From what I understand the government and local authorities have access to gedmatch. People are literally giving up their freedom, and others without their permission. Interesting, the new ways of the world.

[@FamilyHistoryFanatics]

Please watch this video to better understand how much government and law enforcement gather from your DNA, if anything. kzbin.info/www/bejne/imqoladsi8qqY8U

[@Ione1916]

No! Don't go away yet! Come back!

[@FamilyHistoryFanatics]

New videos every Tuesday and Thursday

[@GarryVC]

Thanks, Andrew :) R1b1a1a2a1a1b x U5a1a1 :D You find them in Belgium :p

[@FamilyHistoryFanatics]

Awesome