Apologies for the delay, folks! The days have been running away with me lately! I’ll now take you back to December, and day 2 of I4GG.
Day 2 – International Genetic Genealogy Conference
The second day of the conference had a fantastic speaker schedule. First up was Cece Moore’s keynote which was very inspiring to listen to. CeCe made the point that ‘citizen’ scientists are progressing genetic genealogy via grassroots efforts, the vast majority of whom have no or very little scientific background. Whereas in the recent past people got interested in genetic genealogy because they were already interested in traditional genealogy, now it’s becoming common for people to become interested in genealogy after DNA testing for ethnicity, and then becoming interested in finding their roots.
CeCe also showed us some amazing comparisons for solved Parabon cases from Parabon’s snapshot phenotyping tool, showing us the original prediction side by side with the mugshot of the perpetrator. In some cases, the likeness was truly eerie. You can see some of the comparisons for yourself at https://snapshot.parabon-nanolabs.com/posters.
Next up was Diahan Southard, whose talk was entitled ‘Genetic Genealogy Behind The Curtain. This was a really interesting lecture explaining exactly how DNA analysis works and why ethnicity results are different at different companies. Diahan also explained something that I thought I understood well, but Diahan’s explanation made so much more sense to me. She explained the difference between DNA shared with a match being identical by state and DNA shared which is identical by descent. Identical by state means that you share DNA with someone stemming from a common ancestor a very very long time ago, or that you are basically from the same place. She also said that if there is a segment which you share with a lot of different people it’s much more likely to be an ‘identical by state’ match rather than identical by descent. As the name indicates, identical by descent means you share a common ancestor.
Ancestry doesn’t have a chromosome browser, so this information doesn’t really help you there, but Diahan explained that actually, Ancestry does some of the hard work for you. Its algorithm, Timber, can predict segments that are likely identical by state and downgrades them so matches that look like only identical by state appear at the end of your match list. If you were ever wondering why a FamilyTreeDNA match who is also on Ancestry appears to have a match with you that is higher at FamilyTreeDNA than Ancestry, this is why.
Diahan also gave a really useful tip that I did know in theory, but hadn’t really thought about. Statistically, no group of five DNA cousins will share the same DNA segment, so if you are seeing that in a chromosome browser, it’s extremely likely that you are looking at an identical by state segment.
A great way to see your own pile-ups (that is, identical by state segments) is at Gedmatch – there is a Tier One utility called ‘matching segment search’ which will show you graphically your matches by chromosome, and on some chromosomes, you can see just how many people share that same segment, it’s a pretty cool utility.
Diahan’s website is Your DNA Guide, where you can find a whole heap of useful information, including an entire video library of lessons for a very reasonable yearly subscription. As well as her website, Diahan travels and lectures around the US teaching people in person about DNA and genetic genealogy. She’s also a consultant with LivingDNA. If you get a chance to hear her speak in person, definitely take it!
Next up was the delightful Shannon Christmas, whose talk was ‘Researching and Communicating with DNA Matches‘. Shannon’s talk not only gave some great tips on how to research your DNA matches but also on how to make sure your profile at the DNA company you’ve tested with is the best it can be to encourage people to contact you. For example, having a public accessible built out family tree, providing an email address in your profile (because sometimes the messaging functions on the DNA companies sites are not the best – Ancestry, in particular, is known to be buggy), and making clear in your profile why you tested and that you are open to receiving messages and will respond. He also gave some great tips about making your messages count, and what not to do (as I’ve written about before on here, giving your entire life story to your new close DNA match isn’t the best plan ever, and it’s not really surprising that I didn’t get responses!). He also gave a really great tip that had never occurred to me before, so I’ll share it here – if you are trying to identify a match on Ancestry, take their profile picture and do a Google reverse image search. Sometimes this will lead you to other social media profiles that that person has, and you can identify them that way.
I was really impressed by Shannon’s talk; it was very informative, and he was very entertaining to boot. If you’d like to find out more about Shannon, head to his website http://throughthetreesblog.tumblr.com/.
Kitty Cooper – first session
After lunch, Kitty Cooper talked to us about the newest updates with GEDMatch. I actually didn’t take many notes here, and most of what she spoke about was Genesis, which is now very much live – it’s no longer possible to upload kits to classic GedMatch, you must now use Genesis, for which the link is https://genesis.gedmatch.com/. Kitty did give us some figures; there were (as at December) 1.2m kits on Gedmatch & Genesis, and somewhere between 900 and 1,000 new kits are uploaded daily. One of the much-discussed themes of the weekend was the rising use of Gedmatch by Law Enforcement agencies in criminal cases. Kitty stated that on the day the capture of the Golden State Killer was announced in the media, about as many kits were deleted as were uploaded, but after that deletions quickly tapered off. Kitty has kindly shared the slides from her talk at http://slides.com/kittycooper/whats-new-at-gedmatch-i4gg-2018#/.
Kathleen Fernandes was up next, talking to us about research methods for unknown parentage cases. I was particularly interested in hearing Kathleen speak as she has roots in the Azores, which, in case you aren’t familiar, is a group of Portuguese islands in the middle of the Atlantic ocean. Due to its isolated region, it’s also an endogamous community, and so tracing Azorean genealogy using genetic genealogy can be challenging. This area is of interest to me as I have some friends who emigrated to the US from the Azores in childhood, and it is a place I have visited and found extremely beautiful.
Kathleen went through the basic methodology that she uses to try and identify unknown parents, which is not unlike the procedure I outlined in my recent article ‘What is Genetic Genealogy’, even though she has the added complication of endogamy perhaps making DNA matches appear closer to her than they are in reality. She first uses the trees of her matches to identify a shared ancestral couple, and to do this, she compares the trees of shared matches, identifies common surnames, and focuses on trees containing those surnames. For example, early on in my search, when I honed in on the surnames Toll, Ledgerwood and Whitenack, I did the same thing to try and find common ancestors; my success with this approach was quite high since the surnames were uncommon. By contrast, my current success rate trying to do this with my own surname, King, is zero, as it’s far too common. That said, Baker is a common English surname, and I actually have had success running a search for that surname in my DNA results, which enabled me to get rid of a brick wall caused by too many George Bakers in Devon for me to know which was ‘my’ George Baker.
Once Kathleen has identified a particular couple she creates a master tree using that couple as the base couple, and she builds out that tree both backwards and forwards to try and uncover further clues.
Kathleen also gave some further clues for searches involving unknown parentage. She recommended ‘fishing in all ponds’, which is a tactic I’ve also recommended; it means testing and/or uploading your raw DNA in all places that you can, since you don’t know where your key matches are going to be. If you are a male, or you have a male who would be willing to test, she also recommends getting a Y37 test from FamiyTreeDNA, if it’s a paternal line you are trying to uncover. This would obviously not work if you are a female adoptee and therefore don’t know any paternal relatives, but could work if it’s a grandfather you are trying to identify, and there is a male-line descendant.
When relevant, she also recommended making sure you test your oldest generation, since they are closer to the mystery. Likewise, if you are conducting an unknown parent search, if you have a known living parent that is willing to test, that can help you exclude matches from your own match list. Extending your own (known) family tree as far as you possibly can both backwards, forwards, and even sideways can help you identify how matches relate to you – I have many matches in my own match list that I’m able to identify common ancestors for without too much research, because I instantly recognise surnames of people that siblings of my direct ancestors married.
These are just some of the most helpful hints that I wrote down from Kathleen’s talk. If you are at all interested in learning more, then I’d definitely consider buying the video for her talk when it becomes available (hopefully this February).
Next up was Tim Janzen, whose talk was entitled Maximizing Your Use of the AncestryDNA Test. And I must apologise again here since, for the second day running, my jetlag kicked in with renewed endeavour at 4 pm, so my notes are a little sparse. However, thankfully I have his slides to refer to, so I can tell you a little bit about his talk! Tim started with an overview of what genetic genealogists need from DNA testing, which is principally: 1) family trees for our matches and 2) matching segment data for our DNA matches. Ancestry does not currently explicitly give us the latter set of data, the only way to obtain it is for your matches to either test somewhere that does give that information or to upload their raw DNA to Gedmatch. However, Ancestry does give us something we can work with, in the form of the ‘shared matches’ page for each of your matches. In theory (but not necessarily in reality due to cousin marriages, endogamy, or simply being related to you in more than one way) all those on a match’s shared matches list likely share one common ancestor or a pair of common ancestors with you.
Tim shared how he uses the ‘star’ function that Ancestry has in-built into its DNA matches feature; he stars each match of his that have either uploaded to Gedmatch or tested at 23andMe, FamilyTreeDNA or My Heritage, because then he knows at a glance that he is likely to have access to segment data for that particular match. He also keeps a master spreadsheet of information about each match, and contacts them to suggest they upload their raw DNA to Gedmatch (note from me and not Tim on this subject – if you are going to do this, I strongly recommend you make your matches aware about the other uses of Gedmatch in case they are not comfortable with this). He also shared with us the format of his usual first messages to his Ancestry matches.
Tim also talked a little about Ancestry’s ‘New Ancestor Discoveries’ feature. Unfortunately, this is something that only a few people have access to and is no longer being updated for anyone, so it’s not terribly useful for the majority of Ancestry users. If you do notice you have a section called this on your Ancestry DNA Home Page (it will be under the first section showing your Ethnicity, DNA Matches and DNA Circles (if you have them)), here’s a screenshot from a test I have access to, so you know what to look for. Tim suggested that if you are lucky enough to have this section, that you contact the matches relating to the Ancestor hint (when you click the name you will get a list of people related to that Ancestor), and again, ask them to upload to Gedmatch so you can work out what DNA segments relate to that specific ancestor.
Tim also explained how the Google Chrome extension, Jeff Snavely’s DNA Helper can be used to download your Ancestry DNA match lists. You can find out more information about this at http://www.ancestrydnahelper.com. The extension can be downloaded here: AncestryDNA Helper Chrome Extension. Unfortunately, I understand that this tool is not currently working for Mac users.
Finally, Tim went into some detail about how to use DNAGedcom, and his personal method of Chromosome Mapping. If you’d like to know more about this, again, I’d recommend purchasing the full talk when it becomes available.
Kitty Cooper – second session
The last talk of the day was from Kitty Cooper, who this time spoke about Using DNA and GWorks for Solving Unknown Parentage Cases. She repeated some of the same advice that Kathleen had given earlier in the day to start off with, but added that you should make sure that your research tree is private and unsearchable. This is important, as, for example, you don’t want to include your matches on your research tree and then have them contact you wondering why you have them in your tree! She also gave a great piece of advice for testing out a particular hypothesis, especially when you are working with Ancestry, and that is to add yourself as a ‘fake’ child of a particular suspected set of common ancestors, and then link your DNA kit to that fake child. What should happen is that after a few days, Ancestry’s hints feature should kick into action, and you should start getting hints for DNA Matches, who Ancestry believes share that common ancestor.
GWorks is a tool at DNAGedcom which automates the process of comparing trees to find common ancestors. Rather than re-hash Kitty’s guide on how to use it, I’d recommend you take a look at her slides, which she has very kindly made available here: http://slides.com/kittycooper/using-dna-for-adoption-searches#/. Kitty’s blog is also a wealth of valuable information and I used it often early on in my own search. You can find it at http://blog.kittycooper.com.
In my next post, I’ll recap the final day of I4GG, I’ll try not to keep you waiting so long next time!