Tuesday, January 14, 2020

How Copying Errors Can Really Screw Up ThruLines

I want to illustrate how ThruLines is only as reliable as the family trees it's using, and how even if you appear to have a few DNA matches linking you both to the same ancestor, that doesn't make it accurate. As we know, it's common for the inexperienced genealogist (of which there are many) to blindly copy data from other trees without verifying it. All it takes is a handful of people copying the same error for ThruLines to make a wrong connection that might seem accurate because there's more than one DNA match.

My dad (JB, shown above) is a DNA match with two people, RJ and RH (you can ignore MM in the screenshot above, that's a legitimate connection I've verified). ThruLines suggests that they are both connected to my dad via his ancestor, Giovanni Biello. It does not include Giovanni's wife, so these are allegedly half cousins, but to my knowledge, Giovanni was only married once. I was open minded to the idea I might have missed another marriage, but then I noticed something else.

RJ and RH both descend from someone named Denizi Biello, b. Feb 1862 (according to the 1900 US Census), supposedly the son of my ancestor Giovanni Biello, b. 13 Jul 1847. This would mean Giovanni was only 14 when Denizi was born, which might be biologically possible but it's extremely unlikely. Men didn't marry until they were old enough to support a family, which they would not be at 14, and if they had children out of wedlock, the child was often left at the church as a foundling (meaning the father not identified). So something about this just didn't sound right to me.

Yet, 4 people had added Denizi as the son of this Giovanni to their tree. And in their defense, Denizi's death certificate does indeed say his father was Giovanni Biello and his mother Domenica Scioli.

Denizi's, or Dionisio Biello's civil birth record
Digging through the civil records of Monteroduni, Italy where both men were born, I found Denizi's birth records which finally held the answer. For starters, his original name was Dionisio Biello, and he was actually born several years before 1862, on February 2, 1856. But his parents names were correct: Giovanni Biello and Domenica Scioli. Only, Giovanni was not born anywhere near 1847. He was recorded as aged 38, which would make him born about 1818. It also said Giovanni was the son of the late Dionisio, whereas "my" Giovanni Biello was the son of Lorenzo.

So, two completely different men, as I suspected. I do not know who originally made the wrong assumption that Denizi was the son of "my" Giovanni but then 3 other people copied the error, ThruLines picked up on it and then found two people descended from Denizi. Since there were only 4 tree with the same error, I contacted all of them to let them know - hopefully they'll make the change and the incorrect ThruLines will disappear, but imagine if this error had been copied by more than 3 people! It probably wouldn't have been worth sending a message to each one since many likely wouldn't get the message or make the correction.

Granted, I have to admit that Monteroduni is a small town in rural Italy where there is a ton of endogamy and cousin marriages. I'm pretty sure everyone with ancestry in Monteroduni is related to everyone else there in some way, but it's still important to figure out the correct relationships whenever possible. Biello is actually a very rare surname and given that and the same location, our Biello lines probably intersect at some point, it's just a question of whether it's before or after civil records began in 1809. I'll certainly keep looking.

Monday, January 6, 2020

ThruLines vs DNA Circles

The most common question I see about ThruLines is whether it uses DNA, or family trees, or both, and there seems to be a good deal of misunderstanding and confusion about ThruLines, especially in regards to how it compares to the now retired DNA Circles. The best thing I can say is that ThruLines does things with trees that DNA Circles didn't do, and DNA Circles did things with DNA that ThruLines doesn't do.

I'll start with what DNA Circles used to do.

A screenshot from Ancestry's blog of how DNA Circles
showed everyone in the group shared DNA with each other 
DNA Circles would first look for a group of people who mostly all shared DNA with each other. So let's say you shared DNA with A, B, and C and on top of that, A, B, and C all shared DNA with each other too. In addition, you did not share DNA with another person called D, but D did shared DNA with A, B, and C, and so was included in the group. So not everyone in the group had to match every single other person, but had to match enough people in the group to justify including them. Once this DNA group (or circle) was established, the system would then look for a common ancestor among your trees - and only among the trees of the people in the group. Once the common ancestor was identified, people in the group who didn't have this ancestor in their tree yet (or didn't even have a tree to begin with) would get DNA Circles in the form of "New Ancestor Discoveries". Within the tools of DNA Circles, you could see who all was in the group and who shared DNA (and who didn't) with whoever else (see above right). (Note: I think the minimum for a Circle to be created was 7 people not too closely related to each other, not 5 like the example I'm using, but I reduced it for the sake of ease).

So, DNA Circles was primarily looking at the shared group DNA and only using trees to identify the source of all that shared DNA (which is actually much how Genetic Communities work too, but I digress).

ThruLines works sort of in the completely opposite way.

ThruLines is only showing these matches descend from the
same ancestor based on trees - it does not tell you whether
they share DNA with each other or not
ThruLines looks for a common ancestor between you and an individual DNA match by looking at the entire database of (searchable) family trees. This is something DNA Circles didn't do, because it only looked at the trees of those in the Circle. So ThruLines is taking match A and looking for a common ancestor by trying to compile all the data from available trees (not just the tree of you and match A). It does this for each DNA match individually, so it then separately finds match B and C are also supposedly descended from the same ancestor and it groups A, B, and C based on what the trees say is their shared ancestor, regardless of whether these matches also share DNA with each other or not (and it doesn't even tell you if they do or not). ThruLines does not check to see if A, B, and C also share DNA with each other like DNA Circles did, and it doesn't even involve D because you don't match D. You can sometimes check to see if they share DNA with each other yourself by using the Shared Matches feature, but remember only estimated 4th cousins or closer are included in the Shared Matches list. If you share less than 20 cM with A, B, and C, you can't see whether they share DNA with each other or not.

What does this mean? It means that beyond the fact that ThruLines is only looking for tree connections with your DNA matches, DNA really isn't a part of ThruLines. The groupings are not based on DNA like DNA Circles were because ThruLines doesn't even know, much less show you if the people in the group share DNA with the others in the group or not. Knowing which of your matches also match each other is important for establishing a connection to an ancestor because remember, family trees are fallible and you can't rely on them alone, especially when there's also no known paper trail to confirm it. You need those DNA groups/circles to tie those alleged descendants together and confirm what the trees say, but ThruLines doesn't do this.

Another screenshot from Ancestry blog showing the now
retired New Ancestor Discoveries.
Another thing ThruLines doesn't do, because it's primarily working off trees instead of primarily working off DNA like Circles did, is provide you with Circles, or New Ancestor Discoveries, even if you don't have a tree attached to your test results (shown right). Because ThruLines is working off trees, it needs a starting point in your tree to connect it to other people's trees. Without at least a basic tree with your parents in it (and ideally they want at least 4 generations), you will not get any ThruLines at all. But because DNA Circles was firstly looking at groups of shared DNA, it didn't matter what your tree said or whether you even had a tree, it could tell you that you fit into that established DNA group based purely on DNA. While ThruLines does include "Potential Ancestors" not yet in your tree, this is again based on what trees say and you will not get any if you don't have a tree attached to your results. This is a drawback for adoptees and people looking for unknown biological family since they have no biological tree to add.

Hopefully that helps clarify the main differences between these two tools. They are/were both useful in their own ways, but they are different and understanding their differences is important so you're not making assumptions about ThruLines and letting it lead you astray.

Wednesday, November 27, 2019

AncestryHealth Results are in!

Recently, Ancestry.com announced the introduction of AncestryHealth, providing health and wellness reports for the first time. It costs $49 on top of the original DNA test price, and they are also offering a subscription (currently by invite only) that will include ongoing updates and new reports added in the future. This is very different to how other major testing companies offer health reports, subscriptions are not the norm and frankly I think it's a little cheeky from Ancestry, and I will not be paying the subscription.

But I did pay the one time fee for the current reports to see what they're like and the results are in. For people who tested particularly on the old v1 chip, your DNA has to be rerun through the lab because the raw data isn't ideal for the health reports they're offering, so it took a little bit of time for the results to come in.

AncestryHealth is offering 4 types of reports: Cancer Risk, Carrier Status, Heart & Blood Health, and Wellness Reports.

Under Cancer, there's currently only 2 reports: Hereditary Breast & Ovarian Cancer Syndrome, and Lynch Syndrome. They are careful to note that "No DNA differences found, but other factors may increase your chance of developing cancer." They also detail the fact that their test only includes 27 of the more than 2,400 DNA differences in the BRCA1/BRCA2 genes. None of the genealogy related DNA tests that offer health reports include all known SNPs associated with cancer (or any other given condition). The only consumer DNA test that does is Color.

Carrier Status includes 3: Cystic Fibrosis, Sickle Cell Anemia, and Tay-Sachs Disease.

Heart & Blood Health has 4 reports: Cardiomyopathy, Familial Hypercholesterolemia, Hereditary Hemochromatosis, Hereditary Thrombophilia.

And finally 8 under Wellness Reports: Beta-Carotene, Caffeine Consumption, Lactose Intolerance, Omega-3, Vitamin B12, Vitamin C, Vitamin D, Vitamin E.

Unsurprisingly, there aren't many reports yet and it doesn't tell me very much, at least not much that I didn't already know. It was much the same when 23andMe first reintroduced their health reports too, the only difference is, 23andMe continue to update and add new reports at no extra cost whereas AncestryHealth are charging a subscription for it. I'm not really sure why Ancestry think that this will be competitive pricing, as it will be far more expensive for users in the long run. MyHeritage recently joined the health band wagon too, and like 23andMe, do not require a subscription for new reports.

So is AncestryHealth worth the extra $49? Right now, not really. Maybe in the future as they add more reports, it will become more useful but it's unclear if merely buying the test in the future will gain you those new reports that I can now only get if I subscribe. Seems silly to punish people for being early customers, but at the same time if AncestryHealth Core (the non-subscription) remains the way it is with only 13 reports, it will be a pretty pathetic health report. Ancestry haven't been very forthcoming with the details on the subscription though. Unless they decide to drop the subscription idea altogether, I doubt very much I'll ever recommend AncestryHealth at all. 23andMe provides much more information at a lower cost (in the long run), but remember, even 23andMe is not comprehensive. Even Promethease, with it's hundreds of reports, is not comprehensive because the raw DNA data it's working off of is not comprehensive and only includes a small portion of our full genetic profile.

Monday, October 21, 2019

Yet Another AncestryDNA Ethnicity Update

AncestryDNA's latest ethnicity update
AncestryDNA had promised another update to their ethnicity report this year, despite there only have been one last year (which was still being rolled out the last people earlier this year too) and it seems to be rolling out already, with little to no beta test like last time.

The reference panel has now jumped up to 40,017 samples, but don't assume that means it's going to be more accurate. As ever, it depends on the individual, and just how admixed you are. Also, unlike last year's update, we're not seeing any new percentage groups in Europe (though I do believe there's new ones in the Americas), just a change to the percentages (though there are some new Genetic Communities).

For me personally, the report just keeps getting worse and worse. They have now removed my Scandinavian results entirely, which is a shame because I did have one Norwegian great grandfather so I should be roughly 12-13% Norwegian. Of course, I may have inherited less than that from this particular great grandparent (and most ethnicity reports do reflect that), but I did inherit something from him (a great grandparent is not distant enough to not inherit anything), so the complete wipe away of any Scandinavian result at all (not even anything in Sweden, let alone Norway) is incredibly inaccurate and a shocking degradation.

Meanwhile, they've increased my French results even more to 21% (was 18%), even though I still don't have any recent French ancestry (I had a few French Huguenot ancestors from back in the 1600s, but we're talking about 2 or 3 out of 1024 or 2048 8th-9th great grandparents - a drop in the water that is generally too diluted to be detectable, and certainly shouldn't be as high as 21%).

They do appear to have removed a bunch of the low percentages that were likely noise, or coming from neighboring regions. That seems like a plus at first, but since my valid Scandinavian results were among them, it doesn't seem like a great improvement.

The other only changes were the swap of Germanic and England/Wales/NW Europe in terms of percentage amount. Previously, I had highest results in England/Wales, which was consistent with my tree, now I have highest results in Germanic.

My Italy results remained my the same, which might seem like a positive (at least they didn't go down), but they are remaining stubbornly at 12% when I should be about 32% (and this one I know for sure because my paternal grandfather tested and I share 18% with him, leaving me to share 32% with my Southern Italian paternal grandmother). I suppose the rest of my Italian DNA is showing up in France. If you look below, the map for France does cover Northern Italy, but of course I don't have ancestry in Northern Italy, only Southern, and AncestryDNA are the only ones who seemingly can't figure this out (well, they are the only ones trying to give France it's own category apart from LivingDNA which more accurately give me 0% in France).

In addition to these changes, we've also seen a redefinition of the maps. Although the categories haven't changed, they've just altered the maps to better reflect the regions each category is supposed to cover. At the same time, they've taken away the page that included more in depth details and background on each category, which is a major regression if you ask me. We now only have a short paragraph with info on the region and it no longer lists the "secondary" areas that each category covers, even though the maps still include them.

As you can see above, there is no longer a link to click for more information. AncestryDNA keep removing more and more details and information, as though they think we are too stupid to understand it. People are simply now going to have more questions and understand the results even less.

Also note how the map for France still shows the surrounding areas that this category includes, but for some inexplicable reason, AncestryDNA has decided to only include the "Primarily located in" information, which only lists France, and no longer lists the "Also included in" areas shown in the map, like Spain, Italy, etc. That won't be confusing for people at all.

I really don't know what they are thinking sometimes, and since they rarely ever explain or communicate their choices to their customers (apart from "we think this is better"), I doubt we'll get an explanation. We still never even saw the return of the Average Admixture chart (shown left) when it was lost in last years update, which was so useful for understanding what regions shared a lot of DNA with what neighboring areas (God forbid they actually admit that is a thing).

Having said all that, the report does seem to have improved for people who are less mixed, just like last years update. My half Italian dad's results went from 44% Italian (which is reasonably close to half) to exactly 50%. The rest of his results didn't change much, which means he still has too much in England/Wales and not enough in Germanic (only 5% Germanic when he should be more like 30%).

My paternal grandfather's results have evened out a little bit - his tree is roughly half German and half Scots-Irish (maybe some English), and his update says 51% England/Wales, and 31% Germanic, a slight improvement from 67% England/Wales and 23% Germanic.

My husband is the least mixed of all. He is a British native, born and raised, but half Irish. Grouping his ancestry the way AncestryDNA does, he'd be about 40% English and 60% Irish/Scottish. Last year's update gave him 38% England/Wales and 61% Ireland/Scotland with a random noise level of 1% in Benin/Togo. Today's update has him at 43% England/Wales and 56% in Ireland/Scotland. Not much of a change and still pretty consistent with his background, apart from that random 1% they've now put in Nigeria.

As for my mom's results, her previous update from last year saw no results whatsoever in Germanic even though she should be around 20%. Meanwhile, her Norwegian results, which should have been around 25% (that Norwegian great grandfather of mine was her grandfather) were inflated at 40% last year. She now had 10% Germanic, which is better than nothing, and 27% in Norway, which is almost spot on. The only thing a little off is the 14% in Sweden.

How have your results changed? For the better or worse?

Friday, October 11, 2019

Breaking Down More Brick Walls with DNA and ThruLines

Previously, I detailed how I broke down one brick wall with the aid of my DNA matches, involving a suspected maiden name. This time, I'm going to detail how it helped me prove that my ancestor Jonathan Deaves was the son of Isaac Deaves.

My 3rd great grandmother, Mary W Deaves was always a huge brick wall in my tree. When I first got into genealogy, I started out where my grandmother had left off with her research. She did not even have Mary's maiden name. I was able to find Mary's death record (shown left) which said her father's name was Jonathan Deaves. That at least gave me her maiden name, and her father's name, but there was no mother's name and for years I was not able to find out anything more about her father. I had no idea when he was born or died, though I assumed it was probably somewhere in or around Philadelphia since that's where his daughter lived. I had no idea who he married or when or where, or who his parents might have been. By the time of the 1850 census, Mary was already married and I couldn't narrow down any Jonathan or Jon/John Deaves, Daves, or even Davis in census records as conclusively the one I was looking for. So I'd made a little headway but I was stuck again, just like my grandmother had been.

Deaves seemed like a fairly unique name, although it is also written as Daves or Davis sometimes, which are very common, but I thought I'd give my DNA matches a search for the more unique spelling. Three matches showed up, and they were all descended from an Isaac Deaves or his father, Thomas Deaves. Looking into the records, Isaac Deaves did indeed have a son named Jonathan (named in Isaac's will) but how could I be sure he was the Jonathan Deaves I was looking for? The location was right but the will didn't mention any of Jonathan's children, or any other information about him (though even if it had, I had no other info on my Jonathan so how could I confirm it?). I couldn't find any other records to help confirm it either and I thought three matches were a good lead, but not conclusive. It didn't seem like quite enough, especially since I couldn't find anyone descended from Jonathan himself.

Then ThruLines happened. And suddenly, it was finding connections I hadn't been able to find. It's showing Isaac Deaves as a potential ancestor! I look again, and I (and my mom, who is one generation closer) now have six matches descended from Isaac, one from Jonathan, and three from Thomas for a total of ten, most of whom also matched each other (an important factor) and most of whom share a 15+ cM segment with me or my mom (meaning they are definitely identical by descent). I carefully went through each one and confirmed their descent from these men - it doesn't mean much to have this many matches to one branch if there's an error somewhere and most or none of them are actually descended from these ancestors. But once you do have this many numbers, it becomes less likely they each have an error, especially when they are descended from different generations in the same line because that means they can't have all just copied the same error.

Interestingly, I also have one other DNA match who is descended from a Deaves family in Glouchestershire, England, which is supposedly where Thomas's father was from (another Isaac), but I haven't been able to confirm this yet and with only one match, I can't confirm it with DNA. Hopefully with more research (either on the paper trail or with DNA), I can confirm or disprove it eventually. For now, I'm very happy with my breakthrough.

ThruLines only picked up on 4 people descended from Isaac, I found the
other 2

Monday, September 23, 2019

How to use Ancestry.com's Sharing Links

Many people, both Ancestry.com users and non-users, seem to be unaware of the company's option to provide a link of an individual record that anyone, even those without an account or subscription, can view, so I think it's important I detail how to use it to get the most out of it, both when sharing and receiving. The transcription is limited and you have to use your browser's option to save the image, but the image (when there is one) is available in full size to anyone when using this link. It's particularly useful to use with records that are only available online at Ancestry.com (and in my screenshots, I'm deliberating using a record from a collection only available online at Ancestry.com as an example). The link also does not expire like it does at some other websites so there's no time limit on accessing the record. So how does it work?

Ancestry.com has a help article with details, however it doesn't cover how to get a shareable link without actually emailing it to someone or posting it on Facebook or Twitter, and doesn't detail how to save the image of the record from the sharing link, so that's what I'm going to cover.

Click the "Share" button and select an option

For Ancestry.com subscribers:
1. Open any record on Ancestry.com that you want to share. Maybe you want to share it on Facebook or other social media, or you're adding it as a source to a collaborative tree like FamilySearch or Wikitree.

2. In the upper right area, click the "Share" button (shown above). There will be three options, one to email someone a link, one for posting it to Facebook, and one for posting it to Twitter. If you're actually doing one of these things, then just select the appropriate option. Be aware that when selecting Facebook, it does not seem to give you the option to post it to one of your Facebook groups, let alone a specific topic within a group. So this is only an option if you want to share it in your newsfeed with all your friends. If you do want to post it in a specific topic in a Facebook group, use the following option instead...

To get a shareable link to copy and paste anywhere, close the popup window

3. To get just a shareable link that you can copy and paste to anywhere, select either the Facebook or Twitter option (the email option doesn't work with this for some reason) but instead of posting it to Facebook or Twitter, close the popup window. Either click the little X in the upper right corner of the popup window, or click "cancel" (shown above). The window will close and then a new popup window will come up with the option to copy a shareable link (shown below). Click the URL once and it should select the link, then right click it and choose "copy". You can then paste it anywhere you want.

Copy the sharing link to paste it anywhere

For non-Ancestry.com users:
1. Whenever you see an Ancestry.com link to a record that includes the word "sharing" in the URL, that means it's a shareable link and you'll be able to view it even if you don't have an Ancestry.com account or subscription.

As you can see, the transcription of the record is limited in the shareable view

2. When viewing the record, if it includes an image you want to save, first click on the image or the "click to zoom" option that comes up over it when you hover over it (shown above).

3. A popup window with the full image will appear - you can zoom in to read it better, but it does not matter what the zoom setting is at to save the image. There's no Ancestry.com option to save the image so you have to use your browser's option. Right click the image and select "save image as" or "save picture as" or whatever similar terminology the browser you're using says (shown below). Then name and save the image to your computer as you normally would - make sure you note where you saved it so you don't have trouble finding it later.

Save the image to your computer

Hope this helps, have fun sharing!

Saturday, August 31, 2019

National Geographic's Genographic Project Discontinued

If you've done a DNA test through National Geographic's Genographic Project and been to their site recently, you'll have seen the announcement of the discontinuation of their project. Kits are no longer available for sale, but if you have one, you can still send it in for processing until December 31, 2019. Existing test results will only be available until June 30, 2020. Here's the full announcement:

"The Genographic Project was launched in 2005 as a research project in collaboration with scientists and universities around the world with a goal of revealing patterns of human migration. Since then, nearly 1 million people have participated in The Genographic Project through National Geographic’s “Geno” DNA Ancestry kits. The public participation phase of this research project is ending and, as a result, effective May 31, 2019, Geno 2.0 DNA Ancestry kits are no longer available for purchase. If you have already purchased a kit, you may still send it in for processing until December 31, 2019. After December 31, National Geographic and our processing labs at Helix and FTDNA will no longer accept Geno kits for processing. 
If you have a Helix Geno 2.0 kit with an expiration date that already has passed, you may still request a new kit from Helix prior to December 1, 2019. A replacement kit fee of $25, plus standard shipping rates, will continue to apply. Replacement kits must also be submitted for processing prior to December 31, 2019. For replacement kits please contact Helix customer service at (844) 211 – 2070 or support@helix.com. 
National Geographic currently plans to maintain this site, through which customers may access their results, until June 30, 2020. We recommend that you download a printable version of your results for later reference through the Print Your Results link available on your individual results homepage. Please see our FAQ page for more information."

Full page screen capture
To be honest, I didn't get much use out of my Geno 2.0 results anyway, and this doesn't hugely surprise me. I would take screenshots or save your results as a PDF, and of course download your raw DNA data. This way, you can retain all your reports and results after they are removed from the site. Since the Geographic Project didn't do much (if at all?) in the way of updating reports and didn't offer DNA matching with other testers, we aren't really losing anything if we save our reports and download our raw DNA data.

To download your raw DNA data, just go to your profile (go to https://genographic.nationalgeographic.com/results/myprofile once logged in) and scroll down to where it says "This CSV file contains the raw analytical output from your DNA sample." Tick the box for "I agree to these terms" and click the "submit" button, then follow the instructions.

As for saving your reports and analyses, there is a print option on the site and in the print preview window, you can select "save as pdf" under Destination. However, be aware that this doesn't save any of the maps for some reason (it even saves the "heat map" text with references to the map without the heat map - some use that is!). To do that, I would highly recommend using a Chrome extension called Full Page Screen Capture to take full screenshots of all your reports. With the extension, you can save each page either as PDF or an image. It's a great little tool that comes in handy for more than just this, and it's very easy to use.

Monday, July 1, 2019

ThruLines No Longer in Beta

It's official, ThruLines is here to stay and DNA Circles have been permanently retired. If you're upset by this, check out my article on why ThruLines really is an improvement.

More than that, they've rolled out a few improvements to ThruLines and the match review page. Nothing major, but small things can have a huge impact on your workflow and several of these new items will do just that.

In ThruLines/Common Ancestors:

When viewing pathways with a DNA match to a common or potential common ancestor, clicking on individuals now first brings up a side bar on the right that lists "Relationship Records", which DNA matches have that individual in their trees, and which other trees (not DNA matches) have them in their trees. You can then click on any of those trees and basic data will appear in the same side bar so you don't even have to leave the DNA pages (though of course if you want to bring up someone's full tree you still can by clicking from the side bar). This really helps to streamline your workflow by not having to open as many new pages.

And let's go back to this new "Relationship Records" feature. In the side bar, it will list what records can be found which identify someone as the parent/child of someone else, helping to prove ThruLines pathways may be accurate and not 100% based on tree data. Of course in the example above the only Relationship Record found was for Find A Grave, which is not necessarily reliable - Find A Grave doesn't exact include source citations and often people get linked to the wrong families, etc. So be careful even with these Relationship Records, but they should be useful in general.

Another improvement you might have noticed is the "evaluate" tag shown on anyone who isn't in your tree already. This helps point out (particularly to newcomers) that ThruLines is only a suggestion and you still need to verify (or disprove) it before accepting it as fact.

On the match review page:

We can now view a 10 generation pedigree preview from the match review page like we used to be able to do before all these changes. Granted, you now have to click on "expand tree" which opens a popup box, and I believe this is only available to subscribers, but it's better than having to open a match's full tree in a whole new tab. And while non-subscribers can only view the 5 generation pedigree preview (so I've heard), that is better than the nothing they had before. So this should be an improvement for subscribers, and a big improvement for non-subscribers.

The preview of basic data on an individual in a match's pedigree is also back. When you click on someone in the pedigree preview, instead of opening their full details in the full tree in a new tab, we see the return of the popup box with vital data: birth, death, parents, spouse, and children. These are the basic things we need to more quickly identify a DNA match without getting cluttered with everything else that might appear on an ancestor's profile in their full tree.

But it's more than just the return of what we had before. If you notice in the expanded pedigree preview above, shared surnames are now highlight in green, something we've never had before! Of course, shared surname doesn't necessarily mean that's the branch which your relationship is from but the incorporation of shared surnames into the pedigree preview could be really beneficial for quicker reference.

There are still a few items I have complaints about, but they are mostly regarding the match review page, not ThruLines. It may be worth noting that the match review page, the match list, and the user profile are all still in beta, it's only ThruLines which has been fully rolled out now.

My main issues are:

1. We are we no longer able to select a tree when someone has an unlinked tree. We can click on it, but it opens the full tree view on another page - in the past, we could select an unlinked tree and instead of opening a new page, it would show the surname list, pedigree preview, and map/locations for that tree as though it was a linked tree, and then we could even toggle among other trees if the user had more than one. This was much more useful and without it, it's harder to compare surnames and locations and find a connection with an unlinked tree.

2. The loss of the Shared Locations list. The map is there but it would be much easier to see a list than have to click around the map. I never really used the map in the past, and found the locations list much easier to use.

3. There should be a distinction between "Common Ancestors" that appear in both my tree and my match's tree, versus ThruLines/Common Ancestors being found through combining multiple other trees together, which is much more speculative. If I have the same ancestor in my tree that a DNA match does, that should take priority and be a notable distinction from other ThruLines.

I guess with ThruLines coming out of Beta, this is unlikely to be implemented, but a girl can dream and hopefully with enough feedback, my other two complaints regarding the match review page might change.

Wednesday, May 22, 2019

Why ThruLines is an Improvement (Really!)

I've been hearing a lot of negativity about AncestryDNA's new feature ThruLines (aka Common Ancestors when viewing your match list - they are the same thing, just different ways to view them), which seems to be replacing both DNA Circles and Shared Ancestor Hints. It works by looking for a tree connection between you and a DNA match through other family trees, not just yours and your match's trees.

Naturally, this opens the system and connections up to a lot of room for error. We all know how inaccurate some trees out there can be, how those errors can get copied by the dozens, and I'm not going to pretend that this isn't an issue. Anytime you're working with trees, you absolutely must verify the information yourself before you accept it as fact.

But that doesn't mean ThruLines can't be useful. I've been working with them for a few weeks now, and I have found that the majority of them can be verified with a little work. Some of them I haven't been able to verify at all and remain unknowns, and others I have actually proven wrong, but the amount of them which I've been able to confirm far exceeds the old Shared Ancestors Hints which were limited to looking at your tree and your match's tree only. ThruLines is finding legit shared ancestors that neither the system or myself doing the work manually could have ever found before. Take the following example.

As you can see, Sally has a family tree with only 3 people in it and on top of that, it's a private tree. In the past, this would be a complete dead end. Unless I contacted her and she responded to tell me more about her tree or at least give me access (which I likely wouldn't do for every match at this distance), I would have completely written off this match in the past.

But now, ThruLines has found a common ancestor! Just by adding those mere 3 people to her tree and making her tree searchable even if it's private, Sally has allowed ThruLines to find a connection between us. Of course, I first have to make sure ThruLines isn't leading me astray, so let's look at the connection.

Part of what you're seeing here is the benefit of researching as far down the descendant lines of your ancestor's siblings as possible. When I first looked at our ThruLines relationship, I had only researched down to Samuel Gross, and Sally's tree only went back to her grandparent. The two generations in between were being filled in by ThruLines finding them in other trees. Of course, to verify the relationship, I had to do my own research and fill in the missing generations, but this was much easier since I'd already gotten a head start. I researched all the way down to Sally's grandfather, Robert. Since her tree was private, Robert was originally shown as private, but you can actually click on deceased people in private trees and get basic details like their name and birth, just like you would from clicking on results of a private tree in the search engine results (see below). If they are still living, you get nothing, as it should be - living people's privacy is always totally protected (also see below).

In the past, if a tree was private, we got nothing unless the tree owner chose to share info or an invite with us. But with ThruLines, we can now see which common ancestor we share and get basic data on each generation even from private trees, enough that we can then go and research the pathway ourselves and verify it, which is exactly what happened here.

ThruLines isn't perfect, of course. It's only as reliable as the trees it's using, and I have found errors in the trees it uses on occasion. Additionally, the system itself is not infallible and I have come across the occasional case where ThruLines is assuming two different people are the same person. In those cases, the error is with the system, not the trees it's using.

But we are genealogists - fact checking and verifying data is what we do. So let's do it! You might find ThruLines just as beneficial as I have so far. It has found common ancestors with people that the old system wouldn't have, and with people I never found have by searching manually on my own.

Monday, May 6, 2019

Are they really a 4th-6th cousin?

At all DNA companies, matching with other testers includes an estimated relationship (usually a range) based on how much DNA you share. The key word is "estimated". There's plenty of examples of finding relatives whose relationship to you actually winds up being outside the estimated range.

AncestryDNA has the option to click on how much DNA you share (when you're on the match review page - click the little "i" icon if you're on the match list page) where previously, it just gave us generic info about that estimated relationship degree. Now (and this may only be available on the new beta match list), it gives us specific probabilities of each possible relationship based on the exact amount of DNA you share. What I found is that sometimes, even their statistical probabilities don't stick within the estimated relationship range.

Above left is an example. This is with a confirmed 3rd cousin of my mom's, an estimated 4th-6th cousin. Again, it's not unusual that the actual relationship is outside the estimated range, but AncestryDNA's predicted relationship range doesn't even seem to be following their own statistical probabilities. As you can see, 65 cM shared has the highest probability (32%) of being 3rd cousins (or 2C2R, Half 2C1R, or Half 1C3R), which isn't even within their predicted 4th-6th cousin range. No prediction system is going to be perfect (because there's a lot of overlap between different relationships so no estimate will be exact), so I'm not saying anything is wrong with the system, I'm just pointing out that this new statistical probability option will tell you a lot more than the predicted relationship range, so don't overlook it.

Of course, this doesn't mean the actual relationship will always be the one with the highest probability. Lower down on my mom's list is a known 4th cousin sharing 57 cM, and the highest probability is still 3rd cousins at 28%. 4th cousins only have a 16% probability, but this match falls into that 16%.

These probabilities will especially be great for unknown close relations popping up, or telling the difference between full and half siblings. My grandfather's half niece, for example, has an estimated range of 1st-2nd cousin, but in the probabilities, it lists all the possible relationships, including half niece/nephew at 96% probability. In fact, at the amount of DNA they share, it doesn't even list 2nd cousins in the probabilities, ruling that out completely even though it's a part of the original predicted range.

DNA Painter offers the same type of probabilities, but collected from multiple companies, not just AncestryDNA data. It's a great tool for matches from companies who don't provide these probabilities, but now that AncestryDNA does, their internal probabilities are likely a better option for your AncestryDNA matches, since the data is specific to them.