YDNA resolution of LivingDNA

[I1a3_Young]

Based on perusing the results of other LivingDNA users, it appears that LivingDNA doesn't offer better Y haplogroup resolution than the newest version of 23andMe or Ancestry.com tests.

23andMe and Ancestry test 1691 Y positions. LivingDNA claims 20,000 but in an email to me they said the data was 16,000 lines.

So where are all these extra positions and why don't they get better resolution with them?


I ran a comparison using the Morley YSNP sublcade predictor, which compares SNP calls to the 2013 ISOGG tree. This is old as the tree is updated yearly, but it will suffice for this purpose.



LDNA only offers positive calls. As you can see, they test far more UPSTREAM positions, which are totally useless for haplogroup determination.

The best purpose of the additional upstream testing would be for rare variants such as I* or something like that.

 

[Johane Derite]

Based on perusing the results of other LivingDNA users, it appears that LivingDNA doesn't offer better Y haplogroup resolution than the newest version of 23andMe or Ancestry.com tests.

23andMe and Ancestry test 1691 Y positions. LivingDNA claims 20,000 but in an email to me they said the data was 16,000 lines.

So where are all these extra positions and why don't they get better resolution with them?


I ran a comparison using the Morley YSNP sublcade predictor, which compares SNP calls to the 2013 ISOGG tree. This is old as the tree is updated yearly, but it will suffice for this purpose.

View attachment 9060View attachment 9061

LDNA only offers positive calls. As you can see, they test far more UPSTREAM positions, which are totally useless for haplogroup determination.

The best purpose of the additional upstream testing would be for rare variants such as I* or something like that.

The news from them just keeps getting more and more disappointing, I guess I will end up going with 23andme

 

[srdceleva]

There is still the hope that living DNA has a massive update, so far they are a bit disappointing. I'll be happier when gedmatch finally fully accepts their kits

Sent from my KIW-L21 using Tapatalk

 

[I1a3_Young]

There is still the hope that living DNA has a massive update, so far they are a bit disappointing. I'll be happier when gedmatch finally fully accepts their kits

Sent from my KIW-L21 using Tapatalk

gedmatch seems to be using only a fraction of the SNPs but still yielding similar results as the Ancestry.com file. That seems like the LDNA file has very relevant data for the overlapping SNPs.

I posted some of the differences in my personal LDNA thread.

 

[Twilight]

There is still the hope that living DNA has a massive update, so far they are a bit disappointing. I'll be happier when gedmatch finally fully accepts their kits

Sent from my KIW-L21 using Tapatalk

Keep in mind that LivingDNA is only a year old and therefore LivingDNA has plenty of wiggle room for improvement. If you have any constructive criticism or feedback, I'm sure LivingDNA would like to hear it. Here is what the Helpdesk had to say.


Me:
Hello there LivingDNA, thank you so much for uploading my DNA results. I am really exited to see my autosomal admixture; especially since almost half of my ancestry is from Colonial Quakers and Virginians. Anyways on to the question, my Ydna is plain R1b-U152; exactly the same as my 23andme results. Since you guys are committed to giving me the deepest subclass possible, I was wondering what plain R1b-U152 Ydna means. Do I have a super rare deep subclade or something? If so I'd be honored to be the first one to publicly posses this deep subclade on behalf of the Blackledge family.

I got 3.5% Italian but no Italian ancestry recorded, however according to Geneology I have 4.5% French Canadian. And my 3rd Great Grandmother Dora (Pahl) Krumrey was my only Eastern European ancestress, however records say she's from Poland (West Prussia) and lived in modern day Zlotow. Yet I scored 3% Finland & Western Russian/Baltic.

LivingDNA:
The subclade detail of your results will depend on which of the markers that we test for end up positive for your DNA sample. If the markers that we test do not end up positive for your DNA sample, you will not be assigned a subclade.

If in the future we are able to offer you more information and more refined detail within your results, your results will be updated for free. It is possible that in the near future we will be able to assign you a subclade.
I'm afraid I am unable to tell you if you have a super rare subclade.

The markers on our custom Orion chip have been chosen to provide better representation in a wider range of global populations. We have included roughly 20,000 SNP locations on the Y chromosome, and over 4,000 SNP locations on your mitochondrial DNA (mtDNA).

In order to give a better subclade, we have to redesign our entire chip. As this is a very time-consuming project, unfortunately we are not planning to do this in the immediate future, and are instead focussing on other exciting updates to our ancestry platform.

Individuals may want to download their raw data to try and find information or more ancestry with other testing firms and/or find unknown relatives via "matches" by using a variety of third party tools (for example, GEDmatch).

I can see that you have 3.5% Italian ancestry. If you have been assigned anything lower than 5%, there is couple reasons to explain why. Firstly anything below 5% can be referred to as noise in the sample. This means that the sample may of been so small our fine-scale algorithm cannot distinguish the region fully, if this is the case, then your DNA will be linked to the best reference dataset that we have.

Secondly many populations are themselves formed from the mixture of other populations. Our algorithm will attempt to match you directly to a population or region. Sometimes we may not have a reference sample and in some areas we may have a low sample size for the region, or it could be that there is a further population structure within that population/region that makes it near impossible for our our algorithm to match you to a region. In this case we are likely to assign some ancestry to the populations that contributed to that region and so you will see your DNA linked to an unexpected regions.

 

[I1a3_Young]

You aren't getting anything further on YDNA from them. They said themselves they aren't changing the chip. You probably had a zero value in the next U152 subclade SNP location.

Sent from my XT1080 using Eupedia Forum mobile app

 

[Twilight]

You aren't getting anything further on YDNA from them. They said themselves they aren't changing the chip. You probably had a zero value in the next U152 subclade SNP location.

Sent from my XT1080 using Eupedia Forum mobile app

Yes that is true, I won't be able to see my haplogroup deepclade for a little while, what kinds of U152 deepclades does LivingDNA have avalable? Just curious so I can narrow my haplogroup down. If my deepclade has zero value so far, I am more than willing to wait for the deepclade to be studied; if that's the case it's only one bit of my ancestry and my Ydna came from LivingDNA's NW England component so I can look at that component for the time being . 23andme made it to the market in 2006 and LivingDNA is already meeting their standards; as far as Haplogroups are concerned; and this is just the beginning for the company is still has beta results.

Although despite this haplogroup bug I feel blessed of the results I got. If you get rid of the German Ancestry from East Angelia, LivingDNA does place me as 78% British-Islesman; I had rounded up my ancestry before my LivingDNA results came outv I knew I had a little more Islesmen DNA. Now with my results, I can view my English ancestry living in the Middle Ages so personally I'm satisfied with my results so far.
Just like srdceleva, I have high hopes for LivingDNA and hope that Gedmatch and LivingDNA could team up in the future. Granted Gedmatch may have used less snps, however LivingDNA can take samples from the archeological remains they came from.

Basically what I'm getting at, is that our results will evolve over time. LivingDNA has advertised to show us our deep subclades, however no company or even leader can deliver huge promises in one day let alone a year and therefore time is our friend in this case. We'll get what we were promised eventually.

https://my.livingdna.com/share/b1e72f07-85fb-11e7-ab56-5254002fd1a4

 

[Björnsson]

Living DNA is odd and so is Ancestry DNA. LDNA puts me at Z282 and my younger son at Z284, while ADNA puts my older son at Z283, but Gedmatch predictably shows our genetic affinity. If LDNA is accurate and they're using the same Sirius chip, then why would they go deeper for my son and leave me hanging? At least, I know that I am actually Z284, just because my son was tested. On the other hand, I would like to know if we're still basic Norwegian, or the Scottish subclade of Z284, since our forefather made it to Yorkshire in 1066, rather than Scotland. I'm curious if the Celtic ancestry of Scotland caused their mutation, or it is from drift in time and space regardless, throughout the British Isles. I know it's unrelated, but they provide no specific information about K2a5a mtDNA, just some supposedly generic crap about K that doesn't apply to all subclades.

 

[ntindeo]

Living DNA is odd and so is Ancestry DNA. LDNA puts me at Z282 and my younger son at Z284, while ADNA puts my older son at Z283, but Gedmatch predictably shows our genetic affinity. If LDNA is accurate and they're using the same Sirius chip, then why would they go deeper for my son and leave me hanging? At least, I know that I am actually Z284, just because my son was tested. (...)

Did you try to open the file with Y-haplogroup and check all the SNPs inserted there? It is very possible to find a few other subclades in addition to what they assigned you. You can use Ysog to compare.

 

[Björnsson]

Did you try to open the file with Y-haplogroup and check all the SNPs inserted there? It is very possible to find a few other subclades in addition to what they assigned you. You can use Ysog to compare.

That's exactly what I was doing up all night until 0300 this morning. I searched for positions downstream of Z282 for myself and those of Z284 for my son, but neither showed any, just a lot of upstream values clogging up the file space. I think there's a German lab, called Yseq, that might do this subclade isolation, but I'm not sure they take Living DNA or Ancestry DNA files--their upload slot hasn't any options suggesting that is the case. My goal is to find whether we share the original Norse subclade, or the Scottish one, wondering if there's a clear affinity across the British Isles through Føroyar and Ísland.