Coming Soon: Y-DNA Haplogroups for Family Finder

[Riverman]

Overnight the R-L151 Branch kits that have no downstream derived reads grew by 1,368 kits. Total is now 6,112. It's a tragedy P312 and ZZ12_1 weren't tested by Family Finder v2. ZZ12_1 was discovered in 2014 per Ybrowse and FF v2 was released in the Fall of 2015.

I hope a lot of R-L151 people get a Big Y test in order to discover their most downstream public SNP.

edit: By the way, R-L151 is from 3000 BCE which is the Metal Age

No P312 is a real shame. Such a fairly old and extremely widespread haplogroup. But the range within my upstream haplogroup is Middle Bronze Age to High Medieval, depending on exact branch. Some got really lucky, most are intermediate, and some...well. I got the intermediate kind of result but of course, I would have preferred the Medieval version, since even for members of my upstream main branch, chances are about 1:10 to 1: 20 for them to belong to my exact main branch, yet alone being really close (that's probably more like 1:200 to 1:1000 even from the pool of my upstream main branch). 
Paying for an upgrade which chances are that bad is not really a thing, unless the STR values or other aspects make a tester particularly promising.

[Mabrams]

Ha, my Filipino father-in-law (who only did FF) now has a 13,000 year old Y-SNP.  Cool.

Discover:  R-FT354149's paternal line was formed when it branched off from the ancestor R-Z17112 and the rest of mankind around 1450 BCE

Old, but not as old as 13,000

[Mabrams]

Most haplogroups are not very modern.  FTDNA said as much in the initial announcement.


"This means you’ll only receive a partial haplogroup from an autosomal test. Most customers can expect this haplogroup to have originated in the Metal Age (about 10,000 years ago)."

I think they were being very conservative to suggest 10,000 years ago, but that might be true for some people.

That part was weird. The Copper Age is the first of period with metallurgy which started around 7,000 years ago.

Regardless it was easy to see, based on the results from 23andme, Ancestry, myHeritage, and LivingDNA, that most people would get a very old haplogroup. Most of the haplogroups in FF v3 that are younger than 23andme v5 are only slightly younger.

Big Y has always been the better test that leads to a haplogroup within the last 2,000 years in most cases.

FTDNA is not terribly consistent here on the Metal Age.

At the bottom of many Discover pages is a time graph and 10,000 years ago (8000 BCE) is clearly in the Stone Age.  
Or the Ancestral path has the same info in Table form.  

~~

For my SNP-of-Interest, the FF v3 is 100 years newer than 23andMe v5.   500 BCE vs 600 BCE.    LivingDNA v2 was a bit better at 50 BCE.  

My results from Ancestry Raw Data (V1) via YSEQ CladeFinder was U106 at 2950 BCE. I wonder if Ancestry V2s will be much better.

My BigY700 is at 1800 CE. I have a 5th cousin, and we differ by only one Non-Matching Variant, so that result gave us a very recent year. Two clades prior was at 950 CE.

[ArmandoR1b]

Most haplogroups are not very modern.  FTDNA said as much in the initial announcement.


"This means you’ll only receive a partial haplogroup from an autosomal test. Most customers can expect this haplogroup to have originated in the Metal Age (about 10,000 years ago)."

I think they were being very conservative to suggest 10,000 years ago, but that might be true for some people.

That part was weird. The Copper Age is the first of period with metallurgy which started around 7,000 years ago.

Regardless it was easy to see, based on the results from 23andme, Ancestry, myHeritage, and LivingDNA, that most people would get a very old haplogroup. Most of the haplogroups in FF v3 that are younger than 23andme v5 are only slightly younger.

Big Y has always been the better test that leads to a haplogroup within the last 2,000 years in most cases.

For my SNP-of-Interest, the FF v3 is 100 years newer than 23andMe v5.   500 BCE vs 600 BCE.   LivingDNA v2 was a bit better at 50 BCE.  

My results from Ancestry Raw Data (V1) via YSEQ CladeFinder was U106 at 2950 BCE.  I wonder if Ancestry V2s will be much better.

My BigY700 is at 1800 CE.  I have a 5th cousin, and we differ by only one Non-Matching Variant, so that result gave us a very recent year.  Two clades prior was at 950 CE.

FF v3 being only one SNP younger than 23andme v5 shows how similar FF v3 and 23andme v5 are.

I guess Living DNA v2 got a lot better than v1. I found a report that they went with Affymetrix (Thermo Fisher) UKBiobank Axiom Array  Imputation-aware SNP selection, optimized for GWAS, low-frequency SNPs of European and British ancestry. So that must be why it has such a young SNP.

Ancestry v2 is substantially better than Ancestry v1 and has some U106 subclades.

Both 1800 CE and 950 CE are within the last 2,000 years. As seen in this thread there is at least one person with no match in the last 2,000 years. Anecdotes are just that. They don't give a good idea what most people experience.

[Mabrams]

The ISOGG reports LivingDNA v2 with 34,216 Y SNPs. So if quantity was the main factor, LivingDNA wins hands down.

I am not sure their selection is that great. The download is only for your positive SNPs. so just a fraction of the 34 K. But it seemed like I had a lot of positive basal SNPs. Pre-R-M343. Its a little hard to say because they also use SNP names that I am not familiar with.

My LivingDNA result is only one SNP greater than FF v3, despite 3 to 4X the SNPs. Although I was impressed with my LivingDNA SNP. It's also just one SNP behind my BigY500.

I also tested with LivingDNA v1. GSA chip. On that test, I got the same SNP as 23andMe v5. I read that LDNA had 20,000 YSNPs on v1.

[Mitchell-Atkins]

It appears that this update might be testing for SNP Z12222 (U152>L2>Z49>Z142>Z12222, Z150)  

I've been tracking the Z12222 section of the FTDNA U152 project, and noted that a previously called Z150 sample was changed to Z12222 since 16 May 2024.

Z150 and Z12222 are  phylogenetic equivalents.  So I've started tracking Z12222 branch participants, to see if that number significantly increases as the update continues.

Currently there are 152 branch participants for Z12222

Hoping this is the case because Z12222 is 2 branches down/more recent than Z49, and only makes up ~16% of Z49.   Would definitely help weed out some of the family finder and false positive STR matches in the future

EDIT: Just noted that a known patrilineal match, with a TMRCA of 1838 AD, who hasn't done any y-dna testing at FTDNA (but has done family finder) is now Z49. So that answers that question. Z49 is about the best I can hope for. Oh well.

On the bright side, this legit recent patrilineal Z49 match appears as a family finder match in all 4 of the accounts I administer, so this does show that legit patrilineal matches will show up as Z49. There's not a lot of them to go through so this is a helpful improvement. Who knows? This may help me eventually break through my current 1812 AD brickwall

[ArmandoR1b]

The ISOGG reports LivingDNA  v2 with 34,216 Y SNPs.  So if quantity was the main factor, LivingDNA wins hands down.

I am not sure their selection is that great.  The download is only for your positive SNPs. so just a fraction of the 34 K.  But it seemed like I had a lot of positive basal SNPs.  Pre-R-M343.  Its a little hard to say because they also use SNP names that I am not familiar with. 

My LivingDNA result is only one SNP greater than FF v3, despite 3 to 4X the SNPs.    Although I was impressed with my LivingDNA SNP.  It's also just one SNP behind my BigY500. 

I also tested with LivingDNA v1.  GSA chip.    On that test, I got the same SNP as 23andMe v5.  I read that LDNA had 20,000 YSNPs on v1.

When I had mentioned LivingDNA, prior to the previous post, I had been referring to v1. Just one additional subclade down between v1 and v2 is expected since they have to limit the number of probed positions. That is what these arrays do. Also a lot of SNPs are just phylogenetic equivalents of subclades. So it's not as if every single SNP is a a separate subclade.

Since LivingDNA v1 and 23andme v5 were both GSA arrays and I had already seen results from numerous customers on both I knew they were very similar. They are both limited for R-DF27 but LivingDNA v1 at least reported P312 whereas 23andme v5 doesn't.

I never liked that LivingDNA didn't provide the raw data. Negatives and no-calls are important.

At least FF v3 provides Branch participants in the Haplotree but at the same time there is no way to tell how many of the kits are SNP pack tested or actually are terminal even though they had a Big Y test unless the Big Y Block tree is compared.

It would be interesting to see LivingDNA stats of the haplogroups of all of their testers. Then we could see which haplogroups are definitely tested.

R-DF27 Branch participants at FTDNA, which are almost exclusively going to be SNP pack or individual SNP tested, have gone down recently by 3 from 274 to 271 but because P312 and ZZ12_1 weren't tested by Family Finder v2 that number will never go down to 0.

[ArmandoR1b]

Does anyone know if they have processed FF kits from 2016 yet? I found two that are from early 2017. So they should be getting close to 2016 if they haven't yet. My own kit is from 2012. Any way to find out which year had the largest increase in FF customers?

[Mabrams]

Genealogical Database Growth Slows – The DNA Geek
Scroll down to the graph, and you will see FTDNA had a large increase in early 2017 and 2018

I have seen a couple of kits that were ordered in Dec 2016, but batched in January 2017.

[Mabrams]

Dec 1  245,254
May 19 488,788

That is exactly double.  

Well....if you round off ever so slightly. 


Let's go for triple!!!

[ArmandoR1b]

Genealogical Database Growth Slows – The DNA Geek
Scroll down to the graph, and you will see FTDNA had a large increase in early 2017 and 2018

I have seen a couple of kits that were ordered in Dec 2016, but batched in January 2017.

Dec 1  245,254
May 19 488,788

That is exactly double.  

Well....if you round off ever so slightly. 


Let's go for triple!!!

If FTDNA had 300,000 customers in Dec 2016, some of them might be Y-STR only or mtDNA only customers, and if only half of the remaining are men then there are less than 150,000 kits remaining. That's a lot of ifs though.

We can also look at it with 1,628,438 autosomal customers per the Autosomal Comparison Chart of the ISOGG Wiki If half of them are men and the Haplotree has 110,766 Big Y kits then there are about 436,197 kits remaining. Some of them would already have had Big Y though. Others would have had a prediction from STR testing already. So the total Haplotree count won't change as much as the remaining kits. So maximum about 300,000 kits left as far as pure Haplotree numbers. The predicted STR haplogroup number will change significantly too though.

So I think triple is possible if only looking at it from the 2nd perspective

[Pylsteen]

In my project, there are quite a number of men who tested early (>10y ago), only Y-dna, so they will not get a more refined result; however, be sure to check their Y-DNA matches, some of them may have done a FF!

[Dewsloth]

Ha, my Filipino father-in-law (who only did FF) now has a 13,000 year old Y-SNP.  Cool.

Discover:  R-FT354149's paternal line was formed when it branched off from the ancestor R-Z17112 and the rest of mankind around 1450 BCE

Old, but not as old as 13,000

I think you may have misread my post.  I'm talking about my father-in-law (wife's father):

He's O-CTS11727 and his closest Y ancient is "Dushan 4-1, a man who lived between 7024 - 6643 BCE during the East Asian Neolithic Age and was found in the region now known as Dushan, Guangxi, China.Dushan 4-1.  Shared ancestor 13,000 BCE
You and Dushan 4-1 share a common paternal line ancestor who lived around this time.
Rare Connection, 1 in 740, Only 504 customers are this closely related to Dushan 4-1."

My own father (and also me) is DF19>DF88>FGC11833>S4281>S4268>Z17112>FT354149>FT88174>Z43162>[plus four more steps (15 SNPs) into the 20th century]
and our closest ancient is a Viking buried off the coast of Sweden, "Öland 1028 [VK333], ~895 CE, Shared Ancestor 1400 BCE [FT354149]
You and Öland 1028 share a common paternal line ancestor who lived around this time.
Rare Connection, 1 in 1,600, Only 232 customers are this closely related to Öland 1028."

[Webb]

Does anyone know if they have processed FF kits from 2016 yet? I found two that are from early 2017. So they should be getting close to 2016 if they haven't yet. My own kit is from 2012. Any way to find out which year had the largest increase in FF customers?

Good eyes.  They have not gotten to 2016 yet, as the only FF matches I have who are still M269 are 2016 and older matches.  The two that were updated last week were from 2017.

[rmstevens2]

My dad, who passed away back in 2018, originally tested with the Genographic Project, whose tests were done by FTDNA. He got an R-CTS2501 result via the Genographic Project, which is the equivalent of R-DF41. That was an actual SNP test result and not a prediction based on STRs. I transferred his results over to FTDNA some years ago (probably in 2012 or thereabouts). My dad did the Family Finder test in July of 2014.

Since BY166 is downstream of CTS2501/DF41, I guess he'll eventually get moved to R-BY166 via his FF results. Has anyone seen someone in a similar situation get a Y-DNA haplogroup assignment via Family Finder that supplanted his old Genographic Y-DNA result?