PDA

View Full Version : Changes in the FTDNA Tree



Ravai
01-13-2017, 10:22 AM
Good morning, changes are occurring in the FTDNA tree. Now I am:

U152> L2> BY3485> BY14171> BY3478> BY14173

Has your subclass changed as well?

13557

Regards

angscoire
01-13-2017, 11:29 AM
Not yet . Still down as R-YP326 . YFull have recognised my downstream snip of YP5543 for many months now , FTDNA still do not. These things take time.

RobertCasey
01-13-2017, 03:10 PM
I know that I am FGC5647 and its son FGC5639. However, since FGC5639 was discovered as branch via testing at YSEQ, FTDNA refuses to add my branch FGC5639 and lists it as an equivalent of FGC5647. They will not add branches discovered via Full Genomes or YSEQ which is not very responsive to genetic community and is very predatory towards other vendors. I was also forced to test these branches at YSEQ since FTDNA refused to add these YSNPs for individual YSNP testing. We did get both of these YSNPs into our L226 SNP pack, so the only way to get this branch on the FTDNA haplotree is to have one FGC5647 POS and FGC5639 NEG person to place an order for the L226 SNP (no new information - just a redundant test) or order another Big Y which could also reveal more private YSNPs to test at YSEQ (our only choice). Since FTDNA still refuses to add any new YSNPs that are not in their SNP packs, we go around in circles on getting their haplotree corrected.

To date, the cost of every new branch discovered by Big Y tests is around $800. The cost per branch discovered via testing private YSNPs is around $80 for the first seven branches discovered by this method. Since Dennis Wright got 50 private YSNPs included in the L226 SNP pack, we have also revealed another six branches with SNP pack tests of these private YSNPs. However, the 50 equivalent YSNPs included in the L226 SNP pack have not revealed any new branches. So, load up your SNP pack up with private YSNPs and you will build your haplotree much faster at lower costs than just NGS tests.

Osiris
01-13-2017, 08:30 PM
I'm right above you (BY4242).

They updated me pretty fast once the Big Y results came in for my little clade. BY4242 was born between 1760 and 1972 so I guess I'm pretty much done.

TigerMW
01-17-2017, 06:00 PM
.... FTDNA refuses to add my branch FGC5639 and lists it as an equivalent of FGC5647. They will not add branches discovered via Full Genomes or YSEQ which is not very responsive to genetic community and is very predatory towards other vendors.
Very predatory? Oh my, hundreds of vendors including many favorites like Apple are 10 times more predatory.

Do you understand that what you think is predatory may be just being cautious and following good science from another perspective? If you were FTDNA and stood behind your tests (like they do) and a new branch is added based on outside testing and it turns out to lead a customer down the wrong path of future SNP testing then you as FTDNA would probably decide you'd rather be rely of your lab standards and chain of evidence for branch proofs.

I'm not a fool. It's certainly a lot easier to decide something if it makes more money too but this is hardly in the realms of dastardly.


.I was also forced to test these branches at YSEQ since FTDNA refused to add these YSNPs for individual YSNP testing.
The economics of individual SNP testing via Sanger Sequencing have seen their better day. You are lucky to be in a populous and youthful haplogroup.

I was able to add many SNPs for L513 to FTDNA's Sanger Sequencing without being in SNP Packs. They were adding SNPs at a pretty good clip for much of the first year of Big Y results. To that effect, the early bird gets the worm. FTDNA did go on a long hiatus that is that is now over. I just received another set of emails today on new Sanger Sequenced SNPs available that I submitted in December.


We did get both of these YSNPs into our L226 SNP pack, so the only way to get this branch on the FTDNA haplotree is to have one FGC5647 POS and FGC5639 NEG person to place an order for the L226 SNP (no new information - just a redundant test) or order another Big Y which could also reveal more private YSNPs to test at YSEQ (our only choice). Since FTDNA still refuses to add any new YSNPs that are not in their SNP packs, we go around in circles on getting their haplotree corrected.
It takes more time to work through the SNP Pack refresh and testing routing to get an outside discovered branch on to FTDNA's haplotree. Part of this is just the cost of multi-vendor integration. There are advantages, but there are trade-offs too. Everyone has to make these kinds of decisions.

To date, the cost of every new branch discovered by Big Y tests is around $800. The cost per branch discovered via testing private YSNPs is around $80 for the first seven branches discovered by this method.
I don't think your scenario applies to many subclades. L226 is among a handful of unique subclades. Still, I have hard time pushing SNP chasing unless I'm paying for someone else testing my SNPs. I think NGS testing in general still has a huge advantage of discovering your own line of SNPs.


Since Dennis Wright got 50 private YSNPs included in the L226 SNP pack, we have also revealed another six branches with SNP pack tests of these private YSNPs. However, the 50 equivalent YSNPs included in the L226 SNP pack have not revealed any new branches. So, load up your SNP pack up with private YSNPs and you will build your haplotree much faster at lower costs than just NGS tests.
We are still having good luck in the L513 group with packs breaking up blocks, but it is the exception not the rule. My thinking is to have one SNP Pack cycle include a bunch of the phylogenetic equivalents to give them a good run through and then in the next cycle (which we are on now about to begin in L513) switch over to the youthful SNPs.

RobertCasey
01-17-2017, 10:44 PM
Very predatory? Oh my, hundreds of vendors including many favorites like Apple are 10 times more predatory.

Do you understand that what you think is predatory may be just being cautious and following good science from another perspective? If you were FTDNA and stood behind your tests (like they do) and a new branch is added based on outside testing and it turns out to lead a customer down the wrong path of future SNP testing then you as FTDNA would probably decide you'd rather be rely of your lab standards and chain of evidence for branch proofs.

I'm not a fool. It's certainly a lot easier to decide something if it makes more money too but this is hardly in the realms of dastardly.

I still believe that FTDNA is predatory to other YDNA vendors. This a volunteer driven business where we supply countless labor in order for FTDNA to keep their costs low as possible. It remains predatory when requests are denied based on using genetic data from other vendors. FTDNA readily accepts our conclusions supported by FTDNA genetic information - but rejects the same kind of information when coming from other sources. Also, your FTDNA concentric advice will slow down innovation. FTDNA offered "Walk the Y" way beyond its useful life. Only YElite resulted in Big Y. YSEQ introduced tiered testing via Sanger Sequencing in their SNP packs. Then we got FTDNA SNP packs. WGS is now being offered by FGC and YSEQ at reasonable prices - but only one version of Big Y to date. 23andMe introduced atDNA testing - then we get Family Finder. I am hard pressed to think of any product that FTDNA has rolled out first. The Mass Array is a fresh technology, so that is the only bright spot recently.


The economics of individual SNP testing via Sanger Sequencing have seen their better day. You are lucky to be in a populous and youthful haplogroup.

I am still adding new branches to L226 at 10 % of the cost of yet more Big Ys. Big Y is $800 per branch while YSEQ is $80 per branch discovered. At least 80 to 90 % of the haplogroup R has predictable YSNP branches in the 1,500 to 2,500 range with enough branches to warrant individual YSNP testing. Why are requesting YSNPs to be added if its day has gone by ?


I was able to add many SNPs for L513 to FTDNA's Sanger Sequencing without being in SNP Packs. They were adding SNPs at a pretty good clip for much of the first year of Big Y results. To that effect, the early bird gets the worm. FTDNA did go on a long hiatus that is that is now over. I just received another set of emails today on new Sanger Sequenced SNPs available that I submitted in December.

I too was able to get some of my early FGC YSNPs added, then after Big Y had been out a while, these requests are now denied if they are not FTDNA discovered. Again - predatory behavior. Also, it is very clear that those individuals who promote FTDNA concentric recommendations get preferential treatment.


It takes more time to work through the SNP Pack refresh and testing routing to get an outside discovered branch on to FTDNA's haplotree. Part of this is just the cost of multi-vendor integration. There are advantages, but there are trade-offs too. Everyone has to make these kinds of decisions.

This is a valid point. But I really grow weary of those that downplay the need for competition which brings innovation. FGC skirts the database issue as well, so the world is not perfect on the vendor side as well.


I don't think your scenario applies to many subclades. L226 is among a handful of unique subclades. Still, I have hard time pushing SNP chasing unless I'm paying for someone else testing my SNPs. I think NGS testing in general still has a huge advantage of discovering your own line of SNPs.

NGS testing is always the preferred approach as it is the only means to discover more private YSNPs. But it is just not cost effective - and a lot of individuals who can not afford the NGS cost can make significant contributions for only $100 of YSEQ testing. I am extremely pleased with the L226 SNP Pack though. It is really exceeded my expectations. We are at around 50 L226 SNP packs now with eight new branches from the 50 private YSNPs included. But the cost per branch remains around $700 per branch. But like Big Y, the real value of SNP packs is the ability to chart L226 that is only 25 % thoroughly YSNP tested. The charting (via signature prediction) now charts 75 % of the 500 L226 67 marker submissions. Pretty exciting times compared to the days of only having L226 being tested which remains at 100 % accuracy via signature prediction.


We are still having good luck in the L513 group with packs breaking up blocks, but it is the exception not the rule. My thinking is to have one SNP Pack cycle include a bunch of the phylogenetic equivalents to give them a good run through and then in the next cycle (which we are on now about to begin in L513) switch over to the youthful SNPs.

Didn't you and I learn our lesson with wasteful testing of L21 equivalents, L459 and Z290. We could have tested around ten more Walk the Ys for the same cost. With 50 NGS tests and 50 L226 SNP packs, we now have 100 tests of 50 equivalents with not a single new branch discovered via testing of equivalents. This includes around 20 of my FGC L226 equivalents. I have discovered around 10 more L226 equivalents that have only been tested twice. These are now either new L226 equivalents or equivalents of my recent branches, FGC5647 and FGC5639. They were very kind to add these two YSNPs and DC69 which is the other Casey branch under L226. DC69 was discovered by Big Y. Requests to add the FGC L226 equivalents that have only three tests were denied. Instead, they included 20 L226 FGC equivalents that have remained L226 equivalnets after 50 NGS tests.

I highly recommend encouraging loading up any future version of your SNP packs with as many private YSNPs as possible. Also, these should not be spread out prorated based on Big Y tests. Out of 42 branches, 35 % private YSNPs of NGS testers fall into only four branches. So, you really need to analyze which branches are loaded with private YSNPs and heavily test these bottleneck branches. Also, beware that FTDNA has listed all 50 private YSNPs as branches for both L226 and L555. This really creates a lot of confusion of testers and is pretty misleading to show these as branches. However, we really need to create some way to add private YSNPs to the FTDNA haplotree that are being tested in SNP packs. Another good reason to load up on private YSNPs, they add these private YSNPs to the list to be tested via Sanger sequencing as well. So, the eight new branches can now be tested individually as well.

However, I am really growing weary of defending my point view. It is always best to have a good dialogue of different viewpoints. I miss those days when you shared that viewpoint. It is sad to see the polarization of FTDNA concentic viewpoints vs. 3rd party viewpoints.

TigerMW
01-18-2017, 06:53 PM
I still believe that FTDNA is predatory to other YDNA vendors. .... . FTDNA readily accepts our conclusions supported by FTDNA genetic information - but rejects the same kind of information when coming from other sources.
This is not really true in terms of haplotree and SNP test development. FTDNA seeks customer advice but they have validation steps they must follow before their haplotree can be updated or they make SNPs available.

You just don't like their validation steps since they require in-house testing on the Y DNA tree side. You have a reasonable disagreement with them on this. I agree with you that they should accept YSEQ results into their system, but I think it should be in a formal way, updating their database at the individual level. I'm not sure it would be as helpful as we want as they'd probably charge a transfer price for this.


Also, your FTDNA concentric advice will slow down innovation.
I am not sure why you had to throw in this general criticism of me. I understand if you disagree with some of my points but let's discuss the points rather than assign general negative commentary.


... Why are requesting YSNPs to be added if its day has gone by ?
I didn't say the days of Sanger Sequencing are gone. I said "The economics of individual SNP testing via Sanger Sequencing have seen their better day. You are lucky to be in a populous and youthful haplogroup."
The point I had was not that they were gone for L226, but that L226 is a lucky haplogroup in terms of its size, youth, depth of testing and people like yourself and Dennis Wright, who've driven so much testing and analysis. One at a time SNP chasing can work fine in that environment with your leadership and planning.

One at a time SNP chasing does not work fine in less populous, poorly tested subclades where coordination is not good. When I say it does not work fine the point I am getting to is that it does not work fine for the vendors and then consequently for the testers. SNP chasing can be frustrating and in reality you are still really testing for other peoples' SNPs.

Sanger Sequencing panels are still one at time SNP testing but the below gives you an idea of what the vendor faces in turns of economic viability. This is what I meant by Sanger Sequencing having seen it's better day. It's not cost-effective for broad usage. That's what you'd expect for a technology that was developed in the 1970's.

Here is the Sanger Sequencing R1a-Y9073 Mortzell panel. 14 SNPs for $194.00
http://www.yseq.net/product_info.php?products_id=16824

Here is the Sanger Sequencing R1b-BY2606 panel. 38 SNPs for $395.00
http://www.yseq.net/product_info.php?products_id=11897

Here is the Sanger Sequencing R1b-L1065 McAfee - McDuffee. 5 SNPs & 6 STRs (a nice approach) for $179.00
http://www.yseq.net/product_info.php?products_id=5148

I hit the reload button on my browser so these are current as of right now.

The issue of economic viability and technical feasibility is on the vendor side. That is the point I was trying to get to. I created this chart in late 2013 when I recognized the issue. It's unavoidable, particularly for Sanger Sequencing.
https://dl.dropboxusercontent.com/u/17907527/Hypothetical_Economic_Challenge_of_Individual_SNP_ Testing.pdf

The issue affects SNP Packs as well, but the whole purpose of Mass Spectrometry (Agena MassARRAY) is to provide better cost-effectiveness so SNP Packs can go a little further than one at a time SNP testing.


Also, it is very clear that those individuals who promote FTDNA concentric recommendations get preferential treatment.
I think you are overreacting in this commentary.

Is it not a good thing if an individual provides advice to vendor that helps them build better products that satisfy more people? It is beneficial for both testers and vendors.

If a person gives advice that turns out to be good, is not that just building credibility? If you were a vendor, would you not be more likely to take advice from a credible source?

Sometimes in the process you learn some things. Some times it is from other volunteers (i.e. Rory Cain on request driven updates to FTDNA's haplotree). Some times it is from the vendor (ie. SNP pack development proactively updates FTDNA's haplotree.) I try to pass that on unless it is breaking a confidence.



NGS testing is always the preferred approach as it is the only means to discover more private YSNPs. But it is just not cost effective.
I agree with your first sentence but the second sentence depends. Everyone's budget and goals are different. It's up to them what is affordable or not, but a course of action is NOT cost-effective it you don't reach goals sufficiently. You can't get your own line of SNPs (or can't know) by one at a time SNP chasing until we get down to the high confidence known genelogical records scenario.

If you want your own line of SNPs, one at a time SNP testing is ineffective. Something that is ineffective is not cost-effective even if free. When it comes to genetic genealogy, your most valuable SNPs are the last couple.


Didn't you and I learn our lesson with wasteful testing of L21 equivalents, L459 and Z290.
I agree. I think lightly tested phylogenetic equivalent blocks need to be hit hard but there is some point when it is wasteful. The one thing to keep in mind is that with some technologies, the SNP Pack MassARRAY for example, there is no discernible cost difference for adding one more SNP to a pack product. The only concern is the capacity of the product, hence there is almost no waste unless at capacity.


I highly recommend encouraging loading up any future version of your SNP packs with as many private YSNPs as possible.
I agree but there has to be some analysis of the nature of the private SNPs. This is where haplogroups like L226 are lucky and have relatively few private SNPs in comparison to the youthfulness and testing population size of the haplogroup.

On the other hand we have people with a list of 20 or even 30-40 private SNPs and don't have many folks with close genetic distances. You could load a whole pack with just four or five sets of private SNPs. That's not economically viable for the vendor. It also causes discouragement for the testers if they come up negative on everything or negative on everything but a 4000 year old SNP.


However, I am really growing weary of defending my point view. It is always best to have a good dialogue of different viewpoints. I miss those days when you shared that viewpoint. It is sad to see the polarization of FTDNA concentic viewpoints vs. 3rd party viewpoints.
You are the one that goes on the attack, calling FTDNA "predatory". Perhaps you may not want to attack others if you grow wary of defending your attack. An argument does not need to be personal or inflammatory. It can be just good discussion. If you want to turn to using negative adjectives then you are more likely to see counter-arguments. In any case, agreement is not required for learning. In fact disagreement is better for challenging thought.