Phylotree build 16 now published

[gtc]

Here is the long-awaited build 16:

http://www.phylotree.org/tree/main.htm

[Biblioteque]

gtc: You are very generous to share that with us.

[J Honeychuck]

Re: K1a C497T (T16093C)

If "Mutations between brackets () are recurrent/unstable within the respective clade, or are yet uncertain based on current data," then why do they appear in the tree at all?

TIA for your advice.

[Jessant]

Re: K1a C497T (T16093C)

If "Mutations between brackets () are recurrent/unstable within the respective clade, or are yet uncertain based on current data," then why do they appear in the tree at all?

TIA for your advice.

Transparency.

It's helpful knowing which mutations in that area of the tree have previously been shown to be unstable or recurrent.

Likewise, it's helpful to know which aspects of the tree need further investigation.

[djknox]

Understanding how to use build 16

Hi

I'm haplogroup H with the following two additional SNPs:

G9554A

T10866C


I can find G9554A on this new build as defining HV1a1a and H1e6. T10866C is not listed.

I would appreciate if anyone could explain a few things for me:

1) why does G9554A show up in two different terminal branches? Does that mean that that particular SNP independently occurred randomly twice? Or is the build incorrect?

2) Although I am H, I do not have the terminal SNPs for HV1 or HV1a or HV1a1. I also do not have the terminal SNPs for H1 or H1e. So unless my results have a couple of "no calls" that would put me into H1e6, it would appear as though I do not fit into any of the H1 thru H100 subclads. Does this mean that G9554A must have occurred a 3rd time yet to be identified other than with me?

3) Should I test for G3010A & G5460A just to make sure they were not missed?

4) I tested FMS 1.5 years ago and yet I am still not showing up in any of H1 to H100. Does this mean that no one else tested has G9554A & T10866C? I do have a GD=0 match... so would that not mean that somebody else at FTDNA also has G9554A & T10866C? Why then are we not added as H101?

5) I have 46 FMS matches all designated as H only with GD = 0 to 3. Any idea when some of those testkits will be reclassified according to this build 16?

Any insight that can be offered for the above would be appreciated.

[GST]

I can find G9554A on this new build as defining HV1a1a and H1e6. T10866C is not listed.

I would appreciate if anyone could explain a few things for me:

1) why does G9554A show up in two different terminal branches? Does that mean that that particular SNP independently occurred randomly twice?

Yes, the same mutation can occur independently in many different haplogroups or subclades. Since you do not have the mutations that define H1, G9554A occurred independently in your maternal line, and eventually it will define your subclade of H.

Should I test for G3010A & G5460A just to make sure they were not missed?

No need to retest, the test is extremely accurate and you can be confident that you are not in H1.

I tested FMS 1.5 years ago and yet I am still not showing up in any of H1 to H100. Does this mean that no one else tested has G9554A & T10866C? I do have a GD=0 match... so would that not mean that somebody else at FTDNA also has G9554A & T10866C? Why then are we not added as H101?

Mannis van Oven updates Phylotree, and he looks for at least 3 samples with unique extra mutations to define a new subclade. I think the logic is that the tree should indicates branch points, not tips of the tree. So ten people who are H, and who match each other exactly, would count as one sample for the tree. You can submit your results directly to GenBank using Ian Logan's helpful instructions. There is already one sample that shares G9554A from a research study, so you should get a subclade name when a 3rd sample is published in GenBank.

I have 46 FMS matches all designated as H only with GD = 0 to 3. Any idea when some of those test kits will be reclassified according to this build 16?

Some of these H samples are completely unique and quite rare, so it could take a long time before enough samples accumulate to create subclades for each unique branch within H. But if the rate of testing accelerates, the tree should be filled in faster. There were two recent research studies that published several thousand new FMS samples, mostly from Denmark. So that is encouraging, let's hope we see more large studies like these in the next few years.

[djknox]

OK this is clearer... thanks Gail. And thanks for posting the new build 16. I guess my impatience was premature - I see that there's a lot of effort being done... but from the outside it does sometimes seem static.

Concerning my last question, what I meant was: would not some of these 46 H haplotypes now fall into one of the new subclades now identified on the phylo tree? Ie would some of my GD=3 matches now get re-classified as H-65 for example? If yes, would it be on FTDNA's list to update haplotypes? Put another way, could some of those 46 H matches now be Hxx and now no longer of interest to me from a genealogical perspective?

Cheers
Douglas

[Ann Turner]

Concerning my last question, what I meant was: would not some of these 46 H haplotypes now fall into one of the new subclades now identified on the phylo tree? Ie would some of my GD=3 matches now get re-classified as H-65 for example? If yes, would it be on FTDNA's list to update haplotypes? Put another way, could some of those 46 H matches now be Hxx and now no longer of interest to me from a genealogical perspective?

Cheers
Douglas

Your reasoning is correct, but it doesn't appear that any new top-level branches were added in Build 16 of the phylotree -- 100 is the highest number I see at http://phylotree.org, the same as Build 15. Oh, I have a vague recollection that FTDNA was using Build 14, so maybe some of those generic H matches would disappear if FTDNA updates everyone (a big project).

One thing you could do is write to your FMS matches and ask them if they share any of your "private" mutations (after checking to be sure they have no medical implications), which could become the motif for a new subclade. Then perhaps you could ask them to submit their sequence to GenBank if they haven't done so already.

Note that they must take the initiative to do this, even if you have given FTDNA permission to use your sequence for research. FTDNA doesn't actually submit a sequence unless it's used in a published article.

Ian Logan's website has guidance on submitting sequences to GenBank

http://www.ianlogan.co.uk/Submission.htm

[Bartot]

Your reasoning is correct, but it doesn't appear that any new top-level branches were added in Build 16 of the phylotree -- 100 is the highest number I see at http://phylotree.org, the same as Build 15. Oh, I have a vague recollection that FTDNA was using Build 14, so maybe some of those generic H matches would disappear if FTDNA updates everyone (a big project).

One thing you could do is write to your FMS matches and ask them if they share any of your "private" mutations (after checking to be sure they have no medical implications), which could become the motif for a new subclade. Then perhaps you could ask them to submit their sequence to GenBank if they haven't done so already.

Note that they must take the initiative to do this, even if you have given FTDNA permission to use your sequence for research. FTDNA doesn't actually submit a sequence unless it's used in a published article.

Ian Logan's website has guidance on submitting sequences to GenBank

http://www.ianlogan.co.uk/Submission.htm

Ftdna delay in upgrading from #14 to #16 leaves Ftdna matches as basically useless.
Will they upgrade or end up being like 23andme and not change any tree since 2008?

[djknox]

Ann thanks for your insight... I get it now.

As for updating the database of customers to newer builds, it would seem to me that that is what computers are good at... following a sequential logic paradigm. I could be wrong, but it should not be a monumental task to write an update algorithm provided the database is structured properly. Having said that, I personally know how complicated code can become if its foundations are not well done...

If I were FTDNA, I would start building internal admin tools to handle these sort of things on a semi-automatic basis. It's an investment that will pay big dividends as the industry expands. Otherwise, their client base will collapse out of frustration and bad reputation.

As Bartot eludes to, there's not much purpose in providing genetic dna testing without providing analysis output and tools. I think I have been quite vocal about this in years past and it is where the commercial industry is failing to date. FTDNA needs some big-time IT leadership with appropriate funding.

[Rebekah Canada]

Updating to Phylotree 16

Hi,

It is not big IT/EG wise from our side. We upload a new tree file, start a stored procedure, and push it live when done. We are in hover to go up to a new tree alongside NatGeo as they use data we process and send.

Your reasoning is correct, but it doesn't appear that any new top-level branches were added in Build 16 of the phylotree -- 100 is the highest number I see at http://phylotree.org, the same as Build 15. Oh, I have a vague recollection that FTDNA was using Build 14, so maybe some of those generic H matches would disappear if FTDNA updates everyone (a big project)...

[Littlest bit]

Rebekah Canada said:

It is not big IT/EG wise from our side. We upload a new tree file, start a stored procedure, and push it live when done. We are in hover to go up to a new tree alongside NatGeo as they use data we process and send.

I don't know. I got my FMS results last year and I am listed as J1c3 but I have the mutations for J1c3i which was added to the Phylotree in Sept 2012, build 15. My J1c3i mutations are listed as "extra mutations." If it is so easy to update, why didn't they update to build 15 or is my J1c3i just not added?
http://PhyloTree.org/what_is_new.htm

From my results:
Extra Mutations 309.1C 315.1C 522.1A 522.2C T5442C C12858T A15758G C16519T

Best mtDNA Haplogroup Matches:

1) J1c3i

Defining Markers for haplogroup J1c3i:
HVR2: 73G (185A) (228A) 263G 295T 462T 489C
CR: 750G 1438G 2706G 3010A 4216C 4769G 5442C 7028T 8860G 10398G 11251G 11719A 12612G 13708A 13934T 14766T 14798C 15326G 15452A 15758G
HVR1: 16069T 16126C

Marker path from rCRS to haplogroup J1c3i (plus extra markers):
H2a2a1(rCRS) ⇨ 263G ⇨ H2a2a ⇨ 8860G 15326G ⇨ H2a2 ⇨ 750G ⇨ H2a ⇨ 4769G ⇨ H2 ⇨ 1438G ⇨ H ⇨ 2706G 7028T ⇨ HV ⇨ 14766T ⇨ R0 ⇨ 73G 11719A ⇨ R ⇨ 4216C ⇨ R2'JT ⇨ 11251G 15452A 16126C ⇨ JT ⇨ 295T 489C 10398G 12612G 13708A 16069T ⇨ J ⇨ 462T 3010A ⇨ J1 ⇨ (185A) (228A) 14798C ⇨ J1c ⇨ 13934T ⇨ J1c3 ⇨ 5442C 15758G ⇨ J1c3i ⇨ (309.1C) (315.1C) 12858T

Good Match! Your results also had extra markers for this haplogroup:
Matches(28): 73G (228A) 263G 295T 462T 489C 750G 1438G 2706G 3010A 4216C 4769G 5442C 7028T 8860G 10398G 11251G 11719A 12612G 13708A 13934T 14766T 14798C 15326G 15452A 15758G 16069T 16126C
Mismatches(0): (185G)
Extras(1): (309.1C) (315.1C) 12858T

[gtc]

FYI

On molgen.org James Lick announced on 21 Feb:

"PhyloTree Build 16 was released yesterday. Now the beta version of mthap has been updated with this new build.

http://dna.jameslick.com/mthap-new/

There are about 1,265 haplogroup changes in the new build."

Note that James doesn't update his revision history until mthap is out of beta -- hence no mention of build 16 on that page as yet.

[Bartot]

My Wife tested with Ftdna and got K1a4.
I tested her fasta with JLick program and it states K1a4 using tree#16.

She does not have any mutations that match K1a4"a,b,c,d,e,f,g or h.............yet she has other mutations. Question is, how can I check her other mutations origins?
other extras are 3970T, 12273G and 16304C

[Ann Turner]

My Wife tested with Ftdna and got K1a4.
I tested her fasta with JLick program and it states K1a4 using tree#16.

She does not have any mutations that match K1a4"a,b,c,d,e,f,g or h.............yet she has other mutations. Question is, how can I check her other mutations origins?
other extras are 3970T, 12273G and 16304C

First, just a general word about posting coding region mutations. Some are associated with various medical disorders (your wife's are not). A good preliminary place to check is

http://mitomap.org

You can look up her haplogroup at

http://www.mtdnacommunity.org/human-...phylogeny.aspx

Hint: K is a haplogroup under U8b. Start expanding the U section (at the very bottom of the tree).

Then click on the K1a4 hyperlink to view the records in their database and see if any of them share her "private" mutations.