Haplogroup H

Comment

Comment by Mardon Erbland on May 19, 2010 at 3:35am

Thanks for the information Peiternella. I'm also somewhat unusual on the Y side where I'm Q1a3*.

At FTDNA my mtDNA matches page shows 2 low resolution matches on HVR1, 2 low resolution matches on HVR2 but no high resolution HVR1+ HVR2 matches and no FGS (full genome sequence) matches.

At FTDNA I'm a member of the project group, "H Subclade Discovery". Their goal is to "Define new subclades of H." In order to be accepted for membership into this project, you must have taken an mtDNA FGS test. As I write this, my kit #38126 is one of the 103 kits listed in the "Unassigned Members" cluster. (http://www.familytreedna.com/public/mtdna_hstar/default.aspx?sectio...

Comment by Peiternella Suzzanna Hymen on May 18, 2010 at 10:08pm

Hi Mardon, I have completed the database with all the sequences available at Genbank
Even when I compare you with other sequences to look for patterns, and allowing for a possible back mutation of a defining motif, there are none as yet.
Comparing your motifs,
G16438A,
A16212G, to others in the database, shows that you are unique and that maybe at FTDNA ( they must have heaps of full sequences that are not at Genbank) A8638G shows only 2 on my database you and an H30 but you have nothing else in common with that H30.
.
G16438A are on MitoWheel it says that there are 4 out 3735 on their database and A16212G 13 out of 3735.
Maybe you are another sub branch of H, using A16212G G16438A

Comment by Peiternella Suzzanna Hymen on May 10, 2010 at 8:21pm

Mardon here is an example of an H2a2b1 missing H2 branch mutation 1438A
HM107111 FTDNA Haplogroup H2a2b1 26-APR-2010
A263G 309.1C 315.1C C3388A A8860G A15326G A16235G C16291T A16293G C16400T

Comment by Peiternella Suzzanna Hymen on May 10, 2010 at 7:51pm

cheers Maria and yes I realise about papers can take time. However my partner talks about 2004 being the last paper he has seen on his particular Haplogroup. I look forward to the new paper.

Comment by Maria on May 10, 2010 at 7:35pm

Not to disappoint you, but papers (particularly good papers) take a long time. First you do the actual research, then you write, someone reviews it, then its sent to the journal where it is peer reviewed, it usually comes back with comments, you rewrite it with the needed changes, you send it back and with luck it gets published. If it's not accepted, you start all over again.
Having said that, there should be something more than what we have for the H group. Last week I found an interesting and recent paper on this group. I'll forward the link tomorrow as I have it at work.

Comment by Peiternella Suzzanna Hymen on May 10, 2010 at 7:23pm

Also Mardon, My partner foams at the mouth over lack of up to date papers being written and how the public, who are paying for the sequencing to be done so they can follow genealogy through Phylogenetic study, are being left in the dark and the academics are using this data but not hurry to publish or actually give some data that is actually relevant to genealogy study ,as such.
I trust his judgement on reading sequences more than Phylotree and 'our dragging in the sand' FTDNA. We do not get much info or what is really happening in our projects He is in a private Haplogroup project where they share photos, family news, personal interests and they always seem to be online discussing dna or just random stuff -just like a club -his group at FTDNA, where he is co admin there, are not very active but not as pitiful as our H group.!!!
Hope I have made sense. in both emails. I am home bored and lack focus LOL got my leg in plaster and I am not good at sitting idle. I was so trained for a marathon .Was feeling so fit and ready.... sob sob.
If you like I can email you a sample H3 data base he made for me just so I could look at what data there was around. It is old now and not to current.
excuse spelling and grammer please LOL. Proof reading I should do obviously!!!

Comment by Peiternella Suzzanna Hymen on May 10, 2010 at 7:03pm

AHHH crap!, I put a 1 on the end of H6a1a. Actually this sequence is at genbank as a H6a1. This is quite new as I saw it on Ian Logans reports to Genealogy DNA newsletter
HM124476 FTDNA Haplogroup H6a1 26-APR-2010
T195C T239C A263G 315.1C A750G A1438G G3915A A4727G A4769G A8860G
G9380A T10237C T11204C T11253C A15326G T16362C A16482G.
Now my partner treats the back mutations as:
if they have all the mutations that are in a particular Haplogroup sub branch but are missing that sub branch defining code
ie I have an H3 that shares nearly all the mutations with Hf but is missing 93G and they also share a unique private mutation, so he has classed it among the HF but still named it H3 as he says the phylotree changes as new data comes in. So far I have not come across a major sub branch mutation back-mutation but my partner has had a couple and he showed me their position on his data base. He said you can see the pattern of the mutations and that they are from that particular branch. When they are laid out you see patterns. Our H mutations on Phylotree are a mess-would you not agree?
This was a FTDNA sequence not 23andme.

Comment by Mardon Erbland on May 10, 2010 at 5:26pm

Hi Peiternella,

I'd like to understand better why you've put that haplotype into H6a1a1 instead of stopping at H6a1a*. Can you explain that part a little more?

For mtDNA haplotypes that have an upstream SNP missing when it 'should' be present according to the known downstream SNPs, I'd think first about the possibility of a data error.

I know from personal experience that SNP calls produced by genechip technology are subject to error rates large enough that they cannot be ignored. My mtDNA calls generated from the Illumina HumanHap550+ beadchip that 23andME uses contained several errors when compared to the direct sequencing done by FTDNA for my mtFullSequence test. If you are including data in your database that comes from genechip calls, then I think you can expect some incorrect calls. As I understand it, mtDNA is especially problematic for genechips.

Another possibility is that a human has made a transcription error somewhere in the path between the laboratory and the results that eventually get sent to you.

Your point about back mutations is an interesting one. I don't know how back mutations are supposed to be treated within the mtDNA Haplogroup Tree. I would suppose if a child of a person who was H6a1a developed a back mutation for rCRS position 3915, thereby putting position 3915 back to G from A, then the tree ought to show such a person in a new subclade of H6a1a and not in a new subclade of H6. That's just what seems logical to me. I don't have any idea if there are 'rules' as how back mutation are to be handled in the tree. Do you know?

Mardon

Comment by Peiternella Suzzanna Hymen on May 10, 2010 at 4:34pm

I got that too.But it is not submitted to Genbank as that !!!!. I put that sequence into my data base as H6a1a1 but not because it did not have 5460- they don't have 7325G, 9362 or at least one of them. I see a few sequences that do not have a all or even a branch mutation but can have all the clade mutations- you see a pattern when they are all in a group so it sticks out like a 'horses bottom' when they are or not. I have allowed for back mutation as my partner (has been guiding me) likes to point out that back mutations are not to uncommon and he has also found many wrong branch defining in his W group. Pete has a science degree in Geology but also did a year in Univesity intermediate biology so has a fantastic grasp of the subject. I feel so dumb sometime or actually most the time when he goes on about the maths side of DNA and much much moreLOl

Comment by Mardon Erbland on May 10, 2010 at 6:47am

Is this a 'trick' question? :)

Here's how I'd figure it (without even assuming that the H6 you've provided is correct):

Start at rCRS
A263G, A8860G, A15326G ---> H2a2
A750G ---> H2a
A4769G ---> H2
A1438G ---> H
T16362C ---> H6/H8
T239C, A16482G ---> H6
G3915A ---> H6a
A4727G, G9380A ---> H6a1
T11253C ---> H6a1a

This leaves SNPs T10237C and T11204C unaccounted for.

H6a1a1 requires an SNP at rCRS 5460 so this haplotype is NOT that subclade.

I therefore conclude that the haplotype you've provided is for a yet-to-be-defined subclade of H6a1a. Am I close of have I missed something?

• ‹ Previous
• 1
• 2
• 3
• 4
• 5
• …
• 8
• Next ›
• Page