GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2010-02 > 1267083627


From: Thomas Krahn <>
Subject: Re: [DNA] Hello Mr. Krahn
Date: Thu, 25 Feb 2010 01:40:27 -0600
References: <9a804091002231904y6e27ed67g287b8f27361b1274@mail.gmail.com><4B84BDFA.40407@familytreedna.com> <c6384e011002241610h5cd82278g4f748c7c2f7b8c31@mail.gmail.com><c6384e011002241628w7cc2f292l594124ff14cd4e3d@mail.gmail.com><9a804091002242207w5eff7199q26389595eceb8368@mail.gmail.com>
In-Reply-To: <9a804091002242207w5eff7199q26389595eceb8368@mail.gmail.com>


ab10man2 wrote:
> You said : "L222.2 is not a back mutation. It is a parallel mutation."
>
> What does it mean to say that a mutation is parallel ?
>
Most people from all haplogroups have at the L222 position a T allele.
There are (as far as we know today) two exceptions:

In haplogroup E-M2 the T allele has mutated to G. (=L222.1)
In haplogroup J-L147 the T allele has mutated to G. (=L222.2)

Those mutations are independent phylogenetic events.
However they have both the same location and the same derived allele.
Therefore we consider them as parallel mutations.

>
>
>
> By the way, are you comfortable to say that l222.2 is a usable mutation that
> can help J1 divide their subclade to smaller clusters ? I mean is it like
> l147 or more changeable than that ?
The following is a link to the L222 location on my Y chomosome browser:
http://ymap.ftdna.com/cgi-bin/gbrowse/hs_chrY/?start=15318864;stop=15318910;ref=ChrY;width=1024;version=100;grid=on;id=16a5bc02c5b6419841172f4b65599de2;label=CYT%3Aoverview-Ideogram-Palindromes-DNA%2FGC%20Content-Genes-STRs-dbSNP-JW-KOREF-NA07022-NA18507-HelicosP0-CV-YH-M-PS-PerlegenPrimerPairs-other;h_feat=L222%40yellow

Note the actual DNA sequence in red:
gagagagatagatagatagatagatagatagatagatagatagatag

The L222 mutation looks like this:
gagagagaGagatagatagatagatagatagatagatagatagatag
(note the capital G which was a t before)

If you look at the pattern of the DNA sequence you will see that the SNP
is right at the transition between a dinucleotide repeat and a
tetranucleotide repeat:

ga|ga|ga|ga|Ga|gata|gata|gata|gata|gata|gata|gata|gata|gata|g

Also in the original (ancestral) form of the sequence we have a similar
STR pattern:
ga|ga|ga|gaTa|gata|gata|gata|gata|gata|gata|gata|gata|gata|g

So by the formal L222 T to G mutation we have effectively increased the
(ga)3 repeat by two units to (ga)5
and we have reduced the size of the (gata)10 repeat from 10 to 9 units.

This type of mutation has been recently observed in several other
independent locations.
Other examples of this type are L69/L159 or L149 or L58. From some of
those markers we know
that they are very instable. Apparently similar mechanisms like in a STR
mutation happen and
overwrite single base mutations in the forward or in the reverse direction.

L222.2 has so far been a very reliable SNP in J1 and I have no
indication that it isn't reliable for genealogy and
anthropology. So there is no reason to not use it for your purpose.
However you should definitely be aware that
this marker POTENTIALLY can mutate back some time in the future because
of this STR like mutation mechanism.
So I encourage you to use it, but when you observe some unusual case in
the future that doesn't match into the
genealogical scheme you should remember this and better verify the
phylogeny with independent markers.

It is generally a good idea to not rely on just a single marker. It is
more efficient to verify the consistency of STRs, at least 3 SNPs and
the genealogical background information before to decide to which
(terminal) haplogroup a person belongs.

I hope this helps,

Thomas


This thread: