GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-04 > 1175553032


From:
Subject: Re: [DNA] DYF399 diversity (J2a1*)
Date: Mon, 02 Apr 2007 23:30:32 +0100


Al

I wish I could draw a diagram and post it to the list!
Instead I will try to show how the order of the markers might be changed by
recombination.

The first thing to remember is that the diagram is based on the HUGO
reference sequence of the Y chromosome which is almost entirely based on the
DNA of one anonymous donor. All that is known is that he was from Buffalo,
New York. Perhaps unsurprisingly his haplogroup was R1b1c. Research papers
seem to agree that the overall structure is common to most haplogroups but
we know that there are variations sometimes at individual bases (e.g. the
DYS464 c/g types) and sometimes in the form of duplications and deletions.
Inversions (i.e. where a section between two mirror-image points is
reversed) are probably common but hard to detect with PCR-based tests, so
the actual positions of the C- and T-types of DYF399 may vary between
individuals.

Also the diagram doesn't show the *direction* of the multicopy segments,
which is important. Inversions and conversions can occur when sections are
mirror images of each other (so can form hairpins - see pages 7 to 14 of
Thomas Krahn's presentation), duplications and deletions when they are
facing the same way (see page 40).

Taking the diagram as a starting point we have in P3 - P2 - P1 (section name
- STRs included)
Blue 1 - DYF397
Blue 2 - DYF397
Green 1 - DYF399 (ins G, T-type)
Red 1 - DYS464 & DYS725
Red 2 - DYS725 & DYS464
Grey 1 - NN = new as yet unnamed dinucleotide STR
Blue 3 - DYF397
Yellow 1 - DYF401 to DYF408
Green 2 - DYF399 (T-type)
Red 3 - DYS464 & DYS725
Red 4 - DYS725 & DYS464
Green 3 - DYF399 (C-type)
Yellow 2 - DYF408 to DYF401
Blue 4 - DYF397
Grey 2 - NN

Blue 2 and Blue 3 are mirror-image sections so can line up and the segment
in between flip over to give:

Blue 1 - DYF397
*Blue 3 - DYF397
*Grey 1 - NN = new as yet unnamed dinucleotide STR
*Red 2 - DYS464 & DYS725
*Red 1 - DYS725 & DYS464
*Green 1 - DYF399 (ins G, T-type)
*Blue 2 - DYF397
Yellow 1 - DYF401 to DYF408
Green 2 - DYF399 (T-type)
Red 3 - DYS464 & DYS725
Red 4 - DYS725 & DYS464
Green 3 - DYF399 (C-type)
Yellow 2 - DYF408 to DYF401
Blue 4 - DYF397
Grey 2 - NN

Now Green 1 and Green 2 are mirror-image sections so can line up in a
hairpin, so the Green 1 "ins G, T-type" copy could be replaced by a
duplicate of the Green 2 "T-type" copy.

Finally Red 2 - Red 1 and Red 3 - Red 4 are facing the same way so a
duplication can occur, giving

Blue 1 - DYF397
Blue 3 - DYF397
Grey 1 - NN
Red 2 - DYS464 & DYS725
Red 1 - DYS725 & DYS464
Green 2#2 - DYF399 (T-type)
Blue 2 - DYF397
Yellow 1 - DYF401 to DYF408
Green 2 - DYF399 (T-type)
Red 2#2 - DYS464 & DYS725
Red 1#2 - DYS725 & DYS464
Green 2#3 - DYF399 (T-type)
Blue 2#2 - DYF397
Yellow 1#2 - DYF401 to DYF408
Green 2#4 - DYF399 (T-type)
Red 3 - DYS464 & DYS725
Red 4 - DYS725 & DYS464
Green 3 - DYF399 (C-type)
Yellow 2 - DYF408 to DYF401
Blue 4 - DYF397
Grey 2 - NN
Altogether, there are now 3 yellow sections, 5 blue sections, 5 green
sections and 6 red sections, just as indicated by your results.

How to verify this? I found NN in the reference sequence after discussions
last summer between myself, Bonnie and Thomas about a quite different sort
of duplication in another branch of J2. In the Repping et al 2006 paper
which discussed structural variations the sample from the same subgroup of
J2 was shown to have very clearly 3 "grey" palindromic segments. There
appear to be relatively few STRs in this section but "NN" looks promising.
In my suggested structure there are only *two* grey sections so you should
have only two copies of NN despite all the extra copies of the other
palindromic markers.

On checking some more papers I have just noticed that my sequence of events
parallels a proposed sequence for the deletion which appears to be universal
in haplogroup N i.e. a Blue 2 - Blue 3 inversion followed by a Red/Green -
Red/Green *deletion* (rather than a duplication). Haplogroup N individuals
have lost precicely as many palindromic marker copies as you have gained.
They should still have both of their NN copies, just as you should.
Hmm - another paper gives a different explanation for the hg N deletion
which may also generate a parallel explanation for your duplication, with a
quite different resulting structure but exactly the same number of copies of
each marker. I'll have to write again tomorrow... :-)

Gareth



----- Original Message -----
From: "Alfred A. Aburto Jr." <>
To: <>
Sent: Monday, April 02, 2007 4:02 PM
Subject: Re: [DNA] DYF399 diversity (J2a1*)


> Gareth,
> Thank you so much for jumping in here. I don't think I can digest all
> that you said very quickly. I responded to Bonnie & Rebekah before I saw
> your post. From my post to them you can probably see the great gulf in
> knowledge that exists in my understanding of my palindromic markers.
>
> I don't understand exactly what you meant, in (i), by "inversion". Is
> this that follows correct (going in the order P3 to P1):
> BEFORE INVERSION: DYF397, DYF399(ins G, T-type), DYS464(G-type), DYS725,
> DYS725, DYS464(C-type), N.N., DYF397.
> AFTER INVERSION: DYF397, N.N., DYS464(C-type), DYS725, DYS725,
> DYS464(G-type), DYF399(ins G, T-type), DYF397.
>
> (1) I did NOT take the complement because DYF397 has to be the SAME
> BEFORE and AFTER. You just meant a REVERSAL happens, right?
>
> This is the same picture as before except N.N.(what is this?) is between
> P3 and P2, DYS464(C-type) is on top and DYS464(G-type) is on the bottom
> on the P2 arm, and we have the DYF399(ins G, T-type) on the top lined up
> with a N.N. on the bottom on P1. This would trigger a recLOH event,
> right? But couldn't the DYS399(ins G, T-type) get wipped out, replaced
> by N.N. just as well? Or does the repair mechanism try to repair &
> preserve the DYF399(T-type) instead, by replacing "nonsense" (the
> N.N.)and the DYF399(ins G, T-type) with a 'normal' DYF399(T-type)'s?
>
> So after the recLOH event we have DYS399(T-type)'s replacing the N.N.'s
> at the bottom of P1? This is wrong isn't it? I think what you are saying
> is that at the bottom of P1 we have DYF399(T-type) on the top and N.N.
> on the bottom. So we still have a problem, which is "solved" by P1 & P2
> moving into the "alternative conformation". I'm stuck here. :-( I have
> to think about this much more and how I get the three 188's in DYF408
> and so on ...
> Al
>
>
>
> > wrote:
>
> >Bonnie, Al & Rebekah
> >
> >here's one possible sequence of events (have Thomas's palindromic map
> >handy!)
> >(i) inversion of the section between the lower DYF397 in P3 and the upper
> >DYF397 in P1 (which moves the "spare" copy of DYF399 from between P3 and P2
> >to between P2 and P1)
> >(ii) recLOH replacing the "ins G" T-type copy (i.e. the one with the extra
> >base) with the plain T-type which following (i) is now able to line up
> >opposite it in a different "alternative conformation"
> >(the above recLOH seems to happen frequently in J so perhaps the inversion
> >is a "deep" one shared by several branches?)
> >(iii) duplication of the section between one of the DYS464s in P2 and the
> >corresponding one in P1; because of the inversion this section contains
> >1xDYF397, 2xDYS464, 2xDYF399 (both plain "T" types) and 1 copy of all the
> >"yellow" STRs, precisely the extra copies reported in the results.
> >
> >There may be other recombinational sequences which come up with the same
> >results, I found another one starting with a P1 inversion but it had four
> >steps so is not quite as parsimonious.
> >
> >More DYF399X results from hg J may provide clues as to which sequence of
> >events is correct and when the various recombinations occurred relative to
> >the known subgroup SNPs.
> >
> >Gareth


This thread: