Re: [MLUG - DISCUSSION] Identical Twins

[Mike Miller <EMAIL:PROTECTED>]

On Tue, 4 Jan 2005, Nathan Odle wrote:

> Ok, so on to the types of twins.

I believe that the information you're getting about the timing of the 
zygote split is all theoretical and not well established (especially in 
the case of the conjoined twins).


> So how to tell for sure?
>
> There are a few answers to this question.  One obvious, but not 
> necessarily accurate, solution is to wait until they're born and see how 
> much they resemble one another :)  This isn't nearly as good as most 
> people realize though because a lot of features come about as a result 
> of conditions in the womb - which can both make fraternal twins look 
> alike (Mary Kate and Ashley, anyone?) or identicals appear a little 
> different (they're never EXACTLY alike are they?).

Two things:

First, Mary Kate and Ashley are MZ twins, no doubt.  They say that they 
are DZ, but they either are lying or they've been misinformed.  What is 
the nature of their evidence?

Second, The diagnosis of zygosity by facial appearance is very accurate. 
One expert friend (Nancy Segal) told me that she has no doubt the Olson 
twins are identical and I believe her.  In another interesting example, an 
expert diagnosed 166 pairs by appearance, but the lab suggested that 2 or 
3 pairs diagnosed as MZ were actually DZ.  It turned out that the lab was 
wrong.  Lab errors had created false differences between some pairs of MZ 
twins so they were wrongly diagnosed as DZ by the lab.

Zygosity diagnosis by appearance works well with children and adults, but 
I don't think it works so well with infants!


> Twins with different genders are also a giveaway, because it's not 
> possible for identicals to be different gender, as their DNA is the 
> same.

Right ... almost.  There is a possibility of an error in cell division 
causing a normal XY male and an X female (Turner syndrome, usually called 
45,XO karyotype):

http://taxa.epi.umn.edu/bgnews/1997/msg00266.html

That is very, very rare and so you are basically right, but I just want to 
throw this one out there because it's interesting.


> With that somewhat-unscientific solution aside, there are a couple 
> others. The solution that appears all over the websites is genetic 
> testing.  For $125-175, you can do cheek swabs and send them off to be 
> analyzed.  The results come back in 4-to-6 weeks, and voila you know 
> what you've got.

That should work, but they have to take lab error into account in their 
statistical analysis.


> What our doctor did, was look on the ultrasound for the Corpus Luteum 
> cyst that forms when the ovary releases an egg.  This cyst is perfectly 
> normal, and performs several functions including the generation of 
> hormones, etc.  So how to tell if they're identical?  If you have two 
> cysts, it means she released two eggs.  If there's just one, then she 
> only released one.  One egg == one zygote == monozygotic, or "identical" 
> twins.  Seems that it's just that simple.

OK, but if there is no literature on the accuracy of the method, I have to 
express some reservations.  How do we know he didn't miss one?  How do we 
know that they aren't so close together that two appear as one?


> Now, there's one catch to all this - doctors suspect the existence of 
> yet another twinning mechanism called "Polar Body Twinning", whereby the 
> egg divides....BEFORE it's fertilized.  So what you get is a rather 
> unique situation where the twins share *half* their DNA (the half 
> donated by the egg).  Of course, counting the CL cysts would not be of 
> benefit in this situation because there's still only one egg involved. 
> However, it's thought that if the mechanism exists at all in humans, 
> it's pretty rare.

I don't know that this has ever been established.  A more common and 
interesting phenomenon:  Twins with two fathers - maternal half-sib twins. 
This happened in the Netherlands a few years ago when a woman was 
artificially inseminated with her husband's sperm, but the lab equipment 
wasn't kept clean and another man's sperm was thereby accidentally (we 
believe!) mixed in.  The result - one white child and one half-black child 
(the other man was Afro-Carribean).  This was quite a shock to the 
parents, I'm sure!  There have been other cases by natural insemination, 
but people avoid talking about those cases!


> As an aside, are any of the genetic hobbyists on the list capable of 
> doing the genetic testing thing?  I think it'd be a cool thing for the 
> baby book :)

It sounds like you found a good price.  I would want to know how many loci 
they are typing, and which loci.  One would want to use highly polymorphic 
loci scattered about the genome.  The more the better, but a small number 
(e.g., one locus per chromosome for five chromosomes) would give you a 
pretty convincing result -- math available on request.

Mike
_______________________________________________
discussion mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/discussion