|
|
JAE wrote:
> You can't think of it as a missing chromosome so much as a fused
> chromosome. Losing a chromosome AND all of the genetic material on it
> would most likely be lethal, but the difference between humans and
> chimps wasn't a wholescale reduction in genes so much as it was merging
> two chromosomes into one. This can (and probably usually does) cause
> wholesale problems with meiosis and is thus strongly selected against,
> but it doesn't have to cause a problem if all the parts still line up
> properly.
>
> This is different than an extra chromosome where there's extra genetic
> material which probably causes regulatory problems.
>
> [snip]
To lose or gain an entire chromosome between our development and that
of chimps (our closest current relatives) does seem a little unusual. I
wonder how many monkey or ape speciations ever did this before ? (This
is probably a stupid question, but it may be one of the crucial ones).
snipped
> It's not entirely far fetched and there's nothing completely
> discrediting it at any particular level, but it doesn't have the strong
> hook for me at this point, a strong hook like haplotypes on two chimp
> and gorilla chromosomes
What, exactly, is a 'strong hook?
>that suggest alignment with regions of our
> second chromosome where gene trees, when compared to other regions of
> the chromosome, suggest some reticulation in the resulting phylogenies.
> If I saw something like this, I'd be inclined to think more strongly
> about hybridization and THEN start thinking about population dynamics
> that might lead to fixation afterwards. The Baird paper doesn't
> indicate chromosomal fusion though as the haplotypes they examined
> were, I believe, on chromosomes 12 and 7. I don't instinctively see
> any suppositions that toss your speculation, but I also don't see
> reason to proceed to many of them without first better establishing
> that hybridization was likely.
Arithmetical arguement snipped - "Statistics are always lies"
> Negative assortative mating happens but it's not the rule. What you
> have to have happen is negative assortative mating in the F_o
> generation and positive assortative mating among hybrids in the F1.
> Could it have happened? Sure, but it's putting the cart before the
> horse. Still a long way to go with the molecular evidence to make the
> case that hybridization happened first.
>
> > > Admittedly, I don't know if this is the proper way to score large scale
> > > chromosomal changes, counting them as single allele changes in part
> > > because I don't really know what the intermediate looks like. (One
> > > complete chromosome and two unfused ones with the same genetic material
> > > in the same individual? That's the way I'm picturing it.)
> >
> > That seems to be what most of these models assume. But what if it
> > happenned instantaneously - (48 -> 46) due to a telomeric fusion in the
> > fusion of the gametes? Why is this not also plausible?
>
> It's plausible, but I don't see why hybrids make this more likely at
> this point. Again, this is in part simply because I don't really know
> much about the mechanisms of chromsomal fusion but also because I've
> not seen anything in what I'm more familiar with, comparative genetic
> phylogenies, that makes me think that hybridization was likely. I
> still have to have reason to believe that hybrids of two lines that had
> split long enough to achieve some level of genetic distinction on what
> became our second chromosome occured.
>
> [snip]
>
> > > I'll have to plead ignorance here. It's possible, but I think the
> > > hard-core MCB people have to weigh in on this, if there's things that
> > > favor karyotypical change.
> >
> > Sorry, what's MCB? (Molecular-Cellular Biology.) The geneticists I've
> > talked to about this seem sceptical, I have to admit - story of my
> > life!
>
> Yes, molecular cell biology.
>
> The skepticism is still warranted in large part because there are
> unknowns and unknowns should generate skepticism. These may even be
> testable unknowns and knowing more about the mechanisms of chromosomal
> fusion might make some of the unknowns more palatable if they're better
> ways of explaining what we see, but currently, they're just unknowns.
> I'd have to see some more evidence of hybridization relating regions
> that would have favored chromosomal fusion to really put much behind
> it. However, with this evidence, it would be reasonable to start
> trying to figure out how a hybrid species caught on and replaced parent
> species.
>
> [snip]
>
> > I just think the implications of this are potentially massive.
Well, if a hybrid species 'caught on' and had distinct reproductional
advantages over the parent species, of course it would survive better.
What, to
> > just raise one question, would it do to the molecular clock evidence?
> > Loren Reiseberg wrote to me that he'd expect a real hybridisation event
> > in the history of a species to perhaps halve the estimated divergence
> > date given by the molecular clock, because half the variation in the
> > genetic material seen would have to be accounted for by the initial
> > hybridisation, rather than drift since that point. It's a fascinating
> > thought. If it happenned once, it might have happenned many times. It
> > would certainly throw a completely new light on a'pith-like species if
> > the Pan-Homo split actually happenned only 2.5-3.5My ago.
>From what you've just said, it would certainly 'throw a new light'. But
what, exactly is the molecular clock ? It seems to be a reductio ad
absurdem ...
To take an example:
1) Crocodiles (or whatever) and other vertebrates on the same
evolutionary branch share an ancestor about 500Mya.
2) Crocodiles DNA differs from say, Dinosaur DNa by x%, but from us by
y% - so we're clearly closer to dinosaurs than crocodiles (this is an
illustration, not a fact)
3} Rodents formed their own major branch about 80 Mya.
4) Primates formed another branch (another side from the rodents) about
50 Mya.
So, can we can take the average/mean/median distance and time between
these very major changes and apply it to changes within the last 5Mya -
or even, in Out-of-Africa, to the last 50-100ya?
No
You cannot take very imprecise time estimates, and apply them to the
extremes of the range - in the case of human history, the last 2 My or
so
> I've very skeptical about most molecular clock estimates in general
And so you should be
> (with or without hybridization), but a Pan-Homo split that recent has
> some Himalayian-sized obstacles against it interms of character
> parsimony in the fossil record. The fossil record provides some rather
> compelling reason to set an earlier minimum calibration point, earlier
> than 3.5 million years ago when we already see much more human-like
> forms. Even ignoring those traits associated with bipedalism (and I
> don't think we should ignore these), there was clearly many forms
> rather derived away from those characteristics where Pan retains
> primitive features in the teeth and skull. Either there were massive
> wholescale reversals completely obliterating any traces of change in
> the line leading to Pan OR all of the fossil hominids identified are
> ridiculously homoplastic with our lineage. Neither of these are
> convincing hypotheses in the least and both require a near total
> disregard for parsimony. In most regards (e.g. a pronounced diastema;
> canine honing complex on P_3; flatter cranial base, angle of petrosal
> bone relative to coronal plane; total lack of incisiformation of the
> canines) chimps retain a more primitive dentition more akin to Miocene
> apes and more akin to other Old World primates. By 4 million years
> there was clearly a hominid lineage derived away from this. While
> perhaps not ancestral, this most likely indicates sister taxons of Homo
> had already split off from the line leading to Pan prior to 4 million
> ybp.
You're probably dead right - I am now reading (out of boredom, because
there's nothing else here) Ian Tattersall's book "Becoming Human' -
Amazon sent it to me for free, after a bit of a fuss because I didn't
get what I ordered.
He says: "Recent studies have shown us that we share more of our
attributes with the great apes than we have liked to imagine; but
understanding the differences between them and us establishes the
essential point of departure for interpreting the human fossil record'
Which is, essentially:
"We're different from our closest genetic cousins, but we really have
no idea why, and we've found nothing at all that gives us any clue"
regards
Richard
|
|