MikeBGene wrote:
WRE>Caveat: I'm into organismal biology. This biochemistry
WRE>stuff is not my bag. I'm trying to pass on what I
WRE>understand from what others have written. I particularly
WRE>want to thank Kevin O'Brien for his many posts detailing
WRE>biochemical esoterica in an accessible fashion, and I hope
WRE>that I haven't bent the concepts stated by Kevin and
WRE>others too badly in handling.
BV>In which case, one common ancestor would be obvious. If on
BV>the other hand, life is the inevitable result of
BV>complexity, as Stuart Kauffman suggests, or as Michael
BV>Denton appears to believe, life is a natural phenomenon in
BV>the universe, life must have arisen many times. In that
BV>case one common ancestor (or even 3) would hardly be
BV>likely, would it?
WRE>Life can commonly arise without the necessity that it commonly
WRE>arise many times *in the same ecosystem*. The scenario
WRE>described conforms to a "winner-take-all" situation. Chemical
WRE>resources would exist prior to the first self-replicating
WRE>system that would thereafter be sequestered in instances of
WRE>that self-replicator. While not an absolute bar to further
WRE>novel self-replicating systems, it would certainly reduce the
WRE>likelihood of further ones arising.
MB>But there is no reason to think there should only be one ecosystem.
WRE>On earth, there are enough means of transporting small
WRE>chemical packages from place to place and enough relatively
WRE>inaccessible places that neither a view that there could only
WRE>have been one ecosystem nor a view that there must have been
WRE>more than one ecosystem can be held to be evidently true. I
WRE>was arguing against the claim that "life must have arisen many
WRE>times" with apparent application to life arising *here on
WRE>earth* many times. While I cannot rule out multiple instances
WRE>of abiogenesis, neither does it appear that I am obligated to
WRE>accept that multiple instances occurred here.
MB>I agree with you that there isn't enough evidence to argue
MB>that life must have arisen many times, but I think Bertvan
MB>has a point.
That's OK, I addressed that point.
MB>With abiogenesis, there are two basic schools of thought.
MB>Christian DeDuve (and I suppose Denton and Kauffman) favor
MB>the view of chemical determinism where, given the
MB>appropriate conditions, abiogenesis is inevitable. Others
MB>view abiogenesis as a highly unlikely event that depends
MB>essentially on the luck of the draw. If the chemical
MB>adeterminists are correct, we still can't say that life
MB>must have arisen many times, but this does indeed appear to
MB>be a natural implication of this line of thinking.
MB>Inevitable things tend to be likely and likely things tend
MB>to happen often. Thus, I do think the fact that the
MB>evidence indicates life is monphyletic is a real problem
MB>for chemical determinism (but it is not a problem for the
MB>lucky accident view).
I think that's a non sequitur. Monophyly of life on earth
does not imply singularity of abiogenesis. See the
restoration at the end of this post.
MB>One can, of course, raise ad hoc scenarios where the first
MB>life forms outcompete subsequent life forms, but these are
MB>easily canceled out by other ad hoc scenarios where this
MB>would not happen.
Only if we all agree that all such ad hoc scenarios are
equiprobable. I don't see that as happening any time soon.
MB>If a self-replicating system arose, I agree this would decrease the
MB>likelihood of another self-replicating system arising *IF* the
MB>original uses the resources needed to generate the other.
WRE>So far, so good.
MB>But this doesn't effect *different* life forms from arising.
WRE>Hmm. I don't see how this can be supported. Sequestration of
WRE>significant chemical resources surely will have an effect on
WRE>other abiogenesis events, whether they would result in similar
WRE>results or not. Also, to accept the assertion as given one
WRE>must ignore the possible products of such chemical systems,
WRE>which may inhibit the formation of other self-replicators.
MB>All of this is possible. And given that abiogenesis is
MB>largely imaginary, we can invent scenarios that go either
MB>way. For example, sequestration may prevent other life
MB>forms from arising *nearby*, but the earth is a huge
MB>laboratory. If sequestration is thus to be significant,
MB>these primitive systems must have been quite efficient very
MB>early on and the raw resources provided by the planet must
MB>have been scarce.
I'm not sure I buy that. As Mike points out later, this
question hinges on whether the rate of abiotic production
exceeds the rate at which some self-replicator sequesters
resources. It's a common intro biology problem to figure
out how long it would take for even slow-reproducers, like
elephants, to reproduce beyond earthly resources.
MB>If, on the other hand, the primitive systems were not
MB>efficient (and I see no reason to think they would be) and
MB>the earth was providing plenty of abiotic material (a
MB>common assumption), there would seem to have been plenty of
MB>abiotic precursors to go around. And since inhibitory
MB>byproducts would quickly dilute out through diffusion and
MB>degradation, newly forming systems on the other side of the
MB>planet would appear to me to be quite safe.
Well, I guess we will have to agree to disagree on this point.
MB>For example, supposedly a self-replicating system arose that uses
MB>L-amino acids. Why didn't one arise that uses D-amino acids?
WRE>Modern organisms predominantly utilize L-amino acids. In the
WRE>(rare) cases where a D-amino acid is used, there seems to be
WRE>a specific conversion process to go from an L-amino precursor
WRE>to a D-amino acid. This situation, though, does not constrain
WRE>the initial self-replicating system to exclusive use of L-amino
WRE>acids.
MB>Given this initial self-replicating system is entirely
MB>hypothetical, I don't see how we can put any constraints on
MB>it.
But we seem to be having fun trying.
WRE>The instantiation of the biochemistry needed to
WRE>generate amino acids from other chemical precursors may have
WRE>come later, and with it the observed reliance upon one set of
WRE>amino acids rather than their chiral opposite numbers. Even
WRE>based upon consideration of biochemical systems made from
WRE>mixtures of amino acids of both L- and D- chirality, one could
WRE>argue that self-selective effects could lead such systems to
WRE>accentuate any imbalance in the proportion of chiral isomers
WRE>utilized until either L- or D- isomers were predominantly or
WRE>exclusively utilized.
MB>Yes. Let's say a population of self-replicating systems
MB>begins to favor the L-isomers and begins to predominate.
MB>But if a competing population begins to favor the D forms,
MB>it will quickly carve out its own niche so it is no longer
MB>competing with the L forms. This is how evolution
MB>typically works. Why would these initial life forms all
MB>jockey for the same niche? Why wouldn't they simply
MB>specialize along different trajectories to avoid direct
MB>competition and stake out their own niches?
Hmm. I thought that we were discussing the scenario where one
self-replicator precedes any other, not some scenario where
multiple simultaneously produced self-replicators interact.
WRE>The existence of specific conversion mechanisms in modern
WRE>organisms to make the odd D-amino acid from a L-amino acid
WRE>is suggestive that at some point there might have been such
WRE>mechanisms to convert a surplus of free D-amino acids to
WRE>their L-amino acid counterparts.
MB>What makes you think these conversion mechanisms did not
MB>arise long after the origin of life?
What makes Mike think that my argument is sensitive to that?
MB>Yes, bacteria use a D amino acid to make their cell wall,
MB>but would you argue the bacterial cell wall is older than
MB>the ribosomes?
I don't know which came first. Is that relevant?
MB>The self-replicating system that uses L would not take anything
MB>away from the environment that would prevent the appearance of a
MB>D-life-form.
WRE>This assumes chirality dependence as a characteristic of the
WRE>first self-replicating system. That might be true, but it is
WRE>not necessarily so.
MB>I'm not making any necessary claims.
Good. Neither am I.
WRE>There are several useful properties for
WRE>protein construction that are available only when the proteins
WRE>are composed entirely of either L- or D-isomers, but this does
WRE>not establish that those properties were necessary to the
WRE>first self-replicating system or even to early
WRE>self-replicating systems.
MB>Once again, since the first self-replicating system is
MB>imaginary, one can not make any such necessary claim
MB>about it. I simply question your "winner take all" claim
MB>as it is neither necessary that the winner would take all.
I don't think I ever said it was a necessary conclusion.
I think that it is a reasonable stance, contrary to
Berthajane's assertion.
WRE>If no such dependence is necessary,
WRE>then an initial self-replicator which scavenges both L- and
WRE>D-amino acids would reduce available amino acids to non-useful
WRE>levels for other self-replicator productions wherever the
WRE>first one reached.
MB>Do you have one little piece of data to support this claim?
About as much as for claims of multiple abiogenesis events
having happened. Maybe more. After all, we can observe the
behavior of modern self-replicating systems as to how they
utilize available resources.
MB>What if the global rate of formation of amino acids is many
MB>times greater than the rate at which this primitve
MB>self-replicator sucks them up?
Self-replicators have this property of often following an
exponential growth curve. I think that the suggestion that
abiotic production could match or exceed this is not very
likely.
MB>What's more, life uses only a small subset of all possible
MB>amino acids (only 20; and some of these are thought to have
MB>been acquired into life after its appearance). Given that
MB>abiogenesis experiments produce many other kinds of amino
MB>acids not used by life (often in greater concentrations),
MB>why didn't a life form appear that used these amino acids?
WRE>Why, indeed?
MB>Who knows? But it is a problem for the view of chemical
MB>determinism.
Is it? It doesn't appear to be much of a problem if what
I've seen so far typifies the objections.
WRE>Question: How many of these non-biological amino
WRE>acids are present in, say, our modern oceans? Our modern
WRE>lakes? Marine hydrothermal vents? If they are not present
WRE>in useful concentrations, why aren't they there?
MB>Many would argue that the oxygenic atmosphere has
MB>essentially shut down abiotic amino acid synthesis.
That pretty much corresponds to my point about the products
of an early self-replicator affecting the conditions that
could give rise to others.
WRE>Possible answer: Non-biological amino acids may have been
WRE>*used* by an early self-replicator, but not simply grabbed
WRE>and incorporated into protein structure. These may have
WRE>been processed into other, more useful, chemical products.
MB>And unless they were very, very good at using these and
MB>there were very few amino acids to begin with, I don't
MB>see this as likely.
This will have to be another point upon which we will disagree.
WRE>Second possible answer: Perhaps these non-biological amino
WRE>acids also commonly have R-groups that interfere with protein
WRE>stability. If so, then these amino acids would make
WRE>relatively unpalatable "leftovers" for a second or further
WRE>self-replicating system.
MB>If this is the answer, the implications are mind-boggling.
MB>For what this means is that the twenty biological amino
MB>acids are essentially the only ones that can form stable
MB>(and thus functional) proteins. This then means that the
MB>universal use of these twenty amino acids is NOT evidence
MB>of a universal common ancestor (as is commonly claimed).
Really? I don't recall that the mere similarity of amino acids
used is *commonly* claimed as evidence for common descent.
Some examples would be useful. If it is commonly used (and
established as such), I would join Mike in arguing that it
should not be.
MB>It would mean that if bacteria were found on Mars that also
MB>used this same set, we could not use this as evidence that
MB>they are related to Earth forms.
I would agree on that. More would be needed.
MB>It would mean that once we begin to explore the universe,
MB>we would expect to find life forms that used this same set
MB>of twenty amino acids.
I agree again.
The specificity that implicates common descent is not the
particular set of amino acids commonly used in biological
organisms, but rather in the pattern of association between
genetic information and those amino acids. This association
does not appear to be chemically foreordained. We can
consider the canonical genetic code to be a mapping from
triplets of nucleotide bases (64 possible triplets) into 21
categories (20 amino acids plus a "stop" symbol). This is a
partition, and the relevant computation of possible ways this
can be done with the same distribution of numbers as in the
canonical genetic code is
N! / (n_1! * n_2! * ... * n_k!)
where N = 64 and k = 21, and the number in each set is derived
from the canonical genetic code. This would give us a lower
bound on the number of possible different ways of apportioning
the genetic codes to the amino acids, for this would not
include the possibilities for those codings that do not preserve
the same numerical parameters as our canonical genetic code.
Here are the numbers:
ala 4
arg 6
asn 2
asp 2
cys 2
gln 2
glu 2
gly 4
his 2
ile 3
leu 6
lys 2
met 1
phe 2
pro 4
ser 6
stop 3
thr 4
trp 1
tyr 2
val 4
Plugging these in, I get:
64! /
(4!6!2!2!2!2!2!4!2!3!6!2!1!2!4!6!3!4!1!2!4!) =
1.27E89 /
(24 * 720 * 2 * 2 * 2 * 2 * 2 * 24 * 2 * 6 * 720 * 2 * 1 * 2 * 24 * 1 * 2 * 24)
=
1.27E89 /
5.28E14
=
2.40E74
That's how many possible mappings with the same numerical
distribution as our canonical genetic code there are. As I
mentioned, this is a lower bound of total possibilities.
(I'd appreciate corrections on my math above if there is
a problem.)
If someone knows the relevant math for directly working out
the odds of obtaining the canonical genetic code, I'd
appreciate finding that out.
Note that other possibilities for genetic code mapping might
be possible, including different numbers of bases in codons
(two bases per codon might be too restrictive, but there isn't
any necessary upper limit on number of bases per codon), or
mapping schemes that could vary the number of bases that code
for amino acids. This would further expand the possible space
for genetic codes.
MB>And that's the ironic thing about chemical determinism - in
MB>seeking to explain the features of life as chemically
MB>inevitable, it essentially paints all possible life forms
MB>into one category - ours.
Uh, where's the irony?
MB>The bottom line is that there would be lots and lots of
MB>leftovers after the "winner takes all."
WRE>The bottom line is that given certain favorable assumptions,
WRE>one could postulate plenty of useful leftovers, while under
WRE>other assumptions favorable to a hypothesis of a single
WRE>earthly abiogenesis event, there may not have been much
WRE>leftover or much leftover of any great utility after the first
WRE>self-replicator made its appearance. While I cannot in
WRE>principle exclude the possibility of multiple origins of life
WRE>on earth, neither am I obligated to accept as a given that
WRE>multiple origins of life actually took place here, contrary to
WRE>the assertion made by Berthajane.
MB>That's fair. But you do then realize that your
MB>winner-take-all scenario is dependent on a small subset of
MB>these assumptions?
The size of the subset says precisely zip about the likelihood.
Mike left off the part of my post that addressed what he claimed
was Berthajane's point. I will restore it.
Of course, common descent is not restricted to the idea of a
single origin of life -- all that common descent requires is
that modern organisms are derived from one or a few original
forms (where "few" is whatever number the evidence indicates).
These may have themselves been derived from the first
self-replicator, or from a second or subsequent
self-replicator which eventually displaced its competitors.
Even if life originated in multiple various forms here on the
early earth, what the evidence supports is that only one or a
few closely related instances of these initial systems
eventually gave rise to all life on earth. The fact of the
canonical genetic code and the similarity of the genetic
apparatus across all life argues strongly for a very
restricted set of common ancestors for all life.
Let me also repeat what I said that led to this part of the
thread:
WRE>"- Patterns of differences in sequences of proteins and heritable
WRE>information support the idea that these differences have accrued
WRE>since the time of a last common ancestor."
I don't see anything in our further discussion that would put
this statement in any doubt.
Wesley