🚀 go-pugleaf

       In article <1992Sep22.183733.39095@Cookie.secapl.com>
frank@Cookie.secapl.com (Frank Adams) writes:

>       I think the plurality of molecular biologists believe that
>DNA evolved from earlier systems involving RNA but not DNA.  A
>strand of RNA (with perhaps some restrictions on the base pairs)
>in a soup of amino acids is a self-replicating system.
>Simulations of the conditions believed to have existed in the
>early history of the Earth produce soups of amino acids, so this
>is no way an artificial environment.

       RNA is a 'alphabetic' molecule. It was designed (or
evolved if you like) because it had the property that a large
number of different types of molecules could be put together from
a small number of parts. ('Alphabetic' for the obvious reason
that it shares this property with language.) This
'alphabetization' is not a property of systems, but a property of
tools created by systems, the reason being the wonderful
efficiency which it lends to the system which uses it. Obviously
the cell must constantly transfer information between its
subparts, and must constantly manufacture materials with a wide
variety of properties. The fact that it manages to do this with a
relatively small number of submolecules shows the high degree of
efficiency of the cell as a system. However, to appreciate this
marvel of cellular technology, one must first distinguish between
the builder and its tools.

>       The question of how RNA first came into existence is a
>very interesting one, which we can't answer yet.  Based on our
>current understanding, "just chance" seems unlikely to be the
>answer.  (The chances against are a bit too long, but only by a
>few orders of magnitude; the chance of a cell coming into
>existence by chance are too small by many orders of magnitude.)

       There is no such thing as "just chance". The much
misunderstood notion of "probability" is just a mathematical
method for describing events which are not understood and cannot
be predicted any other way (but through tabulation). Probability by
definition explains nothing. (QM people should face up to this.)

>       However, there is no reason to think that RNA did not
>evolve from some still simpler self-replicating system, for which
>"just chance" is an appropriate explanation.  This is an
>interesting research question.

       RNA does not reduce to a simpler system, i.e. the way a
vertebrate reduces (evolutionarily) to an invertebrate. This
would be like saying a big word like "hamburger" reduces to a
small word like "bun". This is another reason why you know it is
not central, in evolution or in the functioning of the
cell (as with DNA). What happened was that something else
evolved, and learned to store more and more complex information
in RNA  or DNA (if one came first). One of the biggest advantages
of doing this is obviously in the ease with which this stored info
is transferred to offspring -- requiring only a simple copying
mechanism. This would give such systems tremendous advantage over
complex systems which must copy their functionality unit by unit.
Eventually, becoming increasing dependent on this information, it
would be discovered that allowing mutation of this information itself
was a successful strategy -- up to a point. (That is why DNA repair
mechanisms are so complex.) At any rate, controlled mutation of this
storage information itself would become an important aspect of evolution.
But initially it was the mutation of the underlying system. This
type of mutation must still be occurring today, needless to say, or
at least that is a testable hypothesis.

>       Such an RNA self-replicating system will evolve just as
>DNA does: by mutations changing the base pairs.  This will affect
>the formation of proteins around the RNA, which will favor or
>disfavor the replication of the RNA; hence natural selection
>drives the evolution.

       A block or string of material with no internal
organization is not a system. RNA was created _by_ a system to be
easy to copy, and this is why biologists have been able to copy
it: they just use a similar procedure to the one the cell uses.
This has led to the erroneous notion that it is self-replicating.

>       This theory postulates that at some point, the
>self-replicating RNA generated DNA.  Exactly what evolutionary
>process could have led to this is another interesting question.
>At some point after DNA came into existence, the DNA-based
>systems won the battle for survival, and independent RNA-based
>systems became extinct.

>       This is only one theory, and it is still very incomplete,
>but I think it answers the general question of how DNA could have
>evolved.

       I believe Freeman Dyson originated this "dual origin"
theory. Really, it  is overly complex, all just to avoid the
notion that a non-nucleic acid is in control. Besides, as I said,
the notion of RNA as an originating system is absurd. RNA is not
a system. This is like saying a fish backbone created the cells
and brain of the fish so it could get from place to place.

>>      Specifically what reactions. And how do these
>>reactions support your position that DNA is the "brains" of the
>>cell, rather than just an information storage structure?

>       Different parts of the DNA make themselves differentially
>available for transcription by RNA, thus directly affecting what
>proteins get created.

       Are you suggesting the DNA molecule is "aware" that
certain things need to be accomplished, and subsequently
initiates them? If not, then why are you suggesting DNA plays an
active role?

>>      Given [meant "give"] an example of a system which
>>is considered an integral whole, but does not have a hierarchical
>>structure (which doesn't exclude distributed subsystems by any
>>means) and is not centrally controlled in its ultimate functions.
>>When you suggest that we think of something as a "single" system
>>because it is useful, my suggestion is that it is only useful
>>when that single system is centrally controlled.

>       It is also useful if the control is diffused throughout
>the system.  Your (deleted) discussion of multiple control
>systems is a strawman.  In the case of the cell, it appears that
>most of the activity in the cell is part of the control system of
>the cell (besides, often, serving other purposes).  I know of no
>other example of a complex system with a distributed control
>system; but to deny that the cell has such is to ignore the
>evidence.

>       (This may seem to contradict my claim that the DNA is the
>control system for the cell; the truth is somewhere in between.
>DNA is more than just a repository, but it is not a control
>system by itself.  On the other hand, while the control system is
>distributed through the cell, it is not distributed equally; it
>centers on the DNA.)

       The 'evidence' cited in the current literature would
not contradict a central control system. But tell me this: how is
it thought various messenger molecules like RNA know where to go
once they are formed? How do they physically move through the
nucleus and endoplasmic reticulum or whatever? What stops such a
molecule from getting distracted, halfway between reactions,
turning around, and going off somewhere else? How does a specific
catalyst find its way to the exact correct site to unzip the DNA,
while thousands of other, different catalysts, are also  doing
this simultaneously? Notice that there are legions of molecules
all moving around to and from the nucleus with different
purposes. The cell is like a bustling city of molecules in
motion. How is this physical motion controlled by a chain of
nucleotides?

>       We have two broad classes of control systems: hierarchical
>and feedback. A "purely" hierarchical control system runs into
>the problem of infinite regress: [...remaining possibilities
>deleted]

       Why is infinite (or indefinate) regress rejected out of
hand? What is known about the structure of the material world
which, even if you find the word 'infinite' distasteful, makes
you think that there is not regress at least beyond the level of
the cell we can currently view? Isn't this just the current
mechanistic paradigm in physics? Do you think it would be
non-constructive for molecular biologists to wonder if they have
the full explanation with DNA?

>       The paradigm shift I'm talking involves the ideas of
>non-linear systems, complex feedback, chaos, self-organizing
>systems, strange attractors, and so forth.  It looks to me like
>you are the one that doesn't want to accept the new paradigm, and
>so rejects its validity.

       No. The properties listed above are the result of the
interaction of the complex systems nature is made of. Naturally,
when indefinately complex control systems interact, each intent
on maintaining the identity of its own system, self-organization
and chaotic environments result. IOW these areas are not at
all in contradiction to a systems viewpoint. In fact, the study
of cellular automata is highly revealing from this standpoint, as
an abstract model for provoking thoughts about the evolution of
and development of material structures.

>       A key point: we *know* that complex feedback systems do
>not necessarily behave chaotically.  Such systems may exhibit a
>wide variety of behaviors, including complex but orderly behavior
>which is relatively stable in the face of outside perturbation.
>This is a fact. It leaves very little room for the claim that the
>cell is too complex and orderly to be a feedback system.  You
>would have to establish that there is some upper limit to the
>complexity thus achievable; but the theory suggests that there is
>no such limit.

        You're right. It wouldn't appear this is an profitable
argument to pursue -- though intuitively I disagree with you. IMO
the very notion of the stability of systems comes from observation
of such systems in nature. Feedback systems, on the other hand, are
technological creations, i.e. machines, and recognized to possess
inherent instability. To then say natural, self-organized systems are
actually mechanical feedback systems -- just really sophisticated ones --
well, we're back to the self-replicating machine again, and the
general debate of a complex systems vs. a mechanistic approach.

      John E. Snodgrass [snodgras@cyber.net (mail)]