Dear Colleagues:
I have just returned from an stimulating meeting on the philosopher /
scientist De Chardin. His global perspectives are a truly
remarkable synthesis - even after one sorts out the "noise" factors.
It makes the challenges of Bionetworks seem more approachable. :-)
Since several responses have commented either directly or indirectly
on logic, this message will seek to explore some facets of relations
between chemical logic and other logics. Thus, I will address the
first question in my list:
> 1. From an informational perspective what would constitute a
> definitive theory of the information content of molecular bionetworks?
>
First, two historical comments that are not part of the general
scientific folklore or narrative generation (that Stan is so fond
of. :-) )
About 1875, after Kirchoff's laws had penetrated the scientific and
mathematical communities, two deep results on chemical logic came
into existence.
One conclusion was that the networks of Kirchoff could not be used to
describe chemical structures. Nonetheless, the "drawings" of
Kirchoff's networks and the "drawings" of chemical structures look
very very similar. If we call such drawings by the formal
mathematical name of "graphs", then the Graphs of Kirchoff morphed
into groups and became closely associated with differential equations
and factoring of products (polynominals), linear algebra. Thus,
chemical theory separated from mathematical theory.
Secondly, the optical isomer problem of tartaric acid initiated the
general problem of isomers - substances of the same molecular formula
but different physical - chemical properties
So, what is the logic of a chemical structure? This is a crucial
question. If we can not answer this question, how are we to approach
the logic of bionetworks? of genetics? simulation? Is an
alternative grounding (not chemical structures) for these logics
conceivable?
It is important to note that the physical community generally
presupposes that "quantum mechanics" or "symmetry" or
"thermodynamics" provides the logic to solve these problems. I
invite individuals to address the relations between physical theories
and these two questions - it would be a healthy opening for further
discussions.
One of the problems with the logic of chemistry is the relation
between principle of the conservation of matter and the nature of
arithmetic operations. The conservation of matter demands that each
chemical particle is preserved in its individuality.
Example: The atomic number of Hydrogen is one and of Carbon is six.
Thus, we speak of C6 H6 (Benzene) as 6*6 and 6 * 1 as 36 + 6 = 42.
But 42 is neither the molecular formula nor molecular weight of benzene.
Thus, a significant logical separation of the nature of numbers of
chemistry from the nature of numbers of mathematics exists. This
distinction casts a long shadow over the concept of bionetworks.
Ted states:
"In the local context, all you have to do is invent a new science
that sheds new light on phenomenon that are seen locally, namely in
the molecular domain."
In an overarching view, I agree with Ted that "all you have to do is
invent a new science". But, in the case of chemistry, it was also
necessary to invent a new mathematics. This new mathematics is the
basis of bionetworks. In addition one has to invent a method of
scaling because arithmetic operations do not work on atomic numbers.
Ted continues:
"In the FIS context, you have this seeming impossibility squared.
That's because you'll be using this domain to get leverage on the
larger problem of an information science that applies at all domains.
And more importantly, we'd understand what flows among domains and how."
I agree with the sentiment but not the impossibility of doing it.
I think we must seek deeper abstract methods of understanding the
sources of communication. All to often, we tend to fall into
scientific or philosophical jargon and reduce the concept of
communication to the concept of probability and then to the concept
of "transmission" in a linear symbol sequence, and then into
simulation of symbol behavior. Rather than converging to
calculations, we must think in more abstract views that have the
potential to link to calculations at some latter logical step.
This approach of immediate convergence is "going the wrong direction"
in my view. We ought to first diverge into wider philosophical
territory that can become a basis for latter calculations. Pedro has
suggested some possible avenues that deserve attention when he writes:
In our times, I would say that the "bioinformational" themes have
definitely gone beyond the unitary schemes proposed after the
adoption of the information metaphor (codes, signals, translation,
transduction, etc.), for instance, the case of the so called Central
Dogma of molecular biology.
The challenge is to go beyond the "genetic code." The code is now
more than 40 years old. It itself was strongly influenced by
Shannon / Weaver/ Weiner, etc.
Yes, molecular biology has progressed dramatically over the past 45
years - but almost exclusively on the basis of empirical evidence,
not new methods of calculations. The major exception to this wide
claim are the raw combinatorial calculations for comparing sequences
of various polymers. By comparing sequences (proteins, DNA) we have
illuminated the pathways of evolution of species. Remarkable
progress, heavily dependent on statistics but only lightly touched by
information theory.
To return to the logical issues of communication, classical logic is
so "hung up" on the grammar of predication that formal analogy with
chemical bionetworks appears hopeless. Bio-logic appears to me to
vastly deeper and richer than either propositional logic or first
order logic. Pedro appears to agree with this statement when he writes:
"So the �what� of the functional clause should be accompanied by
circumstances such as �how fast�, �where�, �which way�, �with whom�,
�when�, and �how long�."
This comment of Pedro recalls Robert Rosen's text in Anticipatory
Systems on the role of interrogatories in general. The critical
point is that chemical relations among molecules are vastly richer
than the notion of a variable. The conformation of a protein
molecule can occur in many different forms and functional states.
So, the concept of a molecule should not be confused with the concept
of a variable. The pathways of empirical evidence are open to
exploration but the methods to express the formal logic are
deficient. In a practical sense, if a protein of 10,000 atoms
interacts with three substrates, each of 100 atoms, then how can we
express the logic? Empirically, I think that chemists use a private
version of Aristotle's ten categories. In some sense or another, if
the protein and the three substrates come from the same organism, the
molecules "know each other". By this I mean that the mutual source of
co-emergence of the bionetwork has adapted the structures so that
each contributes to the reproduction in a meaning way without
untoward contempt for one another. But modern philosophers have
rejected Aristotle's categories. (See for example, Logical
Properties by Colin McGinn.)
Thus, I sense a deep logical disconnect between the philosophical
approach to logic, the mathematical approach to logic and the
chemist's approach to logic. From an evolutionary perspective,
chemical logic must anticipate both propositional logic and model /
mathematical logic. (I believe that this is an historical fact that
some may find distasteful. Other viewpoints on this critical matter
are desirable.)
Ted continue:
"Toward this end, I'd like to put an early suggestion for limiting
the "bio" semiotic approach a bit. Several smart people here have
noted the internal/external problem of meaning and intent when you
scale biosemiotics down to elements that apparently cannot reason. So
I am very skeptical of Peircian mechanics at the biological level."
In recent discussions on the Pierce listserve, it was concluded that
Piercian semiotics is infinite. Roughly,the reasoning was as
follows: If sign B can replace sign A, then sign C can replace sign
B, and so forth. Simple transitivity of sign replacement. (This
view is foreign to most discussions of Piercian semiotics I have
seen!) So, I must ask Ted for two clarifications.
The internal / external issues have physical and cellular
counterparts. What is the reference?
The critical role of distinguishing logically among internal /
external issues drives me to the concept of organization levels and
the separation of categories by the choice of classes (categories)
and relations.
How are we to interpret the concept of genetic reasoning? By genetic
reasoning, I mean the choices of an organism to "turn-on" some genes
and "turn -off" others in response to the external environment?
"Choices" in this context can be used in the informational sense.
Further discussions of logics are needed; I definitely concur with
Ted on this issue. The challenge is to find ways to communicate
abstractions among individuals with widely different knowledge bases
and experiences. I have not been able to extend the Barwise
approach to chemical systems - has anyone any ideas on how this might
be approached?
Ted suggests:
I suggest we consider artistic and humanity-centric "logics" also, as
we hunt for tools, and be open to a scope that includes internal
conceptual mechanics: desires, intuitions, emotions, creativity.
This sentence reminds me of the seminal works of (Howard?) Simon and
of Christopher Alexander on the concept of design. We want to design
what? A new symbol system? A new perspective logic? A new view of
relations? A better approach to human communication? What would be
an appropriate source of abstractions for such designs? These are
daunting questions!
Thus, I leave these questions as open to individual
interpretations. Perhaps the philosophers can open up these
questions? In what sense can the concept of interrogatives be
leveraged to openings?
Several issues are left un-addressed, including Kevin's interesting
posts and Koichiro's last post in the previous session. More in my
next post.
Cheers
Jerry
On Nov 8, 2005, at 5:19 AM, fis-request@listas.unizar.es wrote:
> Send fis mailing list submissions to
> fis@listas.unizar.es
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://webmail.unizar.es/mailman/listinfo/fis
> or, via email, send a message with subject or body 'help' to
> fis-request@listas.unizar.es
>
> You can reach the person managing the list at
> fis-owner@listas.unizar.es
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of fis digest..."
> Today's Topics:
>
> 1. ON MOLECULAR BIONETWORKS (Jerry LR Chandler)
>
> From: Jerry LR Chandler <Jerry.LR.Chandler@Cox.net> (by way of
> Pedro Marijuan <marijuan@unizar.es>)
> Date: November 8, 2005 5:20:33 AM EST
> To: fis@listas.unizar.es
> Subject: [Fis] ON MOLECULAR BIONETWORKS
>
>
>
> 9th FIS Discussion Session:
>
> ON MOLECULAR BIONETWORKS
>
>
>
> Jerry L.R. Chandler
> Research Professor
> Krasnow Institute for Advanced Study
> George Mason University
>
> Kevin G. Kirby
> Department of Computer Science,
> Northern Kentucky University
> Highland Heights, Kentucky
>
>
>
>
> Introductory Remarks 1:
>
> 12 Questions about relations between molecular bionetworks and
> information theory
>
> (by Jerry L.R. Chandler)
>
> I agreed to write a brief introduction to the subject, knowing that
> I could not do justice to such a complex topic in such a short
> time. As little is known about the subject, I chose to introduce
> topics and to pose questions about potential relations. The
> questions are significant but not ordered by importance or
> priority. Hopefully, participants will contribute to exploring the
> meanings of the questions and addressing conceivable approaches to
> answers. Perhaps one objective could be to discuss the relative
> importance of the 12 questions.
>
> The concept of molecular bionetworks is a modern concept. The
> development of this concept will continue into the indefinite
> future. That is to say, the concept is, at present, well defined
> but in a primitive or early stage of development.
>
> 1. From an informational perspective what would constitute a
> definitive theory of the information content of molecular bionetworks?
>
> Historically, in the 1840�s, Kirchoff�s theory of flow in
> electrical networks grounds the metaphor of the conceptual
> framework for molecular bionetworks. However, the relatively
> simplicity of the flow of electrons in comparison to the flow of
> life restricts the analogy severely. Nevertheless, biomolecules
> are composed from electrical particles and certain electron flow
> patterns are intrinsic to metabolic networks in living systems.
>
> 2. From the perspective of information, how is it possible that
> bionetworks construct the informed flows of electrical current flow
> and how is it related metabolic flows?
>
> Historically, the investigation of empirical basis of molecular
> bionetworks (metabolism) started shortly after Pastuer�s pioneering
> experiments on the causes of fermentation (1870�s). At that time,
> the fermentation of grape juice was difficult to control but of
> great economic importance. (Bad wine sells cheaply!) Quantitative
> analysis of yeast fermentations showed that each molecule of
> glucose generated two molecules of the 2-carbon alcohol and two
> molecules of carbon dioxide. The yeast cell informed the specific
> destruction of the sugar. (If the 6-carbon sugar was merely
> burned, then it produced exactly six molecules of carbon dioxide.)
> Thus, the thermodynamic question of why the yeast cell did not
> completely burn up the 6- carbon sugar entirely into carbon dioxide
> arose.
>
> 3. From an informational perspective, what does this incomplete
> fermentation process of glucose suggest about the role of
> thermodynamics in living processes?
>
> Continuing with the historical roots of biomolecular networks,
> early in the 20 th Century, the exact chemical pathway between
> glucose and alcohol and CO2 was decoded. The code of the cell for
> fermentation was that each chemical reaction was catalysed by an
> informed agents, termed �zwishenferments�. These agents are now
> known as enzymes. Thus, the question of how a cell knows how to
> ferment sugar was regressed to a deeper informational question:
> How does an enzyme know how to inform a chemical change?
>
> 4. From an informational perspective, what is the nature of the
> code that an enzyme contains such that it conducts an informed
> catalytic process?
>
> Continuing with the historical perspective, by the 1950s, it was
> known that biological information was carried in chemical
> sequences, enzymes were composed from amino acid sequences and that
> nucleic acids were composed from nucleotides. In the 1960�s, the
> convergence of experimental work in genetics, biochemistry, x-ray
> crystallography, nutrition and Shannon�s coding theory, generated
> the hypothesis that biological information is encoded in chemical
> structures. This hypothesis remains intact today. During the
> period from 1970 to the present, methods for determining the
> chemical sequences of biomolecules were developed and automated.
> Today, sequencing of microgram quantities of biomolecules is
> routine. The complete DNA sequences of hundreds of biological
> species are known. The sequences of tens of thousands of proteins
> are known. The sizes of proteins, the so-to-speak �cybernetic
> agents,� are often hundreds as time as large as the small molecules
> that they control. Interpolation between protein sequences and DNA
> sequences are routinely conducted via computer programs. The
> relation between the information in DNA or in proteins and the
> substrates for catalytic reactions is deeply mysterious.
>
> 5. From an informational perspective, given solely the DNA and
> protein sequences of an organism, what can we say about the
> biological functioning of the organism?
>
> The classification of chemical structures has been given a new set
> of names in the past decade. Genomics becomes the study of the
> genome, the DNA sequences and properties; protonomic becomes the
> study of protein structures and properties, metabolonomics becomes
> the study of �small� molecules where the sequences of atoms are
> relatively short. (The use of the Greek root, �nomos�, as in law
> (for example, in autonomy), is somewhat misleading in this context.)
>
> Biomolecular networks consist of all the molecules in a cell. One
> crucial feature of such networks is the capacity to generate
> biological functions.
>
> 6. From an informational perspective, what is the nature of the
> information content of a collection of molecules such that they
> generate biological function?
>
> A second crucial feature of such networks is the capacity, in an
> appropriate ecosystem, to reproduce themselves with exactly the
> same chemical structures.
>
> 7. From an informational perspective, what is the nature of the
> relational interactions between the interior and the exterior of an
> organism such that the information is reproduced?
>
> 8. From an informational perspective, is the reproduction of a cell
> a mathematical calculation?
>
> Biomolecular networks function in time. In humans, the generation
> of temporal networks of the heart, the brain, the menses, and so
> forth are intricate flows of orderly information that manifest
> themselves in regular and chaotic rhythms. In simple unicellular
> organisms, the growth rate and cellular division can follow strict
> rhythms. Such rhythms can be modulated, altered, extended or
> stopped by chemical agents that are foreign to the bionetwork.
>
> 9. From an informational perspective, what are the relations
> between chemical structures and temporal rhythms? What is the
> information content of a chemical rhythm?
> In what sense are the rhythms of temporal chemical relations the
> source of biological information? What is the information content
> of a chemical rhythm?
>
> Early in the history of biology, organisms were classified based on
> categories, more or less following Aristotelian conceptualization
> of Phyphrian trees, choices of potential properties. The ten
> Aristotelian categories, the ten highest genera are listed as:
> substance, quantity, quality, relation, place, time, position,
> state, and action, being acted on. Biological function is related
> to these categories. I presume that a property of the molecular
> bionetwork encodes and decodes information concerning these genera.
>
> 10. From modern information theory, how can we improve the
> Aristotelian categorization of properties of organisms, based on
> the qualitative and quantitative attributes of the biomolecular
> networks? How should the concept of information be conceptualized
> such that it will illuminate communication about molecular
> bionetworks or biological function?
>
> Each molecular bionetwork is specific to the particular species in
> which it exists. However, the similarity of bionetworks roughly
> parallels the separation of species in the �tree of life.�
> Virtually all self-reproducing species include a common set of
> small molecules, amino acids, fats, sugars, vitamins, and
> minerals. The degree of similarity of sequences of proteins, RNA
> and DNA vary with the �tree of life� separation. Indeed,
> classification of species, genera and families are now based on
> calculation of sequence similarity.
>
> 11. From an informational perspective, what is the significance of
> the sequence similarities for the theory of biological
> information? For the concepts of encoding and decoding of
> molecular bionetworks?
>
> One approach to the partial analysis of the meaning of the genetic
> code was developed from Shannon�s theory by Dr. Thomas Schneider.
> http://www.lecb.ncifcrf.gov/~toms/
> The theory is specific for one aspect of the molecular bionetwork,
> namely binding of proteins to DNA. His website provides a detailed
> account of Schneider�s philosophy.
>
> 12. From an informational perspective, the encoding processes for
> DNA � protein relationships are somehow related to Shannon�s
> theory. What are the relations between this sort of encoding and
> other metabolic encoding? In particular, can we imagine a
> catalytic � type of encoding that parallels the genetic encoding?
>
>
> ----------------------------------------------------------------------
> ------------------------------------------------------
>
> Introductory Remarks 2:
>
> The Natural Computation Perspective
>
> (by Kevin Kirby)
>
>
> If anything can be called a "science of the artificial" it is
> Computer Science, or so one might think. In fact, such thinking is
> incorrect. And it is precisely in the field of biology -- with
> molecular biology as the best exemplar-- where a clearer
> understanding of Natural Computation will be so valuable.
>
> The initial encounter between computing and nature happens in the
> computational modeling of natural systems. For example, we now see
> many object-oriented models of ecosystems. More deeply into
> computer science, we find questions such as, for example, whether a
> natural language has such-and-such a computational property (e.g.
> whether Dutch is a context-free language). But beyond providing a
> way to talk about models of nature, some have viewed computer
> science as saying something about nature itself, constraining it in
> some way. A very strong form of the Church-Turing thesis can be
> taken as a physical law. And the theory of quantum information has
> made the "it from bit" slogan quite precise. It is in these
latter
> senses that a true �natural computer science� begins to take shape.
>
> One way to further this approach is to turn to biological systems,
> and take a look at the simulation relation. How do we say a
> biological system computes X? Well, we see if there is a dynamics-
> preserving mapping between inputs and states of the biological
> system and a given formal system for X. This relation is usually
> written as a commutative diagram. The simulation relation is
> central in automata theory and was recast into a category-theoretic
> framework by Arbib and Goguen (taking different approaches). But
> the notion of a mapping between a biological system and a formal
> system, seems to be, at first glance, a category mistake! As soon
> as one identifies a fragment of nature as a system, one has locked
> in some set of states, and it is hard to separate the true
> computational power of a living system from what accrues merely to
> our conventional state assignment. This is taken up nicely by the
> philosopher David Chalmers in a response to a very strong statement
> at the conventionality end by Hilary Putnam. (One could see this as
> a recasting of the debate in Plato's Cratylus in computational terms!)
>
> This tension between the formal and the material seems to lie at
> the heart of the field of natural computing. The work of Michael
> Conrad emphasized the special role of biological material to
> explain the fantastic outcomes of evolution, as opposed to any
> power inhering in the class of relatively simple Darwinian
> algorithms. In the mutation-absorption model of the enzyme, we can
> begin to see how computational power emerges from the breakdown in
> the simulation relation, by the failure of commutativity (and, in
> the mathematical sense, the creation of torsion). This seems to be
> the vexing locus of this new field: clarifying precisely what
> happens when formal systems fail to track changes in fragments of
> nature.
>
> In this FIS session on molecular bionetworks, I think we may have
> the opportunity to find a clearer understanding here. Jerry
> Chandler's work deals with the connection between the formal (or
> symbolic) and the biological, and it seems these two perspectives
> may be mutually illuminating, and give us all the chance for some
> brainstorming in this largely unexplored area. Perhaps natural
> computing takes us back to the medieval meaning of natural as
> "sublunar" (to bring up term recirculated lately by Pedro
> Marijuan), dealing with the mutable substances down here under the
> orbit of the moon, illuminated by, but separated from, the
> immutable forms of the celestial world.
>
> One final introductory thought. In imputing computing to a fragment
> of nature -- a ribosome, say -- we should not view ourselves as
> reducing it to the apparently impoverished level of a Turing
> machine or a Pentium 4. Computer Science is in its infancy; the
> ribosome, the cell, the ecosystem: these are all better exemplars
> of computational wonder than our humble devices. If only we had a
> theory�
>
>
>
> ____________________________________
>
>
>
> _______________________________________________
> fis mailing list
> fis@listas.unizar.es
> http://webmail.unizar.es/mailman/listinfo/fis
>
Jerry LR Chandler
Research Professor
Krasnow Institute for Advanced Study
George Mason University
Received on Mon Nov 21 08:04:48 2005
