Steve LaBonne and Mitochondrial genetic codes

HPYockey hpyockey at aol.com
Fri Aug 25 10:18:43 EST 1995


Subject: Steve LaBonne's ideology about the Evolution of the genetic code 

This comment came from: labonnes at csc.albany.edu (S. LaBonne)
Date: 9 Aug 1995 13:59:00 GMT
Message-ID: <40aev4$ffs at rebecca.albany.edu>

Lurkers may get a list of the pertinent references in this post by e-mail
HPYockey at aol.com  I am interested in how many are reading LaBonne's
comments. Perhaps you could ask him for a list of all his publications. 

In article <3vrslq$f54 at charm.magnus.acs.ohio-state.edu>,
Brian D Harper <bharper at magnus.acs.ohio-state.edu> wrote:
>
>     Of  course there are controversial conclusions in the book.
>     If there weren't there would have been no point in publishing it.
>     I think the most important contribution to practical molecular
>     biology is in Chapter 7 on the evolution of the genetic code.
>     I began with the proposal that the original genetic code was a
>     doublet code in which the third nucleotide was silent and which
>     coded for only eight or so amino acids. As the code evolved by
>     enlarging its vocabulary coding theory shows that the number of
>     possible codes decreases and finally must pass through a bottleneck
>     when there are 14 or 15 amino acids in the vocabulary. The
>     vocabulary could then be enlarged to 20 only by applying specificty
>     to the third nucleotide. By analyzing the evolution of such codes
>     by means of coding theory I showed that the existence of
>     mitochondrial and other codes that differ in a few assignments
>     from the standard genetic code is a required consequence of that
>     evolution.
>
>     The question of the separate genetic codes in the mitochondria is
>     usually sloughed off as a trivial curiosity. On the contrary, I
>     suggest that Nature is trying to tell us something if we will only
>     listen. This discussion is a qualitative review of what I have
>     dealt with in full industrial strength in Information and Molecular
>     Biology.
>     -- Hubert Yockey, bionet.info-theory, Jan. 1995
>
>I would be interested in what molecular biology types would have to
>say about this. As a layman it seems to me that this provides very
>good evidence for evolution and also classifies as an entry in the
>"evolutionary predictions" list, i.e with the prediction of the
>mitochondrial code arising directly as a consequence of the evolution
>of the standard code. Best of all, there's no "just so" story
>involved :).

Brian Harper asks for molecular biology types. Guess who responds: Nobody
but Steve LaBonne. "Glad to oblige: it's nonsense.  Non-standard codes,
pace Yockey, have
arisen secondarily in organelles and few protists.  None are found in
organisms that are not right at the tip of evolutionary lineages.
Therefore, Yockey's ideas are not credible, because they would require
that _in all other lineages_, the variant codes magically converged on
the "universal" code.  This is, to say the least, HIGHLY
un-parsimonious.  The evidence tells us clearly that the "universal"
code is the one that was posessed by the last common ancestor of
living things, and that deviations- most of which consist merely of
reassigning a stop codon to an amino acid, a fiarly trivial change-
arose later.  Sorry if this disappoints you. ;-)

Steve: Just a note on English grammar: The word "none", a contraction of
"no one" is singular,  so one says "none is". Glad to help in this regard.


As usual Steve LaBonne presents himself as an expert without quoting any
of his publications. 
In the first place 'phylogenetic evidence' is hardly written in letters of
gold on tablets of jade. Notice the African Eve controversy. Not being
able to follow phylogeny back a few million years puts great doubt about
following it back 3.8 billion years to the Isua formation. The authors who
contributed to that ran their prefered program tens of thousands of times
and took a vote. I suggest that is not very good science.

For my comment on this point see Section 6. The Mutual Entropy of
Homologous Protein Families.

Any knowledge of coding theory or practice shows that a change in the
sense assignment of a letter in the sending alphabet without a
corresponding change in the receiving alphabet is effectively noise. It is
worse than random noise because it makes an error whenever that letter of
the sending alphabet appears. 
For example on page 190-191 of the subject reference one reads:
"It is important to note that this scenario does not require that even a
substantial fraction of the available codes be tested and therefore Jukes'
suggestion passes the third test. The evolutionary process follows an
ascending path, so to speak, to the doublet code saturation bottleneck at
14 to 15 sense code words
This process is irreversible, since all backward steps increase the
vulnerability to mutations to non-sense codons. For the same reason the
paradigm assumes that once an assignment is made it does not change."


 Steve you refer to the argument in Chapter 7 as "Yockey's ideas". If you
had read Chapter 7 you would have found in section 7.2.3 that I cited Tom
Jukes as the author of the idea. I also cited Labouygues, Figereau and
Cullman as the authors of the argument in that section and of the material
in Table 7.1. I showed how these proposals should be treated according to
information theory and coding theory and the results to be obtained. 

You need to pay more respect to your betters!

Chapter 7 applies to a critter that appeared earlier than the urgenote,
whatever that may mean. Chapter 7 concludes that were several of these and
there may have been more than one origin of life event, as Darwin
suggested in the last chapter of the Origin of Species. The standard story
is that several of these early critters got eaten and became mitochodria.
They took their own genetic code with them. The critters with the standard
code appeared and after their meal of simpler critters they continued to
evolve and became what you call the urgenote.  Perhaps their remains
formed the kerogen of the Isua formation. Presumably they later became
Eukaryota, Eubacteria, and Archea. Their ancestors continue to contain
mitochondria with their separate genetic system today. 

I discussed the evidence that the finger prints of a doublet genetic code
are on the standard code. The phylogenetic story begins well after the
emergence from the bottleneck. 

Section 7.3 successfully compares the  results to the codon assignments
found in mitochondria and simple organisms. 

Steve: If you go to the library you will find (with the help of the
librarian) a book called Molecular Biology of the Gene by Watson, Hopkins,
Roberts, Steitz and Weiner  Published by  Benjamin Cummings. Read Chapter
15 The Genetic Code. 
You will find the answers to your questions in that chapter. 


As Ludwig Wittgenstein was oft heard to say: "Wovon man nicht sprechen
kann, darueber man muss schweigen. Steve: "Schweigen sie bitte bis sie das
Buch gelesen und verstanden haben." If you were multicultural you could
read that!

Aufwiedersehen  Hubert P. Yockey  



-- 
Steve LaBonne ******************* (labonnes at cnsunix.albany.edu)
"It can never be satisfied, the mind, never." - Wallace Stevens


*Third, about the genetic code.  I would stand firm in my position --
the phylogenetic evidence strongly implies that the last common
ancestor (urgenote) of all known living organisms was a sophisticated
beast
with (among other things)
 a DNA genome
 the canonical genetic code
 many gene familes (we'll get a good estimate next year when >=1
  complete genome from Eukaryota, Eubacteria, and Archea becomes
  available).
Hence, the genetic code expansion process you propose may have taken
place,
but it did so prior to the emergence of the urgenote.  Alas, the
urgenote is much like the singularity in the Big Bang, and very little
can be determined of history prior to it.  Therefore, your hypothesis
(like all too many early-evolution hypotheses) is untestable.*

In the case of the Big Bang the concept "history prior to it" has no
meaning. The coordinates of space and time are so mixed up by the intense
gravity that they cannot be distinguished. No signal can be sent or
received earlier than the singularity or even from it. A physical concept
that cannot be measured has no meaning. Concepts that cannot be measured
are left to philosophers and theologians. On the other hand there is a
history of the earth before the appearance of life. 

In the first place 'phylogenetic evidence' is hardly written in letters of
gold on tablets of jade. Notice the African Eve controversy. Not being
able to follow phylogeny back a few million years puts great doubt about
following it back 3.8 billion years to the Isua formation. The authors who
contributed to that ran their prefered program tens of thousands of times
and took a vote. I suggest that is not very good science.

For my comment on this point see Section 6. The Mutual Entropy of
Homologous Protein Families.

 You refer to the argument in Chapter 7 as "your argument". In section
7.2.3 I cited Tom Jukes as the author of the idea. I also cited
Labouygues, Figereau and Cullman as the authors of the argument in that
section and of the material in Table 7.1. I showed how these proposals
should be treated according to information theory and coding theory and
the results to be obtained. 

Chapter 7 applies to a critter that appeared earlier than the urgenote,
whatever that may mean. Chapter 7 concludes that were several of these and
there may have been more than one origin of life event, as Darwin
suggested in the last chapter of the Origin of Species. The standard story
is that several of these early critters got eaten and became mitochodria.
They took their own genetic code with them. The critters with the standard
code appeared and after their meal of simpler critters they continued to
evolve and became what you call the urgenote.  Perhaps their remains
formed the kerogen of the Isua formation. Presumably they became
Eukaryota, Eubacteria, and Archea. Their ancestors continue to contain
mitochondria with their separate genetic system today. 

I discussed the evidence that the finger prints of a doublet genetic code
are on the standard code. The phylogenetic story begins well after the
emergence from the bottleneck. 


*Fourth, the section of your book on abiogenesis appears to succumb
to a logical flaw which is an inherent danger in the information theory
approach (this is not a condemnation of this approach, only a pitfall
which must be avoided).   Because you do not explicitly address this
problem, it must be assumed that you have not considered it.

The problem is this:  the IT approach attempts to determine the boundaries
on a given protein, such as cytochrome c.  By looking at extant proteins,
we can determine which amino acids are allowed at certain positions and
which are not.  In your book, you then extend this to predict the
probability of cytochrome c appearing in a random peptide & propose
this as a serious problem for abiogenesis theories.

However, this analysis ignores a biochemical subtlety.  While some
substitutions are absolutely lethal to the function of a protein,
many only cause a significant, but not complete, loss of function.
In a modern organism, virtually any loss in cytochrome c function would
not be tolerated, due to the current energy needs of modern organisms.
But it does not necessarily follow that such substitutions (please
ignore the temporaral directionality connotation of that word) would
not be tolerated in an ancient genome.  I.e., it is quite possible
that the pre-urgenote organisms has a lower metabolic budget, and
hence could survive with a much less efficient cytochrome c (especially
since all their competitors had equally inefficient cyt-c).*

I thought I dealt with that in section 6.1.2. The question of modern
cytochrome c is mentioned in 6.3.1 "Although they are functionally
identical in the electron transfer complex, antibodies have been found in
rabbits inoculated with a wide variety of cytochorme c..."

See above: What is a random peptide?

*In summary, I am proposing that your IT measure is measuring 
modern cytochrome c's, whereas the abiogenesis question is
concerned with anything with cytochrome c-like functionality.
Now again, because we are thinking about pre-urgenote history,
there is no evolutionary evidence to test this.  However, there
is plenty of reason to think this scenario plausible.  First,
the phenomenon of weak activity is well known -- many enzymes
will act a little on substrates other than their intended one.*

PLAUSIBILITY is no criterion for scientific theories. Remember that we are
not the judges of the world. The wave-particle duality of quantum
mechanics is not PLAUSIBLE. It is a total violation of common sense. It
means that a neutron or electron diffracting in a crystal lattice is in
two or more places at once. Yet if we try to catch it in the act we
cannot. This was the subject of a long contest between Einstein and Bohr.
Enstein hated the wave-partical duality to the end of his life.
Experiments done recently support quantum mechanics, like it or not.
Quantum mechanics describes this phenomenon but does not explain it.
Nature is not organized according to university departments. For a poetic
or theological reference Job 38- 41. 
See my remarks below about Sir Karl Popper and the "testibility"
requirement. 

*Second, in experiments by Szostak and others to evolve catalytic
RNAs, a common phenomenon is to observe a peaking of the catalytic
ability of the original pool after several rounds of non-mutagenic
propagation+selection.  Mutagenizing a pool after it has reached
such a state frequently results in the emergence of much better
catalysts.  In other words, many weakly-active catalysts can be
found in the original pool and converted (by mutation) into
strong catalysts, but the odds of finding a strong-catalyst
are poor in random sequence.*

#See above. What is a random sequence? The Szostak et al. paper in Science
269 July 1995 is very interesting. I notice he says on page 368 last
paragraph second column: "We started with only 1.4 x 10^15 of the 10^132
possible N220 sequences. He doesn't know about the Shannon- McMillan
Theorem cited frequently in my book. There are far fewer significant
sequences than the "total possible". I first applied the Shannon-McMillan
theorem to molecular biology in JTB 377-398 (1977). Using the
Shannon-McMillan theorem would support  his thesis not oppose it. 

A further question is catalysis of what? Catalysis is common throughout
both organic and inorganic chemistry. 

*There is, of course, also the phenomenon of "roads-not-taken"
-- canyons on the fitness landscape which require multiple
mutations to cross.  I am in particular thinking of a lecture
I heard Jeremy Knowles give once, in which he described mutagenesis
of triose phosphate isomerase.  Using the crystal structures as a
guide, a substitution was made which shortened one of the key
catalytic residues by 1 carbon (Glu-->Asp).  As a result, the
carboxyl group was too far from the other catalytic sidechains
to function effectively (I think the mutant may have had no
activity, but it could have had a little).  After one round of
random mutagenesis and selection, mutants were found with significant
increases in activity, though not as good as the natural enzyme.
A second round of random mutagenesis (using the first round candidates
as targets) brought the activity almost up to the level of the
original enzyme.  Alas, they stopped there.

Both of these arguments suggest that the observed information content 
of a protein family may overestimate the information content 
of that family, and therefore one must use caution in using IT to predict
the
probabilities of the initial formation of these gene families.*
See discussion on page 328 of Maynard Smith's word game proposal.  

 
What is the role of theory in science?
J. Robert Oppenheimer often said that theorists experiment with ideas as
experimentalists experiment with apparatus in the laboratory. Ideas do not
always work out; neither do all experiments in the laboratory. I followed
the standard procedure in the two cases above, that is, to establish a
proposition and then apply a mathematical theory that pertains to that
problem. For example, if we are dealing with a mechanical system one
applies Newton's laws of mechanics. In that case one must establish that
masses are small compared to the Earth and velocities are small compared
to that of light. If these conditions do not pertain one must use General
relativity. If one is dealing with electricity and magnetism one uses
Maxwell's equations. Again, there are certain conditions that must apply
for Maxwell's equations without going to quantum theory. The point is that
these theories may not be true in a sense that would satisfy 
philosophers. However, for the scientist they are true in the sense that
they are useful in treating a wide variety of problems. In Chapter 7 I
took the proposal of Jukes and the ideas of Labouygues, Figereau and
Cullman  and applied coding theory. These results are important to the
extent that they are useful.

The  primary purpose is to demonstrate how to do these calculations.
The same plan or procedure was followed in Chapter 6. In Tables 6.3, 6.4
and others I made two lists. One list with only the known residues and one
with the residues predicted by Table 6.1.  Of course, if there are too few
functionally equivalent amino acids used in the calculation the
information content will be too large. Any reader is free to use any means
of prediction of what he considers functionally equivalent amino acids and
calculate the result. I did not mean to imply that it is  necessary to
take modern proteins as a guide.

The argument in Chapter 7 is untestable in the sense that it can't be
repeated in a test tube. We are often confronted with that situation. We
cannot repeat the history of geology. We must read the story in the rocks.
In many cases we can find rather exact ages for rocks. Any speculation
about the history of geology must be compatible with those ages. We cannot
repeat the history of the evolution of life.  Creationists frequently say
that evolution is untestable and therefor a pure speculation.

Sir Karl Popper gave the world the "testability" notion applying to
scientific theories. This is fine for philosphers who, like the spectators
at a football game, sit on the sidelines and cheer. It is very different
for those of us who are players on the field. In fact all scientific
theories fail "testablity". Astronomers and nuclear physicists believe
that the energy in the sun and other stars comes from nuclear reactions,
that is, the burning of hydrogen, helium and other atoms to produce the
elements in the periodic table. However, there are only 2/3 of the
neutrinos that should be produced detected on earth. See comment on page
221. Strictly, Sir Karl would say that falsifies the theory. For further
discussion see Chapter 8.

All great theories in science had what seemed at their inception to be
fatal flaws. Newton had to invent calculus  to calculate the orbits
according to his inverse square law of force.
Newton's mechanics and theory of gravity required a force acting
instantaneously through the vacuum of empty space. As the planets move the
force instantly assumes a new value.  Leibnitz and others had
philosophical objections, especially the belief that "action at a
distance" through empty space was impossible. Although this bothered
Newton considerably he nevertheless went ahead with his equation for the
force of gravity.
 If Sir Isaac Newton had had today's accurate planetary orbits that
clearly deviate from exact ellipses he might have thought that data
falsified his theory of gravity. However, deviations from exact ellipses
are accounted for by the gravitational attraction between the planets.
This supports rather than falsifies Newton's theory of gravitation and led
to the discovery of Neptune from these perturbations.
 There were mathematical difficulties.  There is no closed solution to the
third body problem as there is to the two body problem. Since there were
six planets known at the time it could be argued that Newton's theory was
of no practical value. Leibnitz objected strenously to these difficulties.
There is an interesting discussion by Sir John Maddox in Nature 376 p385
(1995). See also the book review of Newton's Principia for the common
reader by Chandrasekhar. Review by David Hughes same issue of Nature page
395.
 
On the other hand, sometimes small deviations are real and their
investigation leads to important discoveries. For example, in the late
19th century astronomers found they could not account for the advance of
perihelion of Mercury by perturbations from Venus and Earth. The correct
calculation of the advance of perihelion of Mercury was the first test
Einstein made of his General Theory of Relativity.

I suggest that the genetic codes that deviate from the standard genetic
codes should not be sloughed off as a minor problem. Nature is trying to
tell us something.


Darwin's theory of evolution required a much older earth than appeared to
be possible at the time. Species seemed to last a very long time and there
were no intermediates.  British geologist Sir Charles. 1797-1875. [His
Principles of Geology (1830-1833) opposed the catastrophic theory of
geologic change.] believed that the earth was several billion years old
but Lord Kelvin did not. Darwin believed that natura non facit saltum yet
he could not explain "sports". The blending or 'paint pot' theory showed
conclusively that his small variations on which the theory depended would
wash out in one or two generations. 

Mendel's laws were criticized because sometimes it appeared that
inheritance did blend.  


The blending or 'paint pot' theory is referred to by Sir Ronald Fisher in
"The Mathematical of Natural Selection" where he shows that the blending
theory is wrong and mutation is correct. Darwin never saw Mendel's paper.
He did not read German very well and was completely lost even in simple
mathematics. Mendel expressed his results in the form of algebra and
ratios. I believe that this is the reason that Mendel's work was ignored
for 35 years. Remember that the Austrian botanist von Naegli was familiar
with Mendel's experiments but did not understand them. Ernst Mayr "The
Growth of Biological Thought" page 723 points out the von Naegli believed
in the blending theory in spite of the fact that he knew of Mendel's work.


I hope you will compare my treatment with that of Eigen. His "error
catastrophe" is often quoted by molecular biologists. It is not only wrong
but Shannon gave the correct treatment in 1948. I discussed this in detail
in Chapter 10.

I hope my book will help those who wish to apply information theory to
specific problems in molecular biology. A number of papers of varying
quality have been published. A particularly egregious example is Mantegna
et al Physical Review Letters 
v73 3169-3172 (1994). This article has received considerble publicity.
Konopka and Martindale in Science 268 p789 (1995) challenged the first
theme of the paper, namely, that according to Zipf's law (sic) "noncoding
regions of DNA may carry biological information". I have sent a Comment
showing that their use of coding theory and information theory is
incorrect. 

I end this e-mail with the following quotation from one of the saints of
biology. 
Sir Ronald ends the preface of "The Mathematical of Natural Selection"
with the following sentence: "I believe that no one will be surprised that
a large number of the points considered demand a far fuller, more
rigorous, and more comprehensive treatment. It seems impossible that full
justice should be done to the subject in this way, until there is built up
a tradition of mathematical work devoted to biological problems,
comparable to the researches upon which a mathematical physicist can draw
in the resolution of special difficulties." 

Henry Quastler published Information Theory in Biology in 1953. It is now
42 years since that time. I hope my book will help those who wish to apply
 information theory to specific problms in molecular biology and that
information theory will play an important role in the future of molecular
biology. 

I appreciate your interest in my book and I look forward to more comment. 
For Harvard to have purchased a copy is very flattering. 

Yours sincerely, Hubert P. Yockey





More information about the Mol-evol mailing list