Customer Reviews

Information Theory, Evolution, and The Origin of Life

5 star:		(4)
4 star:		(3)
3 star:		(1)
2 star:		(0)
1 star:		(2)

 

 

The previous book of Hubert P. Yockey, 'Information theory and molecular biology', published in 1992, was unfortunately out of print for many years. Publication of 'Information theory, evolution, and the origin of life' makes available much of its material. The scope of the new book has moreover been broadened to encompass the hot topics mentioned in the title.

The academic world is divided into narrow compartments, each having its own methods, language, habits and gurus. Communication between them is made difficult by the lack of a common language so they most often ignore each other. When a concept from one of them eventually penetrates another one, it often assumes a superficial form which leads to misunderstandings. Although this may sometimes be better than plain ignorance, it results in rooting prejudices wrongly believed to hold true in other disciplines, and they live long for lack of proper internal criticism. Yockey is at the antipodes of this parochial system. His life-long efforts have been intended to convince biologists that information theory, a discipline originating in communication engineering, is the proper tool for dealing with molecular biology, hence should be at the heart of biology as a whole. He possesses to a high degree the needed didactic talents, as well as an extreme rigor in vocabulary and reasoning. Not only Yockey transcends disciplinary barriers, but also the famous divide between the 'two cultures'. His extremely broad scholarship is not purely scientific, but also historical, philosophical and literary. All chapters of the book bear in epigraph quotations from poets as well as from scientists or philosophers of all times, always wonderfully relevant to the subject matter. Similarly, many excellent quotations pepper the text. Yockey obviously does not think that scientific rigor demands dullness. On the contrary, the book is written in a witty and often caustic style. It abounds with historical anecdotes and comments, often intended to rehabilitate forgotten authors of major concepts or discoveries and to denounce usurped fames (one may disagree with some abrupt judgements).

As an engineer, I am convinced that information theory is the proper tool for dealing with molecular biology. Indeed, genomes communicate the genetic message as sequences of symbols (that these symbols are molecules does not mean they are relevant to chemistry only), and information theory is precisely the science of communication by means of symbol sequences. As a mathematical science, its results consist of theorems which can be thought of as predetermined forms which need 'only' to be filled with biological matter. But Shannon, the founding father of information theory, warned us that this task is far from trivial: 'Seldom do more than a few of nature's secrets give way at one time. [...] A thorough understanding of the mathematical foundations and its communication application is surely a prerequisite to other applications. I personally believe that many of the concepts of information theory will prove useful in these other fields [i.e., social sciences] but the establishing of such applications is not a trivial matter of translating words to a new domain, but rather the slow tedious process of hypothesis and experimental verification.' (from 'The Bandwagon', 1956, in 'Claude Elwood Shannon collected papers', edited by N.J.A. Sloane and A.D. Wyner, IEEE Press, 1993, page 462). Although Shannon made these comments about social sciences, I think that they perfectly apply to biology. Yockey's efforts are precisely aimed at letting biologists understand how relevant and potentially rewarding to their discipline is information theory.

Besides it presents compelling arguments in favor of the use of information theory in biology, the book also considers biological evolution and the origin of life. Since the genome is the medium which transmits the hereditary information through time, information theory is relevant to evolution just as it is to genetics. The book also critically reviews the many scientific and philosophical hypotheses about the origin of life, and shows that none of the alleged scenarios is likely to properly describe the events that actually occurred. The origin of life may well remain unknowable. Similarly, the mathematicians know since Gödel that propositions may be undecidable so, within a given system of axioms, it may be impossible to prove true results. The algorithmic information theory explains why it is so in a surprisingly simple manner: it results from the necessarily finite amount of available information. That the origin of life may remain a mystery forever is the bad news of the book. The good news is that information theory enables dealing with life phenomena by 'measuring, counting and weighting together with reasonings from postulates or axioms', a sentence which is quoted from Socrates at the very beginning of the book. Using information theory, biology can thus reach the status of a quantitative science. Biologists did not yet realize their luck, however. The book bears in epigraph a quotation from Niccolo Machiavelli who very lucidly states how difficult and dangerous it is to '[...] initiate a new order of things. For the reformer has ennemies in all those who profit by the old order, and only lukewarm defenders in all those who would profit by the new order'. Yockey's extraordinary book leaves by no means its reader lukewarm and is in itself a big step towards the new order. Although it is mainly intended to biologists, this book can be fruitful to the general reader who is interested in evolution and the origin of life. There are somewhat technical developments, in either information theory or molecular biology, but they are few and can be skipped. An appendix dealing with information theory, a glossary and an index are provided to help the reader.

Gerard Battail

Dr. Yockey is an extremely clear thinker, and has apparently been thinking about the connections between genetics and the mathematics of information theory for some time (1956 at least). This book, probably a difficult read for the layman, is nevertheless written in an entertaining and unbiased style. Although he slyly sneaks in references to the Bible ("...through a glass darkly...", "...stones that must be rejected by the builder...", etc.), he illuminates with equanimity both creationist and evolutionist theories with the cold light of mathematics. Ultimately, he concludes that life did not happen by chance, although he admits that he has no scenario to explain its origin. He speaks as a pure scientist and should be greatly respected for this.

This book has the very ambitious task of introducing the general reader to the current thinking regarding evolution, the origin of life on Earth, and the question of life on Mars, Europa and elsewhere in the universe.

Dr. Yockey shows that DNA is the genetic information system that compares in almost every aspect with digital data manipulation. DNA represents a code, a program if you will in computer terms that directs life. It also provides for the replication of life, and its evolution into changing forms over time.

The book is aimed at the non-specialist. It is not a text, but a kind of narrative history of significant developments in biology at a fundamental level. There is some mathematics in the book, but it is not a requirement that this be totally understood. The math serves as a proof of the statements he is making.

The book includes a chapter 'Does evolution need an intelligent designer?' This has caused some 'intelligent designers' to use Dr. Yockey's work in support of their argument.

Dr. Yockey concludes however, that there are some things that we just don't know and that: 'The fact that there are many things unavailable to human knowledge and reasoning, even in mathematics, does not mean that there must be an Intelligent Designer.'

This is a very enjoyable book to read. It is well written and clearly shows an intelligent approach to the problem.

I found this book to be an important and valuable resource while researching a possible Ph.D. thesis topic on interactions of DNA with enzyme pathways. Having previously studied about 95% of the math and spent a month as a Visiting Scholar in the most mathematical of the genetics labs at Harvard Medical School, I feel pretty confident that I can recommend the first half of the book to those seeking to build or broaden their professional knowledge of applied mathematics in the biological and biomedical sciences or in bioengineering. Despite its obvious importance to calculating the information content of proteins, protein folding, and cell-to-cell signalling, information theory is rarely covered in the standard biomathematics texts at all. § I think Cambridge University Press ought to ask Yockey to add text material on traditional subjects like Lottka-Volterra population studies, Turing diffusion models, Hopfield networks, and the like. Also, the book needs more exercises, so it would be easier to use for teaching. And wouldn't it be great if it were packaged in Mathematica or MatLab form! § I wish I could say something intelligent about the applications to molecular biology in the second half of the book, but I don't think I've gotten enough biochemistry and molecular genetics yet. One thing's for sure, though, it's written clearly enough that any molecular biologist familiar with the state of the art ought to be able to gauge its worth pretty quickly. Yockey's math is so good it's pretty hard to imagine he flopped on the science. § Maybe some of my own work will arrive in the 2nd edition. I can hope, can't I?

This book, which is the long awaited follow-up to Information Theory and Molecular Biology, is another tour de force in a long history of such insights from Dr. Yockey. As the former head of the U.S. Government's Aberdeen Proving Grounds, Yockey has a demonstrated history of squashing austensibly scientific ideas that superficially make sense, but when given the acid test are found entirely wanting. This book is replete with such deconstructions and they are much needed as they pertain to the current origin of life debate. Let me cite a few examples:

Perhaps formost among them is the idea that life arose from some Urschleim (primeval slime). Not only does Yockey show that this theory cannot be true, he explains exactly why, using mathematical certainty. First, he shows, applying Information theory to Crick's Central Dogma, that because the flow of information can only pass from larger encoding alphabets to smaller ones, but not the other way around, it is impossible for the information which fills the genetic code to have proceded from proteins (the smaller alphabet) to DNA/RNA (the larger alphabet). Ergo, it is equally impossible for any proteins-first theory of life origin to be correct - simply on that basis. Because what matters is not so much the DNA itself, in the scheme of life's continued existence, but the information it contains!

Next, he offers what may be the best summation of evidence in print to show that there simply is no scientific basis whatsoever to conclude that anything like Darwin's "warm little pond" ever existed. But he goes much further, taking evidence from fossil records as to the nature of the earth's atmosphere during the time the Urschleim was presumed to exist, Yockey shows that it is simply not possible chemically for earth to have had the atmosphere that it did and for those ponds to exist. The upshot being, according to cellular biologist and Nobel Laureate Christian Du Duve, without those ponds, the chance of any natural origin of life is zero.

Another strength of the book is the facility with which he ties the procedural activities of the genome to information theory, specifically Shannon's Law. The importance here is his insight into the nature of codes. He begins by demonstrating that the genetic code, in its present optimal form, could not have had a natural origin simply because not enough time has existed since the beginning of the universe to allow for it's actuality strictly in terms of processing.

He furthers this with the following quote from one of his earlier works: "The calculations presented in this paper show that the origin of a rather accurate genetic code, not necessarily the modern one, is a pons asinorum that must be crossed to pass over the abyss that separates crystallography, high polymer chemistry and physics from biology.(Yockey, 1981, 1992)" Then quoting from the book directly thereafter, "The paradox is seldom mentioned that enzymes are required to define or generate the reaction network, and the network is required to synthesize the enzymes and their component amino acids. There is no trace in physics or chemistry of the control of chemical reactions by a sequence of any sort or of a code between sequences. Thus, when we make the distinction between the origin of the genetic code and its evolution, we find the origin of the genetic code is unknowable."

However, Yockey is not arguing for some kind of theistic event. In fact, he takes great pains later in the book to demonstrate that he does not support any theistic conclusion. From his perspective, while it is provably true, based on mathematical certainty, that the genetic code did not have a natural origin, because the universe has demonstrated no ability whatsoever to formulate any kind of code, let alone something as sophisticated as the genome, it cannot be assumed ipso facto that a supernatural event is the only other choice. Because there is no scientific evidence to support that possibility, Yockey is completely unwilling to postulate such, even in off-the-record conversations.

To further distance himself from any hint that he supports Intelligent Design (ID) with is work, he takes-on one of the icons of ID, Dr. Michael Behe, and his theory of Irreducible complexity (IR). The way in which he attempts to show that Behe's theory does not work is to formulate IR as a kind of Gordian Knot that, if Behe is correct, is not computable. Because he can show that Behe's model is computable, he believes he has shown Behe's theory to be incorrect in principle.

However, his complete misunderstanding of Behe's theory leads him to disprove something Behe did not theorize. Behe's IR does not refer to a mathematically unsolvable puzzle, but to a kind of engineering dilemma for which there is no functional step-wise construction. Mechanisms for which there is no gradual, step by step approach to their completion, where every single step is itself a working model, are termed Irreducibly Complex. In other words, IR refers to any mechanism wherein all the parts necessary for its function are similtaneously extant because no partial iteration of the mechanism will function in any way.

I would use the example of a car engine. There is a net of engine parts required for the engine to run. Below that net assembly of parts, the engine will neither start nor run, even in principle. So while an engine is constructed sequentially, none of those sequences, short of a complete engine, will function, as is required by Darwinian gradualism.

Behe uses a simpler example, the mouse trap. His theory states that if you remove any one of the simple parts, it is impossible for the trap to function. The net result of Behe's theory is that IR makes it impossible for any mechanism so possessed to evolve in a gradual way because all the parts have to be there at the start for the mechanism to work. On the other hand, Darwinian Gradualism requires that every step be not only an advancement in function, but a competitive advantage that allows the creature superior ability in the war for continued existence.

Though Yockey confuses Behe's theory with the mathematical version of irreducibly complexity, to his credit, as the aforementioned quote from his book, regarding the impossibility of a network creating enzymes when enzymes themselves must first exist to make the network creating enzymes work [a classic Catch 22], he recognizes the irreducibly complex problem to which Behe refers. As such, while he discusses the it in completely different terms, his own example recognizes, as Behe theorizes, that it is impossible for such mechanisms to come into existence by some natural means.

That little flap is however, of no consequence in the panarama of Yockey's book. Everything he has written on the subject of this book has become a must read for anyone who wants to be completely up to speed on the origin of life question. His original insights are powerful precisely because he goes beyond supposition and hypotheses cum theories, to show with the certainty of mathematical law, why some things cannot be. As a consequence, whenever amathematical biologists finally decide to stop arguing about matters that have already been definitively determined, and consult the wisdom and insights of one of a physicist who is one of the 20th century's great scientific minds, they will devour this book.
John Tomlinson, MA, CHt

Readers should note that the two reviews below dated 1999 and 1998 are for Yockey's 1992 book, not this 2005 one. Once Amazon deletes those reviews, Amazon can delete this one as well.

Hubert Yockey has long studied life's programming from the perspective of information theory. His sceptical conclusions about origin-of-life theories are often cited by proponents of creationism / intelligent design (ID). But in his new book, Yockey is sceptical about some of their theories, too. For example, against Michael Behe he says that protein sequences cannot be irreducibly complex (p 179).

Regarding ID he comments that, according to information theory, "Once life has appeared,... genetic messages will not fade away and can indeed survive for 3.85 billion years without assistance from an Intelligent Designer" (p 181, 184). Okay, but the most interesting aspect of evolution is not the survival of old genetic programs, but the apparent invention of new ones. Does information theory explain how new genetic programs might be composed de novo? Can the process be observed or modeled? An informed discussion of this issue is sorely needed. Yockey's silence about it surprises and disappoints us.

As he ranges widely through the history of evolutionary theory, Yockey often wants to set the record straight. Specifically, several theories and experiments were known already, before the scientists who got credit for them came along. Furthermore, "Darwin did not believe in a 'warm little pond'..." (p 120), and "Oparin was very close to Lysenko" (p 153). If you are interested in information theory and biology, you will probably be edified by Yockey's scholarship in this book.

About 7 years ago I waded into the huge subject of evolution and origin of life. As an electrical engineer I was quickly appalled at the confusion, the endless opining, indulgent name-calling, and quantity of intellectual slop that passed for science in biology.

This was one of the first books I found that brought the level of rigor to biology that people in the hard sciences (Physics, Engineering, Computer Science) are accustomed to. If most discussions about biology seem "squishy" to you, that's because they are. This book is a refreshing departure.

Others here have complained that Yockey is dogmatic. He is, and he's earned the right to be. Most of what Yockey says is as right and as black and white as 1's and 0's. Claude Shannon's 1948 model for digital communication laid the foundation for our modern digital age and Shannon's work is flawless. In this book Yockey shows that the genome is *isomorphic* (structurally identical) with Shannon's model for digital communication.

Yockey says, "Information, transcription, translation, code, redundancy, synonymous, messenger, editing, and proofreading are all appropriate terms in biology. They take their meaning from information theory (Shannon, 1948) and are not synonyms, metaphors, or analogies."

This isomorphism is crucial to establishing any rigorous mathematical framework for modeling evolution in software. It's a required assumption for studying evolution as a computational process; for establishing a sound mathematical framework for the history of evolution (which still for the most part does not exist), and for clarifying the difference between assumptions and truth in evolutionary theories.

Yockey rightly points out that the laws of the genetic code are not derivable from physical laws and that even the simplest genome contains more information than all the laws of physics put together. This is why he says, "I have no doubt that if the historic process leading to the origin of life were knowable it would be a process of physics and chemistry. Thus the process of the origin of life is possible but unknowable."

For this reason he takes the position that the existence of the genetic code must be taken an axiom, a required but unprovable assumption. He shows the vast chasm exists between life and non life. Any attempt to argue otherwise muddles definitions and conflates the laws of physics with the freely chosen rules of codes. As a communications engineer, I understand that two could not be more different. Yockey puts it this way:

"The existence of a genome and the genetic code divides living organisms from nonliving matter. There is nothing in the physico-chemical world that remotely resembles reactions being determined by a sequence and codes between sequences."

The book does contain a few statements I disagree with. Crick's central dogma, which Yockey espouses, has since been shown to be wrong; cells do modify their own genomes based on inputs from the environment. The genome is a read/write storage system, not read-only. And evolution is not a random walk; it is driven by transposition, horizontal gene transfer, symbiogenesis, epigenetics and whole genome duplication, all of which are non-random. Authors like Lynn Margulis, Barbara McClintock and James Shapiro elaborate on these points in detail. However, the mechanisms of evolution are not the focus of this book so this is only a minor issue.

Engineers have very high expectations of theoretical models. As well they should; if they don't, bridges and buildings fall down and people die. Yockey brings the standards that are normally found in engineering to biology. Some biologists will be offended at this; that can only be expected when someone ushers in knowledge from outside their field that highlights the shortcomings of their assumptions.

In any case this book is a tremendous contribution to the field because it opens the door for a systems view of biology, in contrast to the traditional straitjacket of insisting that "it's all just complex physics and chemistry." It's not all just physics and chemistry because genomes literally and not figuratively process information. Living things are information driven and systems oriented. Little progress can be made without a solid foundation in digital communication. Yockey's work lays that foundation.

Just to give a taste of what Yockey's missing, I'd like to quote Stuart Kauffman from his "At Home in the Universe", "Life emerged, I suggest, not simple, but complex and whole, and has remained complex and whole ever since - not because of a mysterious elan vital, but thanks to the simple profound transformation of dead molecules into an organization by which each molecule's formation is catalyzed by some other molecule in the organization. The secret of life, the wellspring of reproduction, is not to be found in the beauty of Watson-Crick pairing, but in the achievement of collective catalytic closure. So, in another sense, life - complex, whole, emergent - is simple after all, a natural outgrowth of the world in which we live." (My only criticism here is in Kauffman calling the molecules dead--after all, the molecules themselves are complex and whole, being comprised as they are of relationships as well.)

And DNA is a lot more complex than Yockey gives credit to (which can probably be attributed to the fact that his information straight jacket can't account for its complexity.) From Martin Ingrouille's and Bill Eddie's "Plants: Diversity and Evolution", "One key feature of the nucleic acids like RNA and DNA is their ability to splice together; parts of the molecule can be looped out or into the sequence of bases. The parts of the sequence excised are called introns and those spliced together exons. Thus, in the evolution of life before the emergence of bacteria, we envisage an 'RNA world' where some molecules are active enzymes, others contain introns and exons and convert themselves, either to RNA by self-splicing, or recombine to yield novel combinations by trans-splicing. Subsequently DNA took the replication and information-storing role, and proteins the catalytic role, and RNA was left as an intermediary. In our 'DNA world' proteins have taken over almost every catalytic activity." This, to me, is a lovely example of how life specializes along lines of polarity.

Yockey, however, doesn't like hyper-cycles or anything that suggests metabolism, but his means of undermining its relevance are totally inane. For example, when he quotes Friedrich Engels, ["Life is the existence of protein bodies, the essential element of which consists in continual metabolic interchange with the natural environment outside them, and which ceases with the cessation of this metabolism, bringing about the decomposition of the protein. If success is ever attained in the preparation of protein bodies chemically, they will undoubtedly exhibit the phenomena of life and carry out the mechanisms of metabolism, however weak and short-lived they may be."]; it's only to glorify the failure of the Stanley Miller experiment, as well as to launch into his rant against dialectical materialism and his animosity towards Operin.

Fine, we may never create life in the test tube, but why can't he just acknowledge that our knowledge base was only in it's infancy during the 19th and 20th centuries rather than concluding, thereby, that it had actually met a dead end requiring that now we have to move on, full bore, to information science? I'll tell you why; since his forte is information, he needs to force everything into his information straight jacket. Had his purpose been the pursuit of truth, rather than the blowing of his own horn, this wouldn't have been the case--but then he wouldn't have had a book!

Information Theory, Evolution, and The Origin of Life

Hubert Yockey, in his book entitled "Information Theory, Evolution, and the Origin of Life", concludes that the Central Dogma proves that proteins can not have originated prior to the development of DNA/RNA. The Central Dogma originally stated by Crick was intended for the modern, studied living world to explain that information flows from DNA -> DNA, DNA -> RNA, RNA -> DNA, and RNA -> protein. This is based on the known mechanisms of transcription and reverse transcription etc. in modern forms of life that we have characterized to date. Yockey shows that because a codon includes three bases each with four possible types (ACGT) that there are 64 possible codes that represent the 20-22 possible amino acids in a sequence. The genetic code is redundant according to the argument so that it is impossible that DNA could arise from proteins. Yockey's ultimate conclusion is that DNA/RNA must have come before proteins and that the ultimate origin of life is unknowable. I find the argument naïve, most likely incorrect and essentially a DNA bias. If we can allow that the genetic code is redundant because of codons, we must acknowledge that in fact the genetic code is the product of complexes of proteins and are a consequence of these complexes. Each base is in fact metabolically constructed by sequences of proteins. It is entirely conceivable to construct new codes for unusual amino acids by altering the protein sequences, something that is being done today by biotechnology companies to generate new peptide based therapeutic drugs. So the information content of proteins is not just 20 amino acids but the trillions of proteins that can be generated through differing sequences which can produce unique catalytic reactions including generating new codes. In addition, the complexes of the proteins contain essential information, e.g., changing the sequence of metabolic reactions or the individual proteins. The protein information space is essentially unlimited and is much more redundant than the genetic code. It is true that the forms of life we characterize today utilize a process that is described by the Central Dogma but it is not true that this is necessarily the way it has always been especially during the origin of life. It has been shown by other scientists that the components of proteins, aminio acids and peptides are readily formed under the conditions of the early earth. On the other hand the bases, nucleic acids, are not formed in this way and are exceptionally unlikely to have existed before amino acids and peptides existed. I would turn Yockey's argument on its head and state that the protein space is so much more redundant that it surely originated prior to DNA/RNA.