Digital Genetics and Evolution Theory

[sciborg]

There is a new book on evolution theory: The Evolution of Aging (2nd ed), ISBN 0978870905, by Theodore Goldsmith that discusses digital information aspects of inheritance and their implications for evolution theory.

 

Darwin tells us that selective breeding and the corresponding evolutionary mechanism of natural selection both depend on the natural variation of characteristics between different individuals. Variation creates differences for selection to select.

 

According to the book, natural variation in more complex species is not actually "natural" in the sense of being a fundamental characteristic of all living things. Instead, because of the digital nature of the genetic code, the "natural" intrinsic situation is that members of a species should nominally tend toward being genetically identical. "Natural" variation in complex organisms is actually created and maintained primarily by the action of a long list of complex evolved mechanisms that process mutations including sexual reproduction, genetic recombination, certain behaviors, etc. The degree of variation produced by these mechanisms is described to be much greater than that produced by the occasional propagatable mutation.

 

This brings up the issue of how all these variation-producing traits evolved. It seems to be a somewhat circular situation: organisms are evolving the means for evolving.

 

Further, variation considered as an evolved design feature is itself incompatible with Darwinian evolution as generally understood. If organisms are striving to propagate their personal designs, then variation is adverse because it acts to reduce the ability of an organism to do that. A Darwinian organism would rather clone itself and therefore propagate ALL of its design characteristics than dilute its design via sexual reproduction and other evolved variation-producing characteristics. Cloning is the "natural", easier, route given digital genetics.

 

So how did these characteristics evolve? Wouldn't an organism that had the variation-producing characteristics be at a disadvantage relative to one that did not (such as one that reproduced by cloning) and therefore "select out?" An organism that happened to possess an advantageous design would certainly seem to be less able to propagate that design. Its descendents would likely be less able to survive, breed, etc. than would a clone. Somehow variation-producing characteristics were able to evolve despite fitness disadvantage, apparently because they convey an evolutionary advantage, an improvement in the capacity for evolution.

 

The question: Is this a plausible idea? If not, where is the fatal logical flaw?

 

If you accept these arguments, Goldsmith then goes on to suggest that aging, seen as a design feature that deliberately limits life span, also enhances the evolution process in several different ways. If variation can evolve, then aging can evolve even though both are fitness-adverse. This is counter to traditional theories of aging that say that aging is an adverse "side effect" linked to some beneficial and therefore evolvable trait such that there is a net Darwinian benefit. ("Beneficial" means a trait that helps in survival or reproduction as in "survival of the fittest.")

 

I realize these concepts might seem, at the least, radical, but they are nonetheless interesting.

[Kygron]

Well for the first part about having mechanisms for developing variation, I can say I was surprised... that it would even be questioned. Of course nature keeps us varied. Because... because... ok, I'm struggling with descriptions, it's just too common-sense for me.

 

As for aging, would you mind sharing an example of why it's beneficial? I can think of one or two, but they're probly not significant enough.

[Mokele]

According to the book, natural variation in more complex species is not actually "natural" in the sense of being a fundamental characteristic of all living things. Instead, because of the digital nature of the genetic code, the "natural" intrinsic situation is that members of a species should nominally tend toward being genetically identical. "Natural" variation in complex organisms is actually created and maintained primarily by the action of a long list of complex evolved mechanisms that process mutations including sexual reproduction, genetic recombination, certain behaviors, etc. The degree of variation produced by these mechanisms is described to be much greater than that produced by the occasional propagatable mutation.

 

Any actual basis for this?

 

Aside from the fact that it's flat-out wrong, I mean. We know how often species mutate, and far from being rare, they're surprisingly common. You, I and every other human has, on average, 5 mutations which affect final protein structure, and many more codon substitutions and the like.

 

Digital does not mean perfect replication, far from it.

 

Further, variation considered as an evolved design feature is itself incompatible with Darwinian evolution as generally understood. If organisms are striving to propagate their personal designs, then variation is adverse because it acts to reduce the ability of an organism to do that.

 

Why don't we build a tower into space out of pasta? Because the material doesn't support it. Same thing here. Mutation isn't a design feature, it's a flaw, a limitation of the system which cannot be circumvented.

 

A Darwinian organism would rather clone itself and therefore propagate ALL of its design characteristics than dilute its design via sexual reproduction and other evolved variation-producing characteristics. Cloning is the "natural", easier, route given digital genetics.

 

One word: parasites.

 

Parasites are usually smaller than the host, more numerous, and under greater selective pressures. As a result, an asexual parasite evolves faster than an asexual host, and will easily be able to counter any new defense the host comes up with.

 

But if two hosts mix their genes, the parasite must now cope with that mixture. Becoming good a dealing with one mix likely precludes becoming good at another, and the jack-of-all-trades is the master of none.

 

This is supported empirically by snails who can reproduce clonally or sexually. In populations with low parasite loads, they reproduce asexually, but when there are parasites around, they use sex. The incompleteness of the copy is outweighed by the copy's increased survival likelihood.

 

Mokele

[Edtharan]

Digital does not mean perfect replication, far from it.

Very true.

 

Digital is easier to correct errors than in Analogue, and in modern devices we include a lot of error correcting systems. But still we get errors.

 

In organisms, variation is an advantage, so there will not be a high pressure to evolve the high fidelity error correcting system like what we have in our digital devices.

[sciborg]

Well for the first part about having mechanisms for developing variation, I can say I was surprised... that it would even be questioned. Of course nature keeps us varied. Because... because... ok, I'm struggling with descriptions, it's just too common-sense for me.

 

I am more comfortable with scientific explanations than "common sense."

 

As for aging, would you mind sharing an example of why it's beneficial? I can think of one or two, but they're probly not significant enough.

The book discusses this in depth but here are some examples of benefits of aging or other evolved life span restriction:

 

- Aids evolution by eliminating older, less evolved individuals and freeing resources for younger more evolved individuals (1882)

- Aids evolution by preventing a few old individuals from dominating reproduction and reducing genetic diversity

- Allows evolution of intelligence and other features that involve acquisition of some property that is not genetically transmitted.

- Acts to control population growth and prevent extinction due to overpopulation.

There are several others.

[sciborg]

According to the book, natural variation in more complex species is not actually "natural" in the sense of being a fundamental characteristic of all living things. Instead, because of the digital nature of the genetic code, the "natural" intrinsic situation is that members of a species should nominally tend toward being genetically identical. "Natural" variation in complex organisms is actually created and maintained primarily by the action of a long list of complex evolved mechanisms that process mutations including sexual reproduction, genetic recombination, certain behaviors, etc. The degree of variation produced by these mechanisms is described to be much greater than that produced by the occasional propagatable mutation.

 

Any actual basis for this?

 

Aside from the fact that it's flat-out wrong, I mean. We know how often species mutate, and far from being rare, they're surprisingly common. You, I and every other human has, on average, 5 mutations which affect final protein structure, and many more codon substitutions and the like.

 

Digital does not mean perfect replication, far from it.[/Quote]

 

The book describes a lot of complex, apparently evolved, design features of organisms that act to create and maintain variation. The author doesn't claim digital is perfect, only that the innate nature of digital messages such as genetic codes does not support "structured" (useful) variation. The useful variation is created by all the evolved design features.

 

Why don't we build a tower into space out of pasta? Because the material doesn't support it. Same thing here. Mutation isn't a design feature, it's a flaw, a limitation of the system which cannot be circumvented.

 

I thought mutations were essential to the whole process of evolution. Without mutations we would all still be one-celled organisms. The "system" is built on mutations.

 

 

One word: parasites.

 

Parasites are usually smaller than the host, more numerous, and under greater selective pressures. As a result, an asexual parasite evolves faster than an asexual host, and will easily be able to counter any new defense the host comes up with.

 

But if two hosts mix their genes, the parasite must now cope with that mixture. Becoming good a dealing with one mix likely precludes becoming good at another, and the jack-of-all-trades is the master of none.

 

This is supported empirically by snails who can reproduce clonally or sexually. In populations with low parasite loads, they reproduce asexually, but when there are parasites around, they use sex. The incompleteness of the copy is outweighed by the copy's increased survival likelihood.

 

Mokele

 

Yes the "pathogen" explanation for sexual reproduction is popular among people that do not believe in "group selection". However, the book contains a long series of arguments to the effect that some form of group selection is the best overall explanation encompasing all the discrepancies with Darwins theory.

[sciborg]

Very true.

 

Digital is easier to correct errors than in Analogue, and in modern devices we include a lot of error correcting systems. But still we get errors.

 

In organisms, variation is an advantage, so there will not be a high pressure to evolve the high fidelity error correcting system like what we have in our digital devices.

 

Yes BUT. The book points out that in an analog system, errors (noise) create small variations just like the variations we see in organisms. In a digital system such as the genetic code, errors do not result in small variations but usually totally mess up the message. The variations that we see are actually created by a series of complex evolved design features that organize, and merge the digital data in the code to produce variations that LOOK like analog variation.

[Edtharan]

Yes BUT. The book points out that in an analog system, errors (noise) create small variations just like the variations we see in organisms. In a digital system such as the genetic code, errors do not result in small variations but usually totally mess up the message.

And it does. There are some error correction system in cells. Also there is a large amount of redundancy (one of the ways cells handle errors), so we can take small amounts of change without massive problems.

 

But sometimes even all this overwhelms an organism's ability to absorb or correct errors. We call these genetic diseases and not all organisms survive to reproduce.

 

Have a look at this: Human Fertility : Wikipedia

 

So, not every sexual encounter will result in an embryo, and even when it does, there is still a chance that it can miscarry. There is evidence that women can become pregnant and miscarry without ever knowing that they were pregnant.

 

A lot of the reasons that it can be so hard to become pregnant is that there are genetic errors. So the "message" can be completely messed up.

 

The redundancy and error checking really are the main reason that we can reproduce. Without these the "message" would be so mucked up that reproduction could not occur after just a few generations.

[Mokele]

The book describes a lot of complex, apparently evolved, design features of organisms that act to create and maintain variation.

 

Describe them. Since I'm far to busy to read the book, explain what it says.

 

The author doesn't claim digital is perfect, only that the innate nature of digital messages such as genetic codes does not support "structured" (useful) variation. The useful variation is created by all the evolved design features.

 

Sounds like the author has no familiarity with the scientific literature. We know what genes cause what variations in many cases, and it always comes back to a mutation in the code.

 

I thought mutations were essential to the whole process of evolution. Without mutations we would all still be one-celled organisms. The "system" is built on mutations.

 

Mutations don't exist because evolution needs them, evolution exists because mutations happen. The imperfect nature of biological data storage in DNA set the stage to allow life forms to evolve rather than remaining static.

 

Yes the "pathogen" explanation for sexual reproduction is popular among people that do not believe in "group selection". However, the book contains a long series of arguments to the effect that some form of group selection is the best overall explanation encompasing all the discrepancies with Darwins theory.

 

Does it mention that group selection was discarded more than 30 years ago because not only was there no data in support of it, but plenty of data that flatly contradicted it?

 

Group selection just plain does not work.

 

The book points out that in an analog system, errors (noise) create small variations just like the variations we see in organisms. In a digital system such as the genetic code, errors do not result in small variations but usually totally mess up the message. The variations that we see are actually created by a series of complex evolved design features that organize, and merge the digital data in the code to produce variations that LOOK like analog variation.

 

False analogy. The author evidently isn't familiar with even the basics of evolutionary biology. Imagine you have a gene controlling size, with two alleles, big and small, so you have two phenotypes, big and small. Now imagine you have two genes, both controlling size; the result would be 3 phenotypes, big, medium, and small. Now try 3 genes, which gives you 4 phenotypes. And add more and more: combined with natural environmental variation due to diet, by the time you hit about 8 genes, you have a result indistinguishable from gradual variation, even with a nice bell curve.

 

Furthermore, not all phenotypic variation is gradual. Look at albinos, for instance. Lots of traits, especially at the cellular level, are simply a case of you have it or you don't.

 

Aids evolution by eliminating older, less evolved individuals and freeing resources for younger more evolved individuals (1882)

 

And those genes would die out, because any animal that *cheats* and doesn't age will have more kids than those aging for a good of the species.

 

That's the key problem of group selection: someone can always cheat, and usually reap great rewards.

 

Aids evolution by preventing a few old individuals from dominating reproduction and reducing genetic diversity

 

How does that aid evolution? Old animals have proven they can survive, young animals haven't. If you're a female crocodile, whose sperm do you want, some youngling whose never even seen a human let along proved he can avoid them, or the century-old 25-foot 2000-lb male who's outwitted 40 years of poachers and even eaten a few?

 

Allows evolution of intelligence and other features that involve acquisition of some property that is not genetically transmitted.

 

This makes no sense. If an animal lives effectively indefinitely, wouldn't that *increase* the pool of cultural knowledge available to a species?

 

Acts to control population growth and prevent extinction due to overpopulation.

 

That's just freaking silly. Do you know how *rarely* an animal dies of old age? It only really happens in animals like elephants or tortoises whose defenses are effectively impregnable. 99.999999999999999% of all animals die before they even begin to get old, due to predators, disease, accident, fights with rivals, low food years, drought, or any of a huge number of reasons.

 

-----------------------------

 

I'll make this really simple: if mutation rate is controlled, picked somehow for 'optimum variation', then one would expect all organisms of a species/population to have the same mutation rate, yes? And that internal mechanisms would hold it constant?

 

However, what would you say if you found that, in animals whose body temperature is determined by the environment, the mutation rate was not only dependent upon temperature, but followed the same rate increase you would expect for any purely chemical reaction?

 

Guess which of these we've observed.

 

Mokele

[Rilx]

Yes BUT. The book points out that in an analog system, errors (noise) create small variations just like the variations we see in organisms. In a digital system such as the genetic code, errors do not result in small variations but usually totally mess up the message. The variations that we see are actually created by a series of complex evolved design features that organize, and merge the digital data in the code to produce variations that LOOK like analog variation.

 

Remember that the genetic code is nothing but a protein memory, not receipt of vital properties. Variation in genetic code causes building some slightly different proteins which in their final position might (or might not) cause small analog-like variations. No corrective design is needed because the operative effect is very small if even exist.

 

"Digital" is misleading; by the "digital" - say separate - nature of molecules and atoms all molecular chemical data is digital. All chemical formulas are "digital" and everything that is made of atoms and molecules is digital.

[sciborg]

I guess this is why the book is a book length document and not a paper. The book provides logical arguments that appear to counter all or most of the arguments made above. The book cites recent scientific papers of quite a few scientists that currently believe in group selection of one sort or another and explains why group selection is becoming more popular. (The selfish gene theory essentially proposes a form of group selection.) The chapter on the digital nature of inheritance and consequences for evolution theory has to be 30 pages and explains in detail why digital genetic messages are different from atoms and molecules. There is discussion of the cheater problem, why immortality would be bad for evolution even in species that seldom die of old age, phenotypic variation, etc. There is no way I can repeat all the arguments, evidence and citations here.