Evolution: A Complexity View | Hacker News Reader

Evolution: A Complexity View

rjlipton.wordpress.com

34 pointsnamlede14 comments

While it's true "the role of sex is not completely understood" in evolution, most current research points to co-evolving parasites as a major force that makes sex a good evolutionary move. It increases inter-generational variation while at the same time preserving viability (because both parent organisms are viable) which merely adding randomness would not do. Parasites are typically incredibly specialized because the hosts always evolve pretty good first-line defenses, so the additional genetic variability that sex produces from generation to generation becomes a considerable added benefit to the host, because the parasite population is always playing catch-up. Like many things in evolution, the war is never won, but sex tips the strategic balance in favour of the host.

That said, this kind of formal, mathematical approach to evolution is becoming increasingly powerful, and it's worth speculating if there might not be a provable theorem connecting random DNA variation and the laws of probability at the molecular level with the fact of macroscopic evolutionary change, speciation, etc... something like Boltzmann's H-theorem in thermodynamics, which attempts to show how micro-physics (kinetic theory) can be used to prove the macroscopic 2nd Law of Thermodynamics.

This would make evolution not a theory but a theorem, which I call "Darwin's Theorem". I've played around with the math a bit (I'm a physicist, not a mathematician) without getting anywhere, and wrote a what-if novel about the possibility (it plays with even more speculative ideas as well), in part with the thought of popularizing the idea, because I think it's worth people thinking about: http://www.amazon.com/Darwins-Theorem-TJ-Radcliffe-ebook/dp/...

While the fact that cross-over during meiosis does in fact contribute to increased genetic diversity leading to resistance to disease, parasites, and other adaptational advantages, one still needs to answer why did sexual reproduction evolve at all? Bacteria and protists are quite good at promoting genetic diversity through horizontal gene transfer, and viral infection with natural DNA replication error rates provide enough sources of "random" genetic variation to keep a diverse gene pool. One can think of competition to share genetic material as a primordial form of sexual selection. Getting from this kind of competition to orchestrated sharing and cross over of large genomes is a huge jump. The real question may be, did increasing organismal complexity necessitate a different type of reproduction to maintain genetic diversity? The answer to this may be yes, but alternative, less sexy hypotheses have to be tested as well. Another question to answer would be – What evolved first: multicellularity or sexual reproduction?

> enough sources of "random" genetic variation

well, that's the kind of thing the GP wants to put in numbers and a mathematical system, to provide substantial arguments.

This paper makes a 0th order assumption that there is an inherent advantage to sexual vs asexual reproduction. There are certainly advantages to both, but the author is correct that the answer may come out of addressing complexity. Algorithmically, there are many approaches to modelling sexual selection (rather than natural selection). Sexual selection occurs much faster than natural selection and may be one of the reasons why we have such apparent gender differences within species. Describing the effect of complexity in terms of "fitness" is one of the largest challenges faced by computer scientists and biologists this century. If you are interested in exploring some different algorithmic models, "Computational Molecular Evolution" is an excellent book (http://www.amazon.com/Computational-Molecular-Evolution-Oxfo...).

I'm confused by your distinction between natural selection and sexual selection. I always thought the latter was encompassed by the former. Are you talking about random phenomena like genetic drift, or that some complexities are not susceptible to adaptation? Did only browse the article, sorry.

Its generally understood that traits that encourage reproduction regardless of whether or not they encourage fitness will outperform other traits (https://en.wikipedia.org/wiki/Sexual_selection). This is why you will see the appearance of coloration and seemingly useless adaptations faster than actually beneficial traits within a population (i.e. a giraffe's neck; its not for reaching those high leaves). Natural selection is generally concerning traits that promote survivability over time, a much slower process. A major hypothesis is that sexual selection may be a driving force behind differentiation of the sexes, probably originating from genetic information sharing among single-celled organisms. A competing hypothesis is that there is an intrinsic advantage to sexual reproduction versus asexual reproduction. I generally favor the former hypothesis because it generally fits the "short-sightedness" principle of evolution.

I agree with you. I wrote an essay once about the need to be sentient about what characteristics of biological evolution one adopts in genetic programming, arguing that sex is arguably just a serendipitous side-effect or at least a sub-optimal solution to the problem of effectively mixing gene material. Much points to this, such as the short-sightedness principle of evolution, and that group or species selection argument is generally hard to make.

However, I am still reluctant to accept your distinction between natural selection and sexual selection. The article you link to even calls sexual selection "a mode of natural selection". The giraffe neck does not have fitness value in the sense of making it a better food processor, quite the opposite. But it does have fitness value in the sense of prospering those very genes, indeed otherwise the giraffe would not exist. Such a definition of fitness does tend to be tautological, but the alternative is no better, because it is inevitably based on a measure of survivability that in the end is necessarily arbitrary.

The basic distinction is this: Natural selection refers to one organism having superior reproduction prospects due to inherent physical factors (strength, speed, height, weight, ability to digest specific foods, ability to detect and respond appropriately to threats, etc.). Natural selection takes place without the awareness of the hosts involved: By the time their relative advantages or disadvantages are in evidence, they are long dead.

In other words, natural selection is about fit, that is, how well an organism "fits" or is suited to and environment. (That it what Darwin meant by fitness: Environmental suitability.)

Sexual selection refers to overt preferences in mating: A female bower bird chooses the male that builds the "best" bower (a completely useless artifice neither ever use), a human male prefers women with larger breasts (pretty much ditto), a peahen prefers the peacock with the more extravagant tail (ditto for sure, in fact, more extravagant tails are liabilities).

Examples are many. The term "runaway sexual selection" is used to refer to exaggerated traits that become prevalent and possibly dominant in specific, local populations (think of human characteristics you associate with certain groups: Sometime, somewhere, someone preferred those slightly larger eyes or slightly curvier hips and over time those became prevalent). These traits are generally completely irrelevant, with zero fitness value.

They exist because we think they look/smell good. Why did we come to prefer one over another? The same random variation that causes changes to fitness causes changes to fitness indicators, but those indicators are not necessarily honest signals.

I am familiar with the gene-centered view of evolution proposed by Williams and popularized by Dawkins with the so-called selfish gene theory. Within this school of thought sexual selection by no means stands in contrast to natural selection and is not labeled an alternative to it. Instead, in fact, sexual selection is argued to be a direct consequence of natural selection as a phenomena you would expect with Darwinisitic reasoning when assuming the gene-centered view. As is made the case, the gene-centered view is the extension of Darwin's original argument, that is, Darwin's argument taken to its logical conclusion with the modern facts at hand, the DNA molecule not the least.

One of the central themes in Dawkins's The Extended Phenotype is the putting into question of the validity of traditional role of the organism in Darwinistic reasoning and why Darwin's original definition of fitness does not hold up. There is actually an entire segment on the many modern (back then) definitions of fitness and problems with each of them. I don't recall what the final conclusion is but it is certain it is not the clear-cut case you make it.

I suppose you have background in a field where your definition of fitness is more firmly established. For instance, I know biological mathematics has come a long way, and I'm not up to speed with what definitions they are using.

Sexual selection is a huge area of research in evolutionary biology. You don't have to like the wikipedia article, but a quick search on google scholar on the term reveals over 2M+ results. Check it out. Peter's description of sexual selection was much better than mine, so thanks! I brought up the Giraffe's neck because its one of the best examples of bad scientific method leading to incorrect evolutionary conclusions. Giraffe males joust as a dominance ritual prior to mating, and generally the giraffe that can exert more force will win and hence mate. There are no advantages to the neck outside of this ritual as far as anyone has been able to determine, hence it is understood to be a purely sexually selected trait.

I'm not sure how you interpreted me as a denier or opponent of sexual selection theory. I was pointing out that that very Wikipedia article you refer to clearly states that sexual selection is a mode of natural selection, which is the case I'm making.

I wrote a lengthier answer to Peter Whittaker. Check it out.

There is an algorithmic technique underlying a lot of this stuff called the "Multiplicative Weights Update Algorithm." It's worth checking out if you want a cool algorithm that applies in lots of different scenarios.

This reminds me of Chaitin's "meta-biology", which uses (self-delimiting) computer programs as genomes, a Turing machine as the environment and "fitness" is defined as how big the program's output is, interpreted as a natural number. Unfortunately it relies on a halting oracle to avoid diverging programs :(

Here's a copy of Chaitin's metabiology draft notes compiled from a series of lectures he's given over the years: https://www.cs.auckland.ac.nz/~chaitin/metabiology.pdf

Reply on news.ycombinator.com