Evolutionary momentum

Hi Xeno.I'm really unclear what you're asking overall, but I would like to address this:

quote:

---

The second face of the question is this: if such cumulative mutations can exist, is it possible that there will be a natural tendency for the phenotype to accumulate? (but independent of the selective advantages)

---

I really don't know what you mean by "natural tendency". However, it is a well-established phenomenon that phenotypical changes in one area will often "force" or cause phenotypical changes in another. The simplest example I can think of is that you'll never see back legs increasing in length without a corresponding increase in length of the front legs. It's called frequency-dependent selection, and it can be observed at the phenotype, genotype, and molecular level.In addition, you can have non-adaptive selection. Sexual selection and Mullers Ratchet come to mind (c.f., peacock's tail).If you could explain a bit more what you're after, I may be able to come up with some specific references for you.

XJ:I think I'm beginning to understand what you're trying to get at here. Are you asking whether consecutive mutations at a "mutational hotspot" that might have a neutral phenotypical effect can accumulate phenotypical change over time?If that's the case, then I would say that theoretically there would be no reason to preclude this from occurring. I can't think of any actual example where this might have occurred, however. I would also hasten to add that cumulative change in a trait - whether from mutation at a hotspot or not - will at some point come under selective pressure even if it starts out neutral. Your bird wing example threw me - increase or decrease in the area of a bird's wing will have an adaptive effect (either positive or negative).I also have a problem with this statement:

quote:

---

If there is a mutative basin, then the whole point is that mutations do not occur at random. It would be a tantalizing coincidence if such basins coincided with useful phenotypical progressions!!

---

I thought you were equating your "mutative basin" concept with mutational hotspot. There are three problems here:1. Mutational hotspots are only non-random in a statistical sense. That is, there is an increased probability that the particular nucleotide site will undergo mutation because of various factors. It is still random in the sense that you cannot predict when or even if a mutation will occur at that site except stochastically over many generations. In other words, IF a specific mutagen is introduced somewhere in the genome, there is a slightly higher probability that it will effect the hotspot site than some other location. It doesn't mean it will.2. Once you start bringing in "useful" in relation to a phenotypical trait, natural selection comes into the game, because "useful" is only possible to be understood in the context of the environment. "Useful" indicates that there has been an increase in marginal fitness. This holds true whether you are dealing with the genome or organism level.3. "Tantalizing coincidence" seems to be a bit loaded. It would certainly be a coincidence if a randomly-induced neutral phenotypical trait/change accidently produced an increase in fitness (in which case it would no longer be neutral) OR if an environmental change caused the neutral trait that has been slowly accumulating to suddenly have a positive (or negative) fitness effect. Still, the only part of this that is non-random is the selection pressure when/if it arises.I hope this addresses your question.[This message has been edited by Quetzal, 12-19-2002]

Hi XJ:Mammuthus pretty much covered what I would have said in response to your next-to-last post, so I'll take a stab at this one.

quote:

---

Ah - I didn't realize this. However, is it feasible to postulate that what we might consider non-coding sites, are actually crucial in some ways? What we might consider "junk" DNA might actually be an integral part of a bigger equation? Going back to the shape of a nose - the fact that we do not understand the epigenesis of the shape of a nose seems to indicate that we have not cracked the code of DNA. As far as I understand, what we do know about gene expression is simply the manufacturing and regulation of protein production - perhaps there are deeper layers of information hidden inside the structure of DNA, and these apparent junk spots are actually parts of this information structure - pure speculation here, but would like to hear other's thoughts...

---

"Junk DNA" is a rather unfortunate term. "Non-coding" is probably a better description. There are a lot of bits of so-called junk DNA (not all, by any stretch) that have a genome-level function, from structural to chromosome silencing. Some bits don't seem to have a function currently (like HERV retroviral insertions), but may end up being incorporated into an exon via mutation. We haven't figured out what everything does, yet, and some of it may in fact be "junk" (like the repetitious Alu sequences).

quote:

---

So this genetic drift would also be a random process, but is it not interesting that such a drift would be expressed as a gradual phenotypical change? The (non-hotspot) mutations that facilitated this drift may indeed be random but the fact that they resulted in a gradual change in a given phenotype may indicate some deeper mathematical pattern - maybe something to do with the homology of the organism.

---

Hmm, genetic drift is a population-level phenomenon relating to a random walk in variation of alleles simply due to chance. It's mostly observed in small, isolated populations. Certain variations may become fixed or even disappear completely. I suppose it's possible that this could lead to gradual phenotypical change - it's been seen to contribute to speciation, for example. However, that aspect is a very rare occurrance, and is often overwhelmed by other factors. In addition, there's no reason that this gradual drift can't reverse itself when alleles that were starting to predominate simply vanished. I'm afraid drift won't help you much.

quote:

---

Yes exactly it would be hitchhiking - I guess my biggest question is how to explain gradual phenotypical changes - it doesn't jive well with my instinct to think of it as purely random - random in the sense that the mutations may occur at random (stochastically as you said) yes - but the fact that these random changes give rise to patterned growth may imply that selection alone may not account for this. This is not necessarily to say that there is intelligence governing the principles of evolution - but perhaps that there may be hidden (at least at present) patterns within DNA that allow for interesting and gradual phenotypical progressions even with a random mutative engine.

---

I think the reason it doesn't jive well is because you're forgetting the key "non-random" part of the equation. Natural selection is the primary engine of evolution. Random mutation simply generates the variation - a nearly continuous process - upon which natural selection operates. Directional selection - where some change in a trait provides a net marginal fitness gain (or loss!) over the generations - has been observed. It can gradually increase or decrease the measure of some trait. Again, however, you're dealing with a population-level phenomenon over lots of generations. I'm not sure how you're able to say that patterned growth of a specific trait may not be explainable by natural selection. As I noted previously, some traits are fitness-neutral in and of themselves, but may vary in proportion to some other trait, as Mammuthus pointed out with his linkage disequilibrium explanation.