[alphazard]

If you take the idea of genes as the target of evolution seriously, then every possible "bargain" between different genes that moves towards a pareto optimal for those genes, will eventually be discovered through the brute force search.

[ajkjk]

brute force search can still be limited in the states it can reach. If there's some limitation on the types of moves you can make, which presumably there is, then you're limited to states that have paths between them.

[philipov]

And in Pareto, there's a rule like that built in because you're only allowed to make moves that increase utility. You're not allowed to move backwards, which can lead to getting trapped in a local maximum.

[jiggawatts]

That’s not precisely correct, because there is some “noise” in the system. Also, multiple genes can have competing effects, so one gene’s individual fitness can be suppressed by another.

[ajuc]

The mutations are like little nudges to throw you off the local maximum.

And there's junk DNA where mutations can accumulate over time without being subject to selection before getting enabled at random to see if they give you an advantage.

I think both the amount of junk DNA and the mutation rate are themselves subject to evolution for the best trade-off.

[immibis]

Note that most "junk DNA" is just DNA with a purpose we don't understand (originally, all DNA that didn't code for proteins). Some of it is true junk, of course.

[teiferer]

That's the exact opposite of random mutation.

[IAmBroom]

Assuming no extinction, climate change, nor heat death of the universe.

Evolution is not a particularly fast optimizer, on the scale of human perception.

[heavyset_go]

Evolution as we know it works on Earth because it is a relatively protected and stable environment, giving genes the time they need to explore a space and adapt to the relatively slow and small changes on Earth.

[NonHyloMorph]

Imho genes are not what if at all should be attributed aggncy here (that might, if at all, be the decentralised sympoietic process of ecosystems)

Just wanted to add this, i see where the formulation is useful as a figure of speech.

[graemep]

Climate change?

Climate change produces a changed environment. It actually drives more evolution to adapt to that changed environment.

Even very sudden changes such as some that caused mass extinctions, it just changed the direction of evolution. The only species that stopped evolving were those that went extinct. No more dinosaurs (bar birds) but lots more mammals.

[tsunamifury]

This assumes multiple infinite axis which certainly don’t exist and is therefore false.

Any early branch can infinitely on that confined branch. It means the billions of other possible branches may never be explored even given infinite time.

Counter would be the evolutionary fill theory where any branch can become any other branch given an environment.

[cgio]

That’s interesting! What other crazy types of bargains have we seen before or could someone imagine?

[jcims]

I still struggle with the brute force search a bit. Just naively a very small gene has 4^500 possible combinations.

[HarHarVeryFunny]

Genetic variation from one generation to the next is incremental - not a matter of tearing it all up and try some something random, not brute force exploring our way through all combinations.

Evolution seems more like building a tree where mostly all you can do is ascend the tree and add finer detail, leaving the trunk and branches (our evolutionary history) in place. It seems unlikely that, say, vertebrates are in the future going to "undo" the major evolutionary developments of the past and lose their skeleton, body symmetry, number of limbs, lungs, alimentary canal, nervous system, brain, etc. We see things like these developing in the evolutionary tree and mostly staying in place once created. Sure some fins turned to limbs, some gills to ears, but once things like that happened they seem to stay in place.

I wonder what evolution would look like if we could see it sped up from the origin of life to billions of years into the future? A building up of complexity to begin with, but those major branches of the evolutionary tree remaining pretty stable it would seem. Continual ongoing change, but of smaller and smaller scope, perhaps - building on what came before.

[thaumasiotes]

> It seems unlikely that, say, vertebrates are in the future going to "undo" the major evolutionary developments of the past and lose their skeleton, body symmetry, number of limbs, lungs, alimentary canal, nervous system, brain, etc. We see things like these developing in the evolutionary tree and mostly staying in place once created.

Try looking at whale skeletons over time. What isn't beneficial gets undone.

[knodi123]

Try looking at the giraffe's recurrent laryngeal nerve. What isn't beneficial is sometimes retained as long as the cost isn't bad enough to impair reproduction.

https://en.wikipedia.org/wiki/Recurrent_laryngeal_nerve#Evid...

[heavyset_go]

What doesn't affect fitness, or has relatively little cost, can and does propagate over time. By definition, nothing is selecting against them.

Similarly, beneficial and complex traits, like eyes, can "regress" if nothing selects against that trait. Plenty of species have lost their sight, making them less generally fit for many environments, because in a certain place and time those species could reproduce even without perfect vision, or just as the result of genetic drift.

[HarHarVeryFunny]

Yes, but it still seems that large scale structure tends to be preserved and it's more localized things like limbs/eyes/ears/teeth that may adapt. A chicken may have no teeth, but it's still basically a therapod.

I'm guessing there may be at least a couple of reasons for this:

1) Large scale structures evolved over long periods of time, involving layer upon layer of genetic change. This isn't going to be undone quickly or by any localized change, and those rare cases where a genetic change/defect does impact some fundamental aspect of the body plan (e.g. a frog with six legs) are very unlikely to be successful.

2) It seems possible that evolution acts to preserve large scale structure that has proved itself over time, and changes to which tend to be detrimental. In the same way that sexual reproduction seems like an evolution hack to evolve faster, then perhaps animals have also evolved genetic hacks to preserve/stabilize large scale structures that are critical to survival.

[thaumasiotes]

>>> It seems unlikely that, say, vertebrates are in the future going to "undo" the major evolutionary developments of the past and lose their skeleton, body symmetry, number of limbs

> but it still seems that large scale structure tends to be preserved and it's more localized things like limbs/eyes/ears/teeth that may adapt

Make up your mind.

[jibal]

"Unlikely" events happen nonetheless. Viruses and barnacles are examples of discarding structures, and it could possibly happen in some vertebrates.

[heavyset_go]

In a billion years, the sun's intensity would have increased such that life and the Earth itself will look very different, assuming life can adapt to living on basically a different planet. There might not be oceans left by then.

[HarHarVeryFunny]

When the environment changes sufficiently to wipe out whole branches of the evolutionary tree, I'd still expect those branches still alive to evolve in incremental fashion. Even if most lineages were wiped out, leaving only extremophiles, then those would still be building upon their own evolutionary history.

[heavyset_go]

I assume we will have sterilizing temperatures and pressures on Earth if all of the water ends up in the atmosphere, and heat releases even more greenhouse gases + subterranean water, leading to a runaway situation like Venus or worse.

[HarHarVeryFunny]

Maybe eventually, but I'm not sure of the relevance to the discussion, which is about the incremental nature of evolution.

[didibus]

It's not brute force search, more like stochastic sampling of random variants. It's a directed search where feedback from the environment prunes and weights the search space, and reproduction is a stochastic process biased by fitness. And fitness is defined by the survival and chances of successful mating, aka reproductive success.

Basically, your sex drive is the main search optimization :p

Edit: This is essentially how genetic algorithms in computer science work. They’re often remarkably effective at finding good solutions without needing brute force.

[Tagbert]

It is possible that not all combinations are equally likely.

[jcims]

Yeah it's working so something's going on .

[treyd]

The actual space is a lot smaller than it looks. Many amino acids have multiple codons that encode for them. You can also exclude cases where you have repeating stop codons (which detatch the RNA from the ribosome).

There's lots of processes that favor certain patterns over others, only considering the biochemistry of the cell, not even the fitness of the animal.

[alphazard]

> Many amino acids have multiple codons that encode for them.

I didn't know this. I suspect this evolved because some amino acids are more useful than others, and increasing the probability of encoding for them was beneficial.

[robbiep]

Have a read (rather than guessing) - it’s fascinating! Your other reply has a good insight but on more of a related topic but the primary reason this exists is for error correction. So approx one third of single nucleotide mutations have no change on the expression of the DNA or protein. And some of those that do change the amino acid are actually conservative; ie changing a basic amino acid to another basic amino acid which may still end up folding in the same way

[thaumasiotes]

There was a paper a while ago documenting that "synonymous" codons took different amounts of time during assembly, causing differences in the folded structure of the protein.

[jcims]

I've spent hours watching Drew Berry/WEHI movies and that whole process just seems like straight up alien technology. Blows me away to think about the scale that it's operating at within my body as I type this.

[20after4]

I think that maybe (with perhaps a very small probability) it actually could be alien technology.