Author Topic: I have an idea  (Read 9011 times)

Offline -venom-

  • Bot Neophyte
  • *
  • Posts: 47
    • View Profile
I have an idea
« on: September 06, 2007, 02:38:19 PM »
have you ever seen that movie called "evolution" were a meteor brings alien life to our planet and it quickly evolves into more and more complex things? well I was thinking is there a way to hype up mutations and reproduction enof to increase evolutionary speed similar to the movie??
If this were a perfect world, nobody would have let me into it.

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
I have an idea
« Reply #1 on: September 06, 2007, 05:19:48 PM »
Evolution is a function of the mutations/generation ratio.  If it's too high, mutations overpower any natural selection and you have a mutation meltdown.  The DNA just turns to non coding goo.  If it's too low, evolution lacks the novelty to break out of local minima (something that's short term beneficial but long term is a dead end).

In addition, the idea that evolution is a nice linear process from a to b is extremely tempting but inaccurate.  It's more like a blind drunk guy trying to walk home.  It's extremely susceptible to dead ends.  On Earth, most major advances in complexity were in response to a previous massive extinction event.  Quite often you need to wipe the slate clean and start over.

And last, there's an undeniable element of time involved.  There's a finite speed that molecules can be moved from place to place with osmosis.  This sets a finite pacing for natural selection to act that can't really be sped up.

So in short, increasing the mutation rate won't work.  Increasing the generational turnover will help, but only up to a point.  If you could somehow dilate time (for real life) or massively increase the cycles/sec (on a computer), you could speed it up, but both are rather hard to do.  And even then, you could run a simulation forever, and not get anything more complex than when you started because evolution is extremely prone to finding dead ends that it can't navigate out of (for computer people, evolution is basically like an imperfect greedy search done in massive parallel).

The movie, in case you're wondering, is extremely naive in its treatment of biology.  Not just evolution.

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
I have an idea
« Reply #2 on: September 06, 2007, 08:00:07 PM »
The problem is that most mutations are delaterious or neutral.  Advantageous mutations are rare and take time to fixate in a population (function of conveyed advantage, generation time and lifetime fucundity) and generally only convey a small relative advantage.  Delaterious mutations on the other hand are generally 100% deadly and (this is key) advantageous mutations don't help you defend against mutations.  Delterious mutations are essentially random death from on high.   Evolution requires a low enough deleterious mutation rate that advantageous mutations have "time" to become relatively widespread in a population relatively quickly relative to the delaterious mutation rate.   If the deleterious mutation rate is too high, advantageous mutations can't fixate before they are destroyed.  

Let me try an example.    Assume 500 out of 1000 mutations are delaterious.  499 are neutral and 1 is advantageous.  (rather optomistic).   Thus, the chances a mutation will adversely impact you is 50%.  Now assume the birth mutation rate is 100%.  Namely, all births mutate.  That's a pretty high mutation rate.  This means 50% of all offspring will have lower fitness, 49.9% will have the same fitness and one out of 1000 will have greater fitness than it's parent.

With a fixed population (like in our sims) average lifetime fucundity is 1 given asexually reproducing organisms (it has to be or the population would increase or decrease over time) but that is an average.  Some die without giving birth.  Some give birth more than once but one or more offspring die before reproducing.   On average, one offspring for each parent survives to reproduce.   At 50% deleterious mutations, if we assumed mutations were the only thing impacting reproductive sucess, then on average, every orgamism has two offspring, one of which dies at boirth and one of which survives to reproduce.   On average.   If we assume preditors and accidents and just being plain unlucky, the ratio is higher.  Lets assume on average, a bot has 4 offspring in it's life, one of which survives to reproduce.  Half are killed at birth due to deleterious mutations and one of the four is killed by something else before he can reproduce.  Lets also assume an advantagous mutation improves a bot's chances of surviving to reproduce by 10% (also very opotomistic) but for simplicity sake, if a bot lives that long, he reproduces 4 times.   Namely, an organism with an advantageous mutation will have on average, a 10% better chance of surviving to reproduce relative to those without it.

What is the probability that when advantageous mutation occurs, it can survive long enough to fixate in the population?   Given the high 100% mutation rate, the chances are bascially 0.   The chances are 55% an organism born with an advantageous mutation will survice to reproduce (by definition, it was born without a delterious mutation, so it's one of the two offsrping that on average, are neutral or advantageous and since it is in fact one of the 1 in 1000 with an advantageous mutation, it has a 10% better chance of surviving than it's neutral sibling, but only 10%.  It can still get eaten, fall in a hole, starve, etc.).   But of the 4 offspring it will have on average if it survies to reproduce, half will die at birth due to delterious mutations.  Their inherited advantage helps them there not at all.  Of the two that don't die at birth, the chances of each surviving to reproduce is similarly 0.55.    Thus, the chances an advantageous mutation survives to the third generation is the probability of the first surviving multiplied by the probabilitiy of at least one of the second surviving:   (.55) * (.55 + (.55/2)) = 45%.  Already less than 50/50.  This is because both things must happen for the mutation to still be in the population at this stage:  the original bot with the mutation must have survived to reproduce AND one or both of it's neutral offspring must also have survived.   The chances of surviving to the 4th generation are lower still and so on.    I may not have the math exactly right, but you get the idea.  The high mutation rate prevents even highly advantageous mutations from surviving long enough to fixate.  It's like trying to survive in a nuclear reactor.    

So, bottom line, you need a low enough mutation rate so that when an advantageous mutation does happen, it can spread in the population before delterious mutations wipe it out...
Many beers....

Offline -venom-

  • Bot Neophyte
  • *
  • Posts: 47
    • View Profile
I have an idea
« Reply #3 on: September 06, 2007, 10:06:36 PM »
oooo lots of knowledge
ok I realize that hyping up the muts wont work.
and I do realize more complicated isn't always better.
but I was just wonderin if there was a way to speed up evo (other then faster cycles)
for porposes like zerosims and the likes
If this were a perfect world, nobody would have let me into it.

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
I have an idea
« Reply #4 on: September 06, 2007, 11:09:04 PM »
Quote from: -venom-
but I was just wonderin if there was a way to speed up evo (other then faster cycles)
for porposes like zerosims and the likes
I think utilizing self-regulating shepard bots as a means to optomize selective pressures towards certain goals holds a lot of promise....
Many beers....

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
I have an idea
« Reply #5 on: September 07, 2007, 01:34:19 AM »
I think the idea of a "frontier" is key.  It's kind of like the books later in the Dune series.  Where something(s) odd comes to known space from the fringes of the unexplored galaxy.  It's on the very edges of survivability that neat things happen, and then those neat things are brought back to the main herd.

Not sure how to set up such a frontier, but I it has to do with extreme conditions and areas of relative ease.  Maybe two different simulated environments networked together.

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
I have an idea
« Reply #6 on: September 07, 2007, 12:23:25 PM »
We are saying the same thing.  A frontier is just a means to create steep environmentally-based, directional selection pressures which can serve to accelerate the fixation of benificial mutations through population bottlenecking, but only if the geneic diversity is there to begin with.    The same exact thing is true of preditors and shepard bots. Think in tems of puncuated equalibrium - cycles, good times and bad.  Genetic drift creates diversity then positive selection operates on it.   But you have to have the diversity for selection to operate, which comes from compounded mutations, which as above, takes time and or larger populations and can't occur too frequently or things turn to mush.  So, you can climb mount improbable in a series of environmentally or preditor-inspired wind sprints (followed by resting periods to build up diversity) or you can utilize a slower, somewhat constant walking pace, but the time it to takes you to get to the top (actually a local maximum) is largely going to be a function of the path you take, not whether you sprint or walk.  The right selection pressures, be they punctual or gradual, keep the path shorter and insure the turns you take are always climbing (as opposed to wandering around in valleys and medows) but the slope has to be always gradual and the maximum altitude you can gain per unit time in the end is going to be limited by a time * population * mut rate equation.
Many beers....

Offline Peter

  • Bot God
  • *****
  • Posts: 1177
    • View Profile
I have an idea
« Reply #7 on: September 12, 2007, 11:50:49 AM »
I don't think you can speed up evolution, if you look in real live how big the chance on a mutation is, well it's really really small. Duplicating DNA for example is an extremely complecate process with different checkups that will check if the dna is correctly copied. Dna consists out of 2 strings if one is wrong it will be seen.
The reason there are still mutations is becouse the DNA is very often copied. How many cells do you think there are in your body. I think the chance that if you'd copy a file from one place to another . The chance on a wrong bit is higher then with DNA a wrong basepair. Rather strange becouse DNA works on molecule level and computers works on still a 'little' bigger level. Also becouse computers only use 0 and 1(binair). DNA uses A, T, G and C (Quaternary)

There is no single specie who speeds up mutation rates to speed up evolution. The lower the better. For every ancetor that has better DNA there will be more with a worse DNA and an huge ammount of death ofspring.

The best for evolution are really small chances of mutation. Althrough I have so say, the rusian space station 'Mir' had much trouble with aggresive fast reproducing bacteria and fungus. Becouse of the radiation coming out of space having higher mutations.
Oh my god, who the hell cares.

Offline Elite

  • Bot Overlord
  • ****
  • Posts: 532
    • View Profile
I have an idea
« Reply #8 on: September 12, 2007, 12:54:10 PM »
The best method I have come across to "speed up evolution" is to cap the maximum bot age.

Generation time is reduced, and genes that confer benefits at a gene-level (ie. that benefit offspring) are selected over genes that benefit individual bots.

Although it is a little artificial, I've has excellent results with this strategy.

I don't think it's 'speeding up' evolution so much as disadvantaging 'bad' (from our and genes' PoV) mutations.

To speed up DB evolution proper, get a faster computer
« Last Edit: September 12, 2007, 12:57:34 PM by Elite »

Offline shvarz

  • Bot God
  • *****
  • Posts: 1341
    • View Profile
I have an idea
« Reply #9 on: September 12, 2007, 12:57:48 PM »
Quote
There is no single specie who speeds up mutation rates to speed up evolution. The lower the better.

Peter, this is incorrect. In order for evolution to occur, diversity must be generated through mutations. If the mutation rate is very low, then new variants are not produced and evolution cannot proceed.

Also, bacteria are known to speed up their evolution when living conditions become extreme. They switch to a less accurate DNA polymerase, which introduces a lot of errors in the hope that they will generate a mutation that would help them to cope with changed conditions. However, it is a desperate and a last-resort mechanism.
"Never underestimate the power of stupid things in big numbers" - Serious Sam

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
I have an idea
« Reply #10 on: September 12, 2007, 02:49:41 PM »
Cells involved in producing antibodies also invoke intentional mutations (albeit very specific ones) to try and come up with novel antibodies to fight an infection.  So it's definitely a sweet spot.  If your mutation rates are too low, you get out competed (assuming the environment isn't a perfect match to what you are now) by more adaptive species.  If it's too high, you lose traction with natural selection and turn to non coding goo.

Offline Welwordion

  • Bot Destroyer
  • ***
  • Posts: 325
    • View Profile
I have an idea
« Reply #11 on: November 19, 2007, 09:16:09 AM »
Its similar to science, doing experiments cost a lot of resources and time, your society (population) needs enough expandable resources (expandable individuals) in order to do the testruns and the better your test(more selective your conditions) the faster you will reach a result .
However what would help to speed up is a kind of lamarckian evolution, an individual testing different sets of gene expressions in order to then discard the failure genes and keep the succesful genes.

Also I recently read that horizontal gene transfer(between in dividuals) may play a larger role to evolution than thought until now, so if you give genes a way  to spread in a population that might also speed up your testing.
(actually that article http://www.physorg.com/news114185292.html is the reason I amback here want to test if an evosim with virusspreading fastens the evolutionary pace)
« Last Edit: November 19, 2007, 09:22:35 AM by Welwordion »

Offline fulizer

  • Bot Destroyer
  • ***
  • Posts: 198
    • View Profile
I have an idea
« Reply #12 on: December 05, 2007, 03:46:25 AM »
houh yes the edge of survival try smaller amounts of bots in a larger area that way the need to actively move to find your food will quilkly become apparent and eventually more efficient things will evolve.
this works for every alife I ever tried
"If this is coffee bring me tea, If this is tea, bring me coffee"

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
I have an idea
« Reply #13 on: December 05, 2007, 09:40:07 AM »
What other alife programs have you played with?

Offline Welwordion

  • Bot Destroyer
  • ***
  • Posts: 325
    • View Profile
I have an idea
« Reply #14 on: December 05, 2007, 03:31:07 PM »
There is yet another possibility to speed up evolution, to apply higher mutation rates at specified  places, that way you can keep a certain structure in genes while mutating specific thinks like conditions or values that are  stored, even in real Dna the ends of the chromosome have higher crossover probability so genes that have a higher need or tolerance for mutations are placed there.