Software model of early evolution that really works.

[jerrywickey]

I have looked over several "evolution" simulations programs. Unfortunately every one I looked at is flawed. The most common flaw is that all permutations of all possible organisms are limited by the design of the software. This isn't representative of observations. The organisms cant grow in any way the program didn't already anticipate. i.e. every organism is made of eight bits. only eight bits. always eight bits.

 

The results from these programs are exactly what the author expects because, the author has unwittingly evolved the program to be successful.

 

 

==================

So I am writing new software, which solves this and the other flaws. But my software is no black box.

 

In this thread I will list the parameters and ask anyone to comment on them. I want to make the parameters closely representative of observations. I will then post the software for anyone to run on their computer. I am the most curious about how many generations will be required between a colony of similar organisms and a consistent and advantageous change in the colony.

 

each organism will have a "genome" made up of two nucleotides instead of four.

 

These "genes" will be greatly simplified also. The "genome" can evolve to include any number of "genes." organisms in my world, may grow and adapt, by collecting as many advantageous genes as their evolution gives them. They may grow to any size. the larger they grow and the greater number of them will increase the length of time to process one generation.

 

since in the real world the average length of a protein coding sequence is about 200 codons. This provides for a possible number of genes somewhere in excess of 20^200 I don't have the computing power to simulate that. In my simplified world a gene is always eight "nucleotides". In my world there are 256 possible gene sequence.

 

I will refer to these "genes" as sequences

 

I will start with a single self replicator.

 

The first replicator has only one "gene" sequence of 0101 1010. these eight "nt" are its entire genome. when this "nucleotide" sequence appears in the genome of any organism it will cause one organism to be added to the colony with an identical genome to its own.

 

Of course mutation may occur during duplication as well as other times in real life but mutations in my world can occur at any time and the mutation routine may be called to act on the new organism immediately after its creation. So I think that adequately covers that possibility.

 

from time to time during an organisms life a random mutation might occur. =R= will be the rate at which these mutations occur. 1 is one mutation per generation. 10 is 10 per generation. 0.1 is once per 10 generations.

 

The mutations can do one of six things

change a 1 to 0

or change 0 to 1

or add one "1" nucleotide at any random point in the genome this makes the genome longer

or add one "0" nucleotide at any random point in the genome this makes the genome longer

or subtract one nucleotide from any random point in the genome. this makes the genome shorter

or it could do nothing

 

A random set of sequences xxxx xxxx will be assumed to be advantageous by any means such as making it easier for the organism to acquire nutrients or making it easier to elude predators or any other advantageous characteristic. all these advantageous mutation really add up to only one thing anyway increased reproduction.

 

So each generation pass will compare the list of advantageous sequences with the sequences found in the genome of each organism. If an advantageous sequence is found the likelihood of multiple reproductions of this particular organism grows. i.e. if three advantageous sequences are found, this generation pass generates three copies of this organism instead of just the one for the reproduction sequence.

 

=A= will represent the number of advantageous sequences.

 

A random set of sequences xxxx xxxx will be assumed to be disadvantageous by any means such as making it more difficult for the organism to acquire nutrients or making it more difficult to elude predators or any other disadvantageous characteristic. all these disadvantageous mutation really add up to only one thing anyway decreased reproduction.

 

So each generation pass will compare the list of disadvantageous sequences with the sequences found in the genome of each organism. If a disadvantageous sequence is found the likelihood of multiple reproductions of this particular organism is diminished. if three disadvantageous sequences are found, there is only a one third chance that this particular organism will reproduce in this generation pass.

 

This is a different way of dealing with limited resources. In a new world full of the chemicals which spawned the new life, limited resources might or might not play a significant role in early evolution. But by including disadvantageous sequences we answer the possibility. There is no difference between the effect of an disadvantageous sequence in my world then the effect of a real life gene mutation which makes an organism less successful at competing with more successful organisms for a limited resource.

 

The effect is the same, reduced reproduction.

 

=D= will represent the number of disadvantageous sequences.

 

A random set of sequences xxxx xxxx will be assumed to be fatal.

 

So each generation pass will compare the list of fatal sequences with the sequences found in the genome of each organism. If a fatal sequence is found the organism is extinguished

 

=X= will represent the number of fatal sequences.

 

all other sequences are neutral, neither causing harm or an advantage.

 

In my model of the early world there is no death except by fatal mutation. A single cell replicator continues to replicate with no arbitrary life expectancy. Death of old age, life span, must have presented some evolutionary advantage at some point to be introduced into the genome of the world as we see it today. In my early world model, death has not yet been introduced.

 

R is the rate of mutation

A is the percentage of sequences which are advantageous

 

(note this is not the rate of advantageous mutations. This is important because, a mutation could cause a disadvantageous sequence to arise but the organism survive anyway. Later this same mutation augmented by another mutation might give rise to a very advantageous mutation.)

 

D is the percentage of sequences which are disadvantageous

X is the percentage of sequences which are fatal.

 

 

Does anyone have any ideas? Any corrections?

 

I will check back in the forum as I am writing this software.

 

What I expect to happen is

 

1) sometimes the very first replicator will mutate fatally. in which case there will be no life

 

2) there will be a lot of mutation which do nothing and genomes will grow, including and carrying many nonsense sequences.

 

3) among all the organisms the most successful will predominate in large numbers

 

4) there will be several groups of successful organisms arising from a few good starts

 

5) each group of similar organisms will share very similar genomes but neutral mutations will cause diversity

 

6) This diversity while neutral by itself may contribute to building advantageous or disadvantageous sequences in later mutations

 

that looks like about it to me.

 

Can anyone see anything else?

 

I am very curious to see the software results. I am starting to write it tonight, right this minute.

 

I would also greatly appreciate any positive comments for my efforts. Any one think this is a good idea? Anyone interested in the software? Please let me know. Post a warm thanks.

 

Thanks from me in advance.

 

Jerry