Login

Join for Free!
118252 members


Increasing Information For Genetic Algorithm

Discussion of everything related to the Theory of Evolution.

Moderator: BioTeam

Increasing Information For Genetic Algorithm

Postby CryoGenesis » Fri Aug 31, 2012 5:20 pm

Hello there, I am writing a program in Java to simulate evolution. It all works as in the virtual creatures are better at things over time depending on environment variables and the creatures reproduce by mixing around the virtual genetic information.
What I want to do is really simulate real world evolution. Can someone explain to me how cells or creatures add more information to themselves? As in they get more complicated in closed systems. I don't know myself but is it mutation? or is it an environmental factor?
CryoGenesis
Garter
Garter
 
Posts: 1
Joined: Fri Aug 31, 2012 5:10 pm

Re: Increasing Information For Genetic Algorithm

Postby Rap » Sat Sep 01, 2012 7:20 am

Its probably a combination of both, but I have no idea of the relative importance of the two. Note that the genetic code is to some degree redundant. That is, there are a number of different DNA "codes" which yield the same protein. Two different strings of DNA which yield identical proteins will nevertheless have a different geometry, and will have different abilities to withstand mutation. If it is vital that a particular sequence remain unchanged, the DNA will evolve to the most mutation-resistant encoding, but if adaptability is required, it will evolve to a less stable encoding. I suppose the same concept might be applied to extending or shortening the DNA. I'm not sure if extending or shortening the DNA will always result in a new species.
Rap
Death Adder
Death Adder
 
Posts: 60
Joined: Tue Jan 18, 2011 5:57 pm

Re: Increasing Information For Genetic Algorithm

Postby wbla3335 » Sat Sep 01, 2012 10:14 am

Information content can be increased by gene duplication, which can occur on various scales from the duplication of quite small bits of DNA to entire genomes. The duplicated DNA can then ultimately end up coding for something new (or causing serious problems that will lead to immediate removal from the population).
wbla3335
Coral
Coral
 
Posts: 227
Joined: Thu Aug 17, 2006 7:20 am


Postby JackBean » Tue Sep 04, 2012 2:06 pm

Rap: DNA does not evolve into mutation-resistant. Simply if the mutation would be lethal then nothing with mutated DNA survives thus it seems like mutation resistant.
Also, the identical DNA strains have identical geometry (whatever that should mean) and, at least in the beginning, they will have the same rate of mutability. But after one of them reaches some point, where it is not functional in original sense anymore, the other one must be preserved, thus looking like mutation resistant (as above), while the first one can freely mutate untill it will get some new crucial function.
http://www.biolib.cz/en/main/

Cis or trans? That's what matters.
User avatar
JackBean
Inland Taipan
Inland Taipan
 
Posts: 5678
Joined: Mon Sep 14, 2009 7:12 pm

Re:

Postby Rap » Thu Sep 06, 2012 1:54 am

JackBean wrote:Rap: DNA does not evolve into mutation-resistant. Simply if the mutation would be lethal then nothing with mutated DNA survives thus it seems like mutation resistant.
Also, the identical DNA strains have identical geometry (whatever that should mean) and, at least in the beginning, they will have the same rate of mutability. But after one of them reaches some point, where it is not functional in original sense anymore, the other one must be preserved, thus looking like mutation resistant (as above), while the first one can freely mutate untill it will get some new crucial function.


See

http://en.wikipedia.org/wiki/Synonymous ... _evolution

and google "synonymous substitution mutability" and "silent substitution mutability" (without quotes)
Rap
Death Adder
Death Adder
 
Posts: 60
Joined: Tue Jan 18, 2011 5:57 pm

Re: Increasing Information For Genetic Algorithm

Postby Cat » Sun Sep 09, 2012 5:20 pm

Rap wrote: If it is vital that a particular sequence remain unchanged, the DNA will evolve to the most mutation-resistant encoding, but if adaptability is required, it will evolve to a less stable encoding.


I tried to get this idea through in several prior posts and doesn't seem to take... DNA does NOT mutate or evolve toward a goal!!! If the selective pressure is lifted, that result in diversity. For example, phenotypic diversity of dogs as compared to little diversity in wolves. If the selective pressure is applied to a diverse population, those that survive establish new race, etc.

The latest genetic data from experiments suggests that genes accumulate loss of function mutations over time while new archeological data suggests that in species on the verge of extinction (possibly from gene loss) whole genome duplication occurs that leads to "overnight" so to speak disappearance of one set of species and appearance of new ones. It's also useful to note that many new phenotypes result from loss of function in some gene (usually a repressor) and not due to acquisition of new genetic information...
Cat
King Cobra
King Cobra
 
Posts: 631
Joined: Thu Feb 14, 2008 7:40 pm

Re: Increasing Information For Genetic Algorithm

Postby Rap » Mon Sep 10, 2012 3:39 am

Cat wrote:
Rap wrote: If it is vital that a particular sequence remain unchanged, the DNA will evolve to the most mutation-resistant encoding, but if adaptability is required, it will evolve to a less stable encoding.


I tried to get this idea through in several prior posts and doesn't seem to take... DNA does NOT mutate or evolve toward a goal!!! If the selective pressure is lifted, that result in diversity. For example, phenotypic diversity of dogs as compared to little diversity in wolves. If the selective pressure is applied to a diverse population, those that survive establish new race, etc.


I agree, the paragraph I wrote is not the best, but I don't think its wrong. Applying selective pressure to a previously diverse population is not the only way for change to occur. If a new selective pressure is applied to a previously non-diverse population, the tendency will be for the selection of DNA better adapted to the new environment. That "better" DNA can be encoded as any number of "synonyms", which yield the same proteins, but which may have different susceptibilities to mutation.

If the new selective pressure is constant in time, then any deviation from the "optimum" will be selected against. The most mutation-resistant synonym will be selected for.

If the new selective pressure is short lived, (but extends over a number of generations), and is then replaced by a different one, which is short lived, and then another one, etc., then a synonym which is less stable will be selected for. The species will have more diversity because of this, and will be less likely to be driven to extinction by the changing environment.
Rap
Death Adder
Death Adder
 
Posts: 60
Joined: Tue Jan 18, 2011 5:57 pm

Re: Re:

Postby JackBean » Mon Sep 10, 2012 3:08 pm

Rap wrote:
JackBean wrote:Rap: DNA does not evolve into mutation-resistant. Simply if the mutation would be lethal then nothing with mutated DNA survives thus it seems like mutation resistant.
Also, the identical DNA strains have identical geometry (whatever that should mean) and, at least in the beginning, they will have the same rate of mutability. But after one of them reaches some point, where it is not functional in original sense anymore, the other one must be preserved, thus looking like mutation resistant (as above), while the first one can freely mutate untill it will get some new crucial function.


See

http://en.wikipedia.org/wiki/Synonymous ... _evolution

and google "synonymous substitution mutability" and "silent substitution mutability" (without quotes)


Yeah, so? How is that related?
http://www.biolib.cz/en/main/

Cis or trans? That's what matters.
User avatar
JackBean
Inland Taipan
Inland Taipan
 
Posts: 5678
Joined: Mon Sep 14, 2009 7:12 pm

Re: Increasing Information For Genetic Algorithm

Postby Cat » Mon Sep 10, 2012 6:22 pm

Rap wrote:If the new selective pressure is constant in time, then any deviation from the "optimum" will be selected against. The most mutation-resistant synonym will be selected for.

If the new selective pressure is short lived, (but extends over a number of generations), and is then replaced by a different one, which is short lived, and then another one, etc., then a synonym which is less stable will be selected for. The species will have more diversity because of this, and will be less likely to be driven to extinction by the changing environment.


That is not true. There is no such thing as "mutation-resistant synonym".

Under constant selective pressure organisms with mutations making them capable of withstanding it will become majority. ANY and ALL mutations beneficial for survival will go on and any ORGANISM (not gene) without any genetic mutations necessary for survival will be eliminated. GENES of that said (dead) organisms will remain in population of heterozygous individuals. Those genes can re-emerge when selective pressure is lifted.

I will give you an example:

You plate some bacteria on Kanamycin and wait. Most of the bacteria will die. However, a few days later you discover few very small colonies appear that are Kan resistant. If you were to investigate the reason for this, you will discover that those bacteria have mutations in ribosomal RNA that make them resistant to Kan. However, there are several possible mutations that cause resistance and any/all of those can be present. Also, such mutations occur REGARDLESS of selective pressure. It's just that they might not be apparent in absence of selective pressure. You would not know that some of your bacteria became Kan resistant until you grow it in presence of Kan...

Also, you are misunderstanding synonymous substitution. It's random as any other mutation. It may affect levels of translation of some genes. If it does so in any great extent, especially if it changes splice site, it will be considered a mutation. So, that part of Wiki is just not true.
Cat
King Cobra
King Cobra
 
Posts: 631
Joined: Thu Feb 14, 2008 7:40 pm


Return to Evolution

Who is online

Users browsing this forum: No registered users and 7 guests