Pick DNA components from various species and add them into a computer simulation that lets you see what the new species would look like/perform. new animals/plants could be pitted against one another in a virtual survival of the fittest war theatre. inputs from players all round the world would create the necessary diversity for evolution and the ultimate survival winner.
-- williamsmatt, Sep 18 2008

Spore -- Create completly new creatures http://www.spore.com/
[xxobot, Sep 18 2008]

Impossible Creatures -- Combine existing creatures http://www.microsof...mpossiblecreatures/
[xxobot, Sep 18 2008]

not as simple as that. But [+] for the idea...

oh and... hasn't this been baked already... see spore and numerous other games
-- xxobot, Sep 18 2008


How many species at this point have their entire genome decoded, let alone knowing what each gene is for? The processing power needed just to compute a change in an existing decoded species is too big for any home cluster of processors, let alone compiling a new one.

I'd like to see a glow in the dark bunny take on Dolly the sheep.

God games are fun.
-- Giblet, Sep 18 2008


Genes are not 'for' anything - they just are, and, if in combination whith other genes, they tend to produce developmental arrangements that allow the host to assist them in their replication, then all well and good, roll again.
-- zen_tom, Sep 18 2008


//a computer simulation that lets you see what the new species would look like/perform//

Sounds like magic to me.
-- Wrongfellow, Sep 18 2008


What everybody already said. It's not theoretically impossible (genes are just code), but we're orders of magnitude away from being able to do it at present.

Given one gene for, say, a transcription factor (which regulates other genes), we have no way to accurately predict the structure of the protein it encodes.

If we cheat and derive the structure of the protein by crystallography etc, we still can't deduce (except in a few special cases, and with a large error) what DNA sequence it will bind to.

If we cheat again and do experiments to see which DNA sequence it binds to, we still can't tell (in general) whether it will up-regulate or down-regulate the gene it controls.

If we cheat again and do experiments to find out whether it up- or down- regulates the gene it controls, we're then back to square one, one layer further in.

And all this is just local metabolism. We don't even know how to start asking how genes determine the shape of jawbone or the aggressiveness of a goldfish.
-- MaxwellBuchanan, Sep 19 2008


why can't we just wade in, give the programe the constraints we want to give it (hence saving us the time taken in actually knowing) and the just TELL everybody through an agressive marketing campaign that this is 'the real thing'?
-- williamsmatt, Sep 22 2008


Shhhh, //the aggressiveness of a goldfish.// Shhhh, ah, you mean the spirit/soul/character factor? Shhhh. As what I know, a robot presses on aggressively because the human controller yank the joystick more than enough!
-- rotary, Sep 22 2008


[williamsmat] You mean use misleading marketing to gloss over the lack of any kind of scientific rigour in the DNA-analysis service you're selling to a gullible public? Well, yes, that usually works (e.g. homeopathy, vitamins, herbal supplements, chiropracty, fish oil, etc., etc...)
-- hippo, Sep 22 2008


The basic idea is a very good one. Genetic engineering is widely used in the scientific (and agricultural) communities and has been utilised to a very low amount of its possibility.

A computer simulation would be a very interesting, and very possibly popular tool. The coding would be extremely difficult if the simulation were to actually take DNA though, as we still only know a very small amount of the functions DNA performs, and even with several projects, only know what 5-10% of DNA even does.

The idea is amazing though. Especially if you factor in multiple generations, mutations, and other genetic factors to make the simulation even more interesting. Croissant for originality and the thought of a program like that.
-- bluefood2010, Sep 23 2008



random, halfbakery