Tuesday, February 12, 2008

The Crux of the Matter

The line of comments is getting too long, so I thought I would pull things together again.

(1) I agree with Marco that it is plausible that there be unexpressed genetic ‘subroutines’ that might be expressed under particular environmental stresses, leading to changes of phenotype that mimic Lamarckian evolution.

(2) I agree with Marco that is plausible that there be control mechanisms that under the pressure of particular environmental stresses allow increased levels of mutation in particular ‘subroutines’ that correlate with responses to the particular stress.

(3) I agree with Marco that a filter can plausibly be applied to these randomly mutated subroutines to allow only some to be expressed.

(4) I further agree with Marco that gametes containing selectively expressed altered sub-routines could possibly be tested with respect to some parameter *correlated* with fitness to survive a particular environmental stress, before they are used.

So what are we still arguing about?

I think all this amounts to a plausible model for something which is tolerably close to Lamarckian evolution, but does not amount to *directed* mutation towards a phenotype that will provide better fitness under specific environmental conditions. And Marco still seems to be pressing for classical *directed* mutations. He says:

However this really opens up a door to: a theory that postulates selective 'good' mutations in the direction of better fitness ...

But it doesn’t. And 'it really opens up a door to' is not an argument. There is no physical way you can selectively mutate a particular gene such that the organism will be better able to cope with a particular condition. How can you selectively let through mutations that *will be* useful? How can an organism’s control systems have any foreknowledge of how a change in the gene will alter the fitness of the phenotype? My answer is, it can’t. To find out what the ‘program’ does, you have to ‘run the code’.


That's where we're up to. Anyone else want to join in? It really is good fun :)

5 comments:

Marco Parigi said...

However this really opens up a door to: a theory that postulates selective 'good' mutations in the direction of better fitness ...

I unreseverdly withdraw this statement on the grounds that it isn't scientific.

(4) Is essentially what I thought this statement was saying - but (4) says it so much better and in a tidy scientific format.

Marco Parigi said...

I would like to add that my vision of "regressive evolution" is kind of as follows:

A) A new kind of stress is applied to a population of organisms.

B) The organisms respond by generating gross mutations a,b,c,d,e,f and g in offspring. These mutations are of reasonably large effect and offspring a might be a smaller phenotype, while offspring b might be bigger (ie the mutations will be going in different directions).

C) The surviving offspring (say b and e) will have a new gene that tags the stress signal to the mutation that it was given and adds a weighting.

D) Those that don't survive will be deemed an unsuccessful response to that stress and the weighting will be lost in a standard Darwinian way.

E) When that same stress signal happens again many generations hence the mutation response will be given a higher weighting to that which survived last time.

F) Thus the second time for every a,c,d,f and g mutation generated in offspring two b and two e mutations will be generated when that same stress signal is applied.

G) Thus appropriate mutations to different stresses will be learnt even if the stress only repeats every hundred or thousand generations.

Dave said...

I am not remotely qualified to participate in this discussion, but I am interested in mutants.

Dr Clam said...

Especially Kitty Pryde, if I remember correctly...

Or do I have you and me confused again? The 80s were such a blur...

Dave said...

In the 80's, I was all about Storm. Hubba hubba. Then Halle Berry kinda ruined the fantasy.

Nowadays I'm swinging back round to KP.