The Naked Scientists

The Naked Scientists Forum

Author Topic: Does the mathematics of evolution work?  (Read 1617 times)

Offline dhjdhj

  • Full Member
  • ***
  • Posts: 78
  • Thanked: 1 times
    • View Profile
Does the mathematics of evolution work?
« on: 13/01/2016 13:26:54 »
I don't know whether this question has been asked before in this forum, but it is something that has troubled me for some time. As I understand it evolution is achieved by random mutation followed by natural selection. Its the word random that gives me problems. I can understand that there could be circumstances where by a mechanism is created within a living being to make it mutate to either take advantage of that circumstance or to mitigate the effects of it, but what are the chances of a random mutation doing that. If we take the higher mammals for instance there doesn't appear to have been a significant helpful mutation for may thousands of years. Has any one done the maths? Are there sufficient generations between us and the start of life on earth to create enough helpful mutations to get to where we are?


 

Offline Atomic-S

  • Hero Member
  • *****
  • Posts: 935
  • Thanked: 18 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #1 on: 14/01/2016 06:40:43 »
Given enough time, there would accumulate a 50% probability that the air in this room will self-liquefy and fall to the floor by chance.  But it could take a while. 

It is unclear to me that the genetic variations that are thought to have allowed the finches in the Galapagos Islands to differentiate themselves based on their various habitats occurred by chance during the time interval in which the adaptations were occurring. It would be just as feasible, and in my view significantly more likely, that the finches at the beginning of the adaptation period were externally largely alike but contained within their genes unexpressed variations already, that later were naturally selected by habitat to create distinctive populations.  That this view is more likely to be correct than the idea that the changes occurred by random chance during the time is suggested by the fact that experiments with putting a population of wild wolves through a selection process based on behavioral characteristics, created distinct populations (varying not only in behavior but also fur color) within a single human lifetime.  It is extremely unlikely that this differentiation was made possible by simply random chemistry, such as might occur as a result of natural radioactivity, because if that were the cause, we should have expected a great many failing examples that produced unviable animals.  Thus, the more reasonable hypothesis is that the population already had these variations within its genes to begin with, and circumstances favored the expression of one or another set of genes.

 

Offline evan_au

  • Neilep Level Member
  • ******
  • Posts: 4123
  • Thanked: 245 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #2 on: 14/01/2016 12:33:55 »
Quote
there doesn't appear to have been a significant helpful mutation for many thousands of years
A mutation that has spread rapidly in humans is persistence of lactase in adults, allowing drinking of milk in agricultural communities. The genes for lactase are not new, but there have been several mutations in the regulatory regions that would normally turn off this gene in young children.

In species that have been bred by humans, rare mutations are identified, and intentionally bred into the domesticated line, to the extent that it can sometimes be hard to identify the wild ancestor of domesticated species. In some cases, these mutations have been seen in historical times, such as seeing one branch of an apple tree that produces apples of a different color.

Probably the most rapid area of helpful mutations is in the ongoing battle against viruses, bacteria, and parasites. Because these organisms have a much faster lifecycle than humans, they accumulate helpful mutations much faster than humans.

We know that there have been helpful mutations because not everyone succumbs to every disease (or humans would now be extinct). Part of this protection is due to mutations, as well as shuffling of existing genes.
  • The many varieties of the Major Histo-Compatibility genes provide a variety of protections against novel invaders.
  • Some people have mutations which provide a degree of protection against HIV infecting white blood cells.
  • Many people have mutations for sickle cell disease, thalassemias, and G6PD deficiency which provide a degree of protection against the malaria parasite (at least when they only have 1 copy of the gene)
  • Some people have mutations that may provide some protection against tuberculosis (again if you only have 1 copy)
  • You will notice that some of these mutations do not have a totally good outcome - just a relatively good outcome in the presence of some severe endemic disease 
There are undoubtedly many other protective mutations which have not come to attention of DNA researchers because they are rare in the population, they don't have the negative effects of some of the above mutations, and they may protect against some disease which has not yet become endemic.

Hopefully, with the plunging costs of DNA sequencing, we should be able to sequence a child and their parents, measuring the incidence of new mutations, and perhaps even discover what proportion of them are advantageous, deleterious, and neutral.
 

Offline puppypower

  • Hero Member
  • *****
  • Posts: 556
  • Thanked: 43 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #3 on: 14/01/2016 13:24:40 »
The problem with the random assumption for mutations and selection is there are more random paths that can cause problems in existing systems, than there are progressive paths that can create useful change in systems that can coordinate with existing systems. For example, say I had automobile with 5000 components. What we will do is randomly change 1 in 5000 components. For each random component picked, we will randomly change that component in terms of making it better or worse, which may involved various random directions.

Say we change the radiator, if we make it worse the car will fail sooner. If we make it better, not every better radiator is useful. It may cool better, but it now may not fit under the hood properly. Random will make more things go wrong than go right.

If you look at random drug testing, there are side affects, even with useful changes in medicines. The development of drugs is not purely random, but uses logic and experience, which is not assumed for life. But in the end, medicines have side affects to go along with useful change. I suppose selection will not pick the side affects, and will not pick all those mice poisoned and made sick during early testing, which could take years. This path is too destructive to explain the healthy persistence of life.

If we look at the DNA, after the DNA is duplicated in cell cycles, there are proofreading enzymes that move along the DNA, checking the genes for typos, making sure based pairing is properly. The final changes in the genes, called mutations, implies oversight by the spell checker enzymes. Is this oversight deliberate or due to being overworked?

In terms of base pairing, if the base pair is not proper, there is a substantial energy increase; lingering potential, at the level of one or more hydrogen bonds. It appears to me a proofreading enzyme should be able to see these typos, by its energy profile. The enzyme will lower the potential by making perfect base pairing. Why would they ignore a typo when they are easy to see? One possibility is say you are proofreading and come to a homonym (word that sounds the same but is spelled differently). This gene change may fit in terms of the energy sound, but change the meaning. The heard of deer ran away from the bare.


The DNA is the most hydrated material in the cell. Integrated within each base pairing is hydration water, as well as water that extends from this hydrated water into the aqueous continuum of the nucleus. This water defines a unique fingerprint for the base pair.

http://www1.lsbu.ac.uk/water/nucleic_acid_hydration.html

Quote
The processing of the genetic information within DNA is facilitated by highly discriminatory and strong protein binding. It has been shown that the interfacial water molecules can serve as 'hydration fingerprints' of a given DNA sequence [889].
The normal 'hydration fingerprint' of the DNA will clearly be disrupted by DNA damage and this will facilitate repair protein attachment. The hydration spine (see above) is capable of carrying messages, as facilitated proton movement down the water wire, between binding sites in a similar, if complementary, manner to the electron transfer through the DNA residues [2258] and so coordinate the repair process.

My theory is, external global potentials in the cell water, for example, due to specific protein accumulation due to repeat environmental stresses, can impact the aqueous finger print, from the outside, to create a homonym. If the aqueous matrix near a mutation is made to look like it fits the energy sound, the proofreader will move on.
« Last Edit: 14/01/2016 13:26:30 by puppypower »
 

Offline dhjdhj

  • Full Member
  • ***
  • Posts: 78
  • Thanked: 1 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #4 on: 14/01/2016 19:26:19 »
I am aware that minor mutations occur within species. My question and it is a question, relates to mutations that fundamentally change a species. The arguably two most successful  mammals, humans and rats both have a common ancestor 60 million ago, and there have many intermediate species for both mammals each more successful that the previous one. The last major human mutation created modern man about 200,000 years ago and we saw off the Neanderthals. Some mutations like a giraffes' neck can be seen to gradually improve the animals ability and one can see how that might occur through natural selection over time. But mutations like the pepper moth which switches from one completely perfect camouflage to another completely different perfect camouflage, in one mutation as its environment changes, is harder to understand. So my question stands. As I understand it Darwin's random selection is the current theory and Lamarck's alternative theory has been discredited, but does the maths in Darwin's theory work?     
 

Offline evan_au

  • Neilep Level Member
  • ******
  • Posts: 4123
  • Thanked: 245 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #5 on: 14/01/2016 21:40:50 »
Quote
Lamarck's alternative theory has been discredited
Lamarck lost the first round, but has made a bit of a comeback with the discovery that epigenetic markers that are created during an individual's lifetime can be passed down to offspring.

Quote
the pepper moth which switches from one completely perfect camouflage to another completely different perfect camouflage
This example is better explained by mutations that increased in frequency in one environment, and not in another (smoggy city vs cleaner countryside). When the environment changed (anti-pollution legislation), the frequency of genes changed in the different populations.

It is not a perfect camouflage in either case; it is a matter of better or worse. Even a 1% difference in effectiveness can spread widely in 100 generations.

A more extreme example was seen in recent experiments where a changing environment switched on pre-existing inactive genes, which then spread through the population. (As I recall, this was done with a species of freshwater fish adapting to a salty environment, something which one assumes would have occurred multiple times throughout history.)

Quote
I am aware that minor mutations occur within species. My question and it is a question, relates to mutations that fundamentally change a species.
In the theory of evolution, mutations that fundamentally change a species mostly occur through a series of minor mutations within a species, combined with some non-genetic factors that cause isolation between groups of the same species (eg the course of a river, splitting a continent, individuals landing on different islands, etc).

So the question here is asking the theory of evolution to create a distinction which the theory of evolution does not really recognize.
 

Offline dhjdhj

  • Full Member
  • ***
  • Posts: 78
  • Thanked: 1 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #6 on: 15/01/2016 21:26:38 »
I'm glad to see Lamarck has been reinstated and I think we are nearly on the same page. My question really was relating to the concept of randomness. No one can seriously doubt the passing of the genetic information from generation to generation, but what triggers beneficial mutations? Something in the environment must have a bearing. Camouflage is an interesting subject of its own and one where I am in my comfort zone I was CEO of a military CCD ( camouflage, concealment, and deception) contractor for twenty years before I retired and I can assure you the pepper moth is perhaps not perfect, because the background scene is varied, but it is pretty damn good, the problem all creatures seeking to hide, and the military is that partial camouflage is no better than no camouflage, it's about getting over a survival threshold. So camouflage only works when it is virtually complete. It's possible to improve bit by bit, but there is no benefit without a pretty major first step. It is possible to argue that small mutations occur until there is benefit but why? evolution surely can't plan. My own gut feeling is that probably Darwin and Lamarck were both part right and that the living environment causes some mechanism within the gene pool to kick in to mutate beneficially.
 

Offline chris

  • Neilep Level Member
  • ******
  • Posts: 5339
  • Thanked: 65 times
  • The Naked Scientist
    • View Profile
    • The Naked Scientists
Re: Does the mathematics of evolution work?
« Reply #7 on: 16/01/2016 11:04:23 »
Martin Nowak, from Harvard University, looks at this sort of question through the lens of mathematics.

You might find his work of interest, including this recent paper published in eLife in which he considers social insects:

http://www.thenakedscientists.com/HTML/interviews/interview/1001617/
 

Offline dhjdhj

  • Full Member
  • ***
  • Posts: 78
  • Thanked: 1 times
    • View Profile
Re: Does the mathematics of evolution work?
« Reply #8 on: 16/01/2016 19:57:16 »
Thanks Chris. very interesting article there does seem to be much more to evolution than is yet understood. I think my question is perhaps currently unanswerable. 
 

The Naked Scientists Forum

Re: Does the mathematics of evolution work?
« Reply #8 on: 16/01/2016 19:57:16 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums