Variation and Variability

The idea for this thread came from this post in the creationist sub-forum. Thanks to philadiddle for explaining to me how to link to it.

http://www.christianforums.com/showpost.php?p=35700766&postcount=6


Here are the points I want to concentrate on:

Some genes can be altered but it should be understood that there is a big difference between an adaptation and a random mutation.

Click to expand...

However, most of the wide variety the we see in all it's vast array most likely resulted from the recombination of unaltered genes in rich gene pools.

Click to expand...

First credit where credit is due. The first statement is quite correct. In fact we can leave out the word "random". There is a big difference between an adaptation and a mutation, period. In fact,there is a difference between a mutation and a variation, never mind an adaptive variation.

The second statement is also true in the terms it is stated. Most of what people identify as variation in a population owes more to recombination of unaltered genes in rich gene pools than to new mutations producing new variations.

However, what this statement overlooks is the difference between "variation" or "variety" as commonly perceived and "variability" which is the actual focus of evolutionary studies.

Let us start back at basics. A mutation is a change in DNA sequence. The first thing we should note is that a mutation does not necessarily produce a variance in a genetic trait. Due to redundancy in the genetic code, a change in DNA sequence may still produce the same amino acid, which produces the same protein, and the same non-varying trait. This fact enables geneticists to study whether or not evolution is occurring. Since these synonymous mutations produce no variation, they are immune to selective pressures. A comparison of the fate of synonymous to non-synonymous (variation-producing) mutations helps determine whether selection pressures influence the inheritance of the non-synonymous mutations.

They don't always. A variation may be quite neutral in its survival benefit. So even when a mutation does produce a variation, the variation is not necessarily adaptive or maladaptive.

So it is very true that there is a difference 1) between a mutation and a variation and 2) between a mutation and an adaptation, and 3) between a variation and an adaptation. A lot of confusion does arise from treating these terms as synonyms when they are not.

Now to the second statement.

Again, much of what people see as variety is due more to recombination than to new mutation. Let's take 8 features and give them each two possible alternates:

gender (male, female)
hair colour (dark, light)
hair texture (curly, straight)
eye colour (blue, brown)
skin colour (dark, light)
ear lobes (long, short)
nose width (narrow, wide)
lips (full, thin)

Simple math tells us that when these are randomly assorted we get 512 different combinations of these eight traits.

Traits are not always assorted randomly. A single gene which controls several traits means those traits tend to be inherited as a package deal. Ditto when two genes are found close to each other on the same chromosome. And even without these qualifiers, we see common associations of certain traits. Blue eyes in people with dark skin and hair pigmentation is much, much rarer than among people with lighter skin and hair pigmentation.

The fact that some combinations of traits come as package deal rather than being randomly assorted explains some features of evolution. If out of a bundle of five traits all dependent on the same gene, one is adaptive, the others will be inherited along with it, whether they are adaptive or not. Some could even be moderately maladaptive and still be preserved if the one supported by positive selection is of significant benefit.

My main point, however, is that while recombination explains how we get many combinations of different traits in a species and so explains much of the variety we see, it does not deal with the sort of variation evolution studies.

Evolution does not study primarily how different traits are combined. It studies how much variation exists in one trait and what happens to the proportional distribution of the variants of one trait. To keep us from becoming confused, I suggest we call the latter a study in variability, rather than a study of variation.

Variability is not about differing combinations of two or more traits depending on two or more genes. It is about how many different variations of a single gene we find in a population, how those variations are distributed in a population, what changes occur in this distribution from generation to generation, and what causes influence these changes in distrbution.

In the example of recombination above, we noted two alternatives for each trait. But we know that for some of them there are many more than two alternatives.

The first question to be asked is "Why is any gene expressed in more than one way?" What causes a gene to be expressed in more than one way? One answer appears to be a variation in DNA sequence in the gene itself. Another would be a variation in DNA sequence in a regulatory section of the genome that tells a gene when to be active and when not to be active. I am sure there are other possibilities. But these two at least depend on differences in DNA sequence.

Why do these differences exist? One could say the species was created with alternate forms of the DNA sequence. If not, the variation in the DNA sequence must have come about through a mutation. So even when there are only two alternate sequences, one possible explanation of why they exist is mutation.

Also, even with only two alternate expressions of a gene, you still have differences in distribution and sometimes changes in the distribution of the alternatives. The latter, by definition, is evolution.

The next question to be asked is "How many variants of a single gene exist in a population?" For some genes the answer is none. All members of a species show the same DNA sequence and express the same gene in the same way. For others we can find two, three, a dozen, and even in some cases, hundreds of different variants of the same gene affecting the same trait.

This is variability. Not combinations of different traits dependent on different genes, but how many ways a gene can be formulated in its DNA sequence, how many different variants of the traits governed by the gene show up in the populations, and what happens to the distribution of these variants in the population over time.

So although the point on recombination is valid in one framework, it is not the sort of variation which is the actual focus of evolution. The actual focus is on variability.

Once this concept is understood, it poses some problems for a recent creation of Genesis "kinds" But I will need to leave that for a later post as I am running short of time.

philadiddle said:

In each thread the post # is clickable, and will take you to a page with just that post on it. Then cut and paste url.

Click to expand...

Thanks. These things often turn out to be incredibly simple when you know about them.

Continuing from OP

One viewpoint often expressed by YECs is that the original created kinds were endowed with a wide range of variability, so that they could adapt to varying conditions.

One corollary of that is the view that each species within the kind lacks some of the original range of variability. Only the original created kind had the full range of variability, and even if we totalled up the full range of existing variability among all the species in the kind, we might still not see the full range of originally created variability as some variants may have been permanently lost and not exist in any species today.

Once we understand that variability is not about how different traits dependent on different genes are assorted and combined, but about how one gene is sequenced and expressed differently in different individuals and sub-populations, this poses some interesting implications for these viewpoints.

Key here is that cells, with very very few exceptions, are either haploid or diploid i.e. they contain one or two copies of the species genome and no more. No matter how many variations of a gene there may be in the population at large, in a haploid species, each individual can carry only one of them. In a diploid species each can carry at most two of them. And nothing prevents individuals in a diploid species from carrying duplicates of the same genetic variant rather than two different variants.

This means that unless one allows for mutations, the number of variants of a gene in the population at large sets a minimum figure for the population of the original kind.

To illustrate. If we find in a population 20 variations of a particular gene, and each individual can carry only two of them, it takes a minimum population of at least 10 to carry all of the variant forms of that gene in that population. And that is assuming that not a single one of the individuals carries duplicates of the same variant, and that no variant is found in more than one individual. Both of those are highly improbable assumptions.

This becomes a significant problem if a gene shows a lot of variability: hundreds of variants show up through the population. And given that a "kind" is often now represented by a plurality of species, we need to look not only at the variability in one species but in all the species in a kind. If "frog" is a "kind" that is over 3,000 species, each with many individuals.

There is a huge potential for variability in so many copies of the frog genome. What size would the original population of frogs need to be to include that much built-in variability? It would seem the orginal created kind must have been a population of at least several hundred individuals.

It would seem to me that on this basis alone, one would have to recognize mutation as a source of new ranges of variability. In short, for many kinds the total range of variability within the kind as represented in a cluster of species could not have been built-in at creation. At least some of it would need to be introduced by mutation.

The problem is intensified if we look at the creation of humans. Nothing says that the original populations of lions or penguins or crickets was limited to two individuals. But the biblical depiction of human creation is that of two individuals, one of which was created from the other. This would seem to limit the original variability of all human genes to a maximum of two variants. Every extension of variability would have to come from mutation.

A final consideration is the reduction of variability due to he flood (if one insists on a global flood in which all terrestrial organisms save those on the ark died).

Since the range of variability depends on the size of the population, a reduction of the population to a very few individuals destroys most of the created built-in variability anyway. So if only two frogs survived the flood, any great range of variability would be reduced to no more than four variants at most. Again this presupposes that both individuals are heterozygous for every trait that shows variation, and that no variant appears in both of them.

The current range of variability has to be reconstructed in post-flood generations via mutations. This means mutations have to be given a much larger role and a much more positive role in generating variability than many creationists seem ready to acknowledge. Built-in variability simply doesn't do the job if most of it gets destroyed in the flood anyway.

Assyrian said:

Re horses alleles, the problem with geegees is that they are a clean animal (I think) so Noah could have had up to 7x2 or even 14x2 sets of alleles on board the ark, depending on whether he took 7 individual animals or seven pairs.

Click to expand...

No, horses are not clean animals per Mosaic dietary laws. (Leviticus 11: 1-8) A clean animal had to chew the cud and have a cloven hoof. Pigs have cloven hoofs, but do not chew the cud, so not clean. Camels chew the cud but do not have cloven hoofs, so not clean.

I don't know if horses chew the cud, but they don't have cloven hooves, so not clean.

Assyrian said:

So horses are in then. (Would Solomon have kept a stable of pigs?)

Click to expand...

Maybe. If he was really, really fond of truffles. (I mean the mushroomy kind they use pigs to find, not the candy sort.)