Pseudogenes and Evolutionary Process

Factual Questions
1

A question inspired by this thread, in which pseudogenes are discussed as a proof of evolution. In this linked articles by Ben, they are discussed in Section 5.1.

My question is this: if pseudogenes are neutral or moderately deleterious (in the given example it was suggested that humans have lost much of their sense of smell by this process), what would account for them being distributed thoughout the species? My understanding is that mutations typically affect only one individual, but become standard for a species due to survival of the fittest. How do pseudogenes fit in with this theory?

2

The phrase “survival of the fittest” primarily applies to the concept of weeding out the deliterious genes/traits/whatever. If a gene sequence provides neither benefit nor harm, it is, effectively, invisible to natural selection. As such, with no mechanism for its removal, it would thus likely persist from generation to generation, eventually becoming commonplace.

3

I had a good response typed out, but Darwin’s Finch beat me to it. So instead, I’m going to say that, if memory serves, Richard Dawkins discusses pseudogenes at some length in his book The Selfish Gene. My copy is at home, so I can’t look it up just now. However, that book is an interesting read for anyone at all interested in evolutionary biology.

4

Why would it become commonplace? IOW, if some ancestor of today’s humans had a mutation that negatively impacted his sense of smell, why wouldn’t the percentage of people with bad senses of smell exist today in proportion to the one guy out of however-many-it-was that had the mutation? Why does everyone have this mutation?

5

Believe it or not, there’s a mild advantage to not having a highly developed sense of smell. Whatever energy previously went into developing that highly accurate olfactory organ is now available for other things, like finding food or a mate. If the disadvantages aren’t enough to outweigh the cost, this is better.

6

Let’s fix that:

“If the disadvantages aren’t enough to outweigh the benefits, this is better.”

7

During the last 100,000 years humans have done rather well, surviving. A very large percentage of humans procreate, and a comparatively large percentage of their progeny survive to procreate as well. Times are good, for humans. Since the billions of humans that are alive now are all descended from at most a few millions of humans then, all genetic characteristics are being conserved, not just the best of them. Divergence into different groups has been thwarted by migration over and over, during that same period. The mixing of genetic information among humans is very high.

Mildly maladaptive qualities have very little impact. Coincidental association with highly desirable traits could outweigh the small disadvantage. The number of failures to produce descendants is simply not high enough to eliminate much that is not specifically, and strongly counter survival. With the selection process muted, the random dispersal of the poor sense of smell gene would allow it to be widely expressed, even without specific benefit, since all genes are increasing in numbers during the period. When specific stresses occur, the sort on other criteria will increase the incidence of neutral genes in the smaller population, unless they are specifically associated with that stress.

I recall that the human nose is not conducive to scent discrimination, but is very effective in preventing the inhalation of water, when diving and swimming. The true neutrality of the characteristic is variable, depending on the specific needs of the human in question. Success in large numbers among one group for unrelated reasons would encourage the dissemination of the gene as well.

Tris

“Human beings are the only creatures that allow their children to come back home.” ~ Bill Cosby ~

8

Maybe I’m wrong, but I find this hard to believe. In any event, you would then say that we could expect to find universal distribution of pseudogenes only in cases in which they conferred a definite advantage, not where they were merely neutral.

Trisk

I have probably failed to understand your post, unless you only meant to answer the question with your last paragraph. There, you seemed to be saying that the loss of sense of smell is somehow connected with the ability to block water. This did not seem to be the case from Ben’s article - he suggested that sense of smell was dependent on the number of olfactory receptor genes (some of which had ceased to function in humans), not on the structure of the nose. In any event, you - like ultrafilter - would appear to be implying that pseudogenes would only exist (in all members of a species) in cases where they gave some survival advantage.

Is this the case?

9

Disclaimer: I’ve bought into the perspective detailed by Richard Dawkins in his books The Selfish Gene and The Extended Phenotype. Also, I’m not a biologist, just an educated layperson.

First off, what I said about the advantages of not having a highly developed sense of smell is my understanding of what Dawkins says. I’ll try to dig up a cite tonight.

Secondly, pseudogenes make a lot of sense if natural selection is taken to act on replicators rather than organisms (as is outlined quite well in the books above). In this case, pseudogenes are just along for the ride. Since they don’t deleteriously affect the survival rates of other genes, there’s no selection pressure against them.

Of course, a professional biologist should probably comment on this.

10

It would become commonplace simply because it is not being removed from the genepool. As individuals with these sequences reproduce, they become more prevalent in the pool. Given enough time and interbreeding, it could eventually become universal. Because it requires time for the sequence to become so pervasive, the degree to which it is found in a population would yield clues as to when it originated.
As for the sense-of-smell thing, I can offer as anecdotal evidence, individuals I know who could tell you exactly what brand of shampoo you used the last time you washed your hair, just by sniffing. It would seem that there are still individuals with rather acute, if not quite bloodhound-like, senses of smell.

11

I don’t understand this. I understand that, in absolute numbers, the amount of people with this gene will increase. But why would the percentage of the population that has these genes not remain constant?

12

There’s also another possiblility. If a particular trait is in competition with a different one, the more beneficial will win. Applicably, there’s a theory that in our evolution our proto-human ancestors enjoyed an aquatic phase where they under went several adaptive changes to include hair loss and straightening of the spinal column. Also was a general streamlining of the body to include the down-turning of the nostrils. Perhaps at this point since being able to swim quickly was more important than being able to smell a genetic mutation may have taken the opportunity to become a dominant trait.

13

Because people intermarry with other people, so that, eventually, everyone’s related and has similar genes.

14

inertia

Your post seems to imply that pseudogenes are genes that can control two or more functions, and that the mutations merely cause them to adopt a more beneficial one. This is not the case, according to what I’ve been told, and in fact were it so, it would defeat the whole argument being made from them.

toadspittle

I don’t think what you are saying is true. Unless you mean that all these mutatitions that cause pseudogenes are dominant, in that anyone with one parent with such a mutation will have them. Is this the case?

15

IzzyR, I’ll explain everything when I get the chance. You’ve asked an excellent question, but in order to do it justice I’ll have to give you a thumbnail description of some basic evolutionary biology which I can’t explain off the cuff.

-Ben

16

First off, bear in mind that if a gene (or pseudogene) is characteristic of a species, what’s happened is that it’s become “fixed” by virtue of being present in all members of the population. Take human hemoglobin, for example. Since all humans have human hemoglobin genes, then matings between humans will always produce children with human hemoglobin genes.

I realize that this sounds like stating the obvious, but once you phrase it that way you realize that one good question to ask about evolutionary biology is, “How do new genes end up being present in all members of the population, so that they become part of the genome of the species?” In order to ask this question, you have to start from highschool Mendelian genetics. If there is a gene for brown hair (B) and a gene for blond hair (b), then in high school you learned how to determine what percentage of offspring would be BB, Bb, or bb given parents that were BB, Bb, etc.

Since we’re talking about populations, you can mathematically extend the concepts of Mendelian genetics to deal with a large population, rather than just two parents. Thus if you know that 10% of the parents are BB, 32% are Bb, and the remainder are bb, and they mate more or less at random, you can calculate the percentages of offspring that will have each combination of genes. These equations are called the Hardy-Weinberg equations, and over several generations the population will reach the Hardy-Weinberg equilibrium, where the percentages of parents with each combination of genes equals the percentages of offspring with each combination.

Now, the Hardy-Weinberg equations, in and of themselves, are a little simplistic. For one thing, you have issues of genetic crossover, which I mentioned in the FAQ. If you want to calculate the percentage of offspring in each generation with two brown hair genes and two blue eye genes, you have to take into account the fact that those two genes might be on the same chromosome, and thus they will tend to be passed on together. (Conversely, genetic crossover will disrupt this linkage, so you can add another correction factor. Bear in mind that these corrections aren’t some post hoc fudge factor to make the facts fit evolution. We know that genetic crossover really happens and we can experimentally measure the value of the correction. We also know that the principles of Mendelian genetics upon which the HW equations are based are well established by experiment. In the end, we need to include extra correctional factors in the HW equations to make them describe reality in detail. You could draw an analogy to the equations for motion: the base HW equations are like Newton’s laws, and the factors like crossover are like Einstein’s more accurate corrections which come from Relativity.)

One mathematical correction which must be added involves statistical fluctuations. In practice, some couples will have more children than others, or freak accidents will, purely at random, wipe out slightly more BB children than Bb children, etc. These statistical fluctuations are more extreme for small populations, and the math for treating them was well-known from statistics before it was applied to population dynamics.

Well, I’m afraid I have to leave the computer right now. I’ll finish the argument later.

-Ben

17

Now, as far as our discussion is concerned, we’re not so much worried about the number of BB vs. Bb vs. bb offspring so much as we’re concerned with the percentage of genes in the population that are B vs. b. Let’s suppose we start with a population that is 50% B and 50% b. (For example, 25% BB, 50% Bb, 25% bb.) All things being equal, the next generation will also be 50%B, 50%b. But if a statistical fluctuation causes one gene to be present in an excess of 1% of the frequency of the gene in the parental population, then the next generation is 50.5%B, 49.5%b.

In the next generation, a statistical fluctuation can happen again- let’s suppose it’s 1% again. If it happens to B, then 1% of 50.5% is .505%. But if it happens to b, then 1% of 49.5% is .495%. You can see that the statistical fluctuations favor the majority gene, and thus the effect can snowball until 100% of the genes are B. This means that b has been bred out of the population, and the B gene is now considered a part of the standard equipment of the species. This process is called genetic drift.

At this point, it’s probably occurred to you that if a single individual in the population gets a beneficial mutation, the snowball effect of all the individuals who don’t have the mutation could breed the beneficial mutation out of existence. In the short term, this does happen: maybe none of that mutant’s offspring inherited the chromosome with that gene on it, or maybe a string of bad luck in successive generations of the family wiped them out. But in the long term, remember that the beneficial effect of the mutation causes its bearers to have more offspring than average. If bearers of the mutation have 2% more offspring than average, that will fight against the fluctuations and snowball the beneficial mutation until, several generations down the line, it’s present in all members of the species. Even in the case of selectively neutral mutations, the snowball effect can amplify a single mutation until it takes over the entire population. (Remember, if the gene has a string of good luck in which it is favored by the fluctuations, it can rally against the other gene.)

As far as pseudogenes go, let’s talk about the vitamin C synthase gene. Once the primate ancestor was eating a fruit-rich diet, there was no need for that species to synthesise its own vitamin C. Inactivating mutations could occur in the synthase gene, but they would be neutral. Although they might be occasionally bred out by statistical fluctuations, they wouldn’t be actively bred out by deleterious effects. Over time, given the number of generations involved and the number of individuals in which mutations can occur, sooner or later some of the mutations will become fixed in the population, and the gene will become a pseudogene.

With regard to the odorant receptors, I’m assuming that when the human ancestors started depending on sight more than smell, a larger portion of the brain was devoted to vision, at the expense of olfaction. At that point much of the vast array of odorant binding proteins would serve no fuction, since the neural connections in the olfactory glomeruli would no longer be present for processing the information gathered by the receptors. At that point mutations in the receptors would be neutral, and would be fixed by genetic drift, turning the receptor genes into pseudogenes.

-Ben

18

According to an article on artificial sight in last month’s Discover, for human beings to have eyesight as keen as an eagle’s, our eyes would have to be as big as tennis balls (to accomodate bigger retinas and lenses, I guess). I’m also guessing we’d need a larger visual cortex in the brain. And the brain already uses more food and oxygen per ounce than any other organ.

19

Ben

I have two problems with applying “genetic drift” as you describe it to pseudogenes (neutral mutations).

  1. To the same extent that genetic drift would favor a gene that was present in more then 50% of the population, it would disadvantage a gene that was present in less than 50%. This means that while a pseudogene/neutral mutation would benefit from this phenomenon once it already achieved 50%, it would be “harmed” by that same factor before this point. So this force would cancel itself out overall, and the likelihood of a pseudogene becoming universal would remain the same as if this process did not exist.

  2. The very concept of genetic drift as you describe it does not appear to be valid for pseudogenes. The reason for this is that you have ignored the fact that the larger the population is the less likely it is to deviate from the norm by a given percentage. This will cancel out the mathematical process that you described.

Consider the following illustration: Imagine that there are three groups, A, B, & C. Each one has exactly the same number of people, and none has a survival advantage over the other. Population growth may fluctuate randomly. Clearly one cannot say that genetic drift would cause any one of these three group to increase in proportion.

Now suppose that Scientist X has decided that for his purposes he will classify A and B together as Group AB, contrasted to Group C. What will happen to these populations? Well, using the process that you described, one might think that Group AB will increase in proportion, as they are twice the size as Group C. If they, by statistical fluctuation, increase in a given year by 1% more than Group C, and the next year Group C increases (again, by statistical fluctuation) 1% more than Group AB, Group AB will have increased its overall proportion. This contradicts our initial assumption that the populations will remain stable. Further, suppose Scientist Y, studying these same groups has decided that for his purposes he will break down the groups as A and BC, the exact same genetic drift process would tell him that Group A will decrease its proportion. Paradox!

The answer is simply that it is not reasonable to assume that a statistical fluctuation will produce the same margin change in the larger group as it does in the smaller groups. As the population grows the variance becomes smaller, and the deviation is less. This compensates for the process you described.

Genetic drift would applicable if the “statistical fluctuations” were not that, i.e. if they were actual survival advantages. Suppose 50% of the time there was a survival advantage which favored the larger group and 50% it favored the smaller group. In this instance, you would be right in that eventually the larger group would dominate. This would not violate the statistical laws, as the deviation of the larger group is not a statistical fluctuation and is not governed by such rules. But the point of pseudogenes is that they involve mutations that do not affect survival, making any differences in survival rates random. As such the process you describe in not applicable.

The only thing left, as I see it, is that a given neutral mutation could simply “get lucky” and survive and prevail against the odds until it successfully “snowballs” into total dominance. But these odds (which would have to be enormous, unless the species population was tiny at the time of the mutation) would have to be factored in to calculations involving mutation rates of pseudogenes.

jab1

I’m unsure if you intended your post in connection to the OP, or just a general comment to ultrafilter. If the former, I’ve failed to understand it, and wonder if you could explain what manner of mutation you might be referring to.

20

I was just pointing out that if we had bigger, better eyes, the cost would be either a larger skull or fewer brain cells devoted to the other senses or to reason. (Ask any mother if she wished her baby had had a bigger head…)