The Rise of Human Chromosome 2: Beyond the Deme

| 52 Comments

This essay is the third of a series authored by Dave Wisker, Graduate Student in Molecular Ecology at the University of Central Missouri.

This series of essays counters common Intelligent Design/Creationist arguments against the fixation of the fusion that produced human chromosome 2. To recap briefly, ID/C’s argue that the fusion would have resulted in a chromosome with two centromeres, which would then be torn apart during meiosis. In “The Dicentric Problem” I explained how the presence of two centromeres does not necessarily interfere with proper segregation. ID/C’s also claim that heterozygotes for the fusion suffer from greatly reduced fertility, preventing the fusion from ever becoming fixed. I explained how this is not true for centric fusions in “The Fertility Problem” . Using realistic values for human populations, “Fixation Within a Deme” showed that the fixation probability of the fusion within a local breeding population, or deme, may be between 4.5 and 10 percent. This final essay will look at the probability of fixation for the fusion in the species as a whole.

Lande (1979) looked at the problem of fixation for chromosomal rearrangements with heterozygote disadvantage. He constructed a model with populations subdivided into numerous, semi-isolated subpopulations, or demes, in which spontaneous chromosome rearrangements with heterozygote disadvantage occur. In addition, the model included random extinction and colonization of demes. Lande did not include mechanisms like meiotic drive (see “Fixation Within a Deme” for explanation) in his model, letting genetic drift (i.e., chance) be the primary force for fixation within a deme. He found that the biggest obstacle to overall fixation was establishment of the fusion within one deme. The overall fixation probability turned out to be the probability of fixation within that first deme:

Once established in a deme, a negatively heterotic [one with a selective disadvantage for heterozygotes–DW] gene arrangement can spread in homozygous form through a subdivided population by random local extinction and colonization. For this process, the fixation rate in a species composed of many semi-isolated demes is approximately equal to the rate of establishment of new gene arrangements in a single one of its demes.

In “Fixation Within a Deme”, I showed that meiotic drive plays a major role in influencing fixation of centric fusions in humans. In addition, the probable structure of early human populations is very similar to that used in Lande’s model. It is therefore reasonable to suggest the overall fixation probability for the human chromosome 2 fusion is at least the value of 4.5 -10% determined earlier.

Interestingly, there may be another factor that influenced the rise in frequency of human chromosome 2. Hedrick and Levin (1984) suggested a mechanism that greatly increases the probability of fixation for negatively heterotic chromosome rearrangements during migration and colonization of other demes. They pointed out that the process of colonization often involves small groups of individuals, so genetic drift in the form of the “founder effect” plays an important role. Furthermore, the individuals in these founder groups may be closely related. So, if a chromosome rearrangement becomes established in a deme, then any founder groups derived from it will already have the rearrangement at a very high frequency to start with. If the group establishes itself in an unoccupied area, fixation would be instantaneous for that deme. If the group joins another existing group, given that most demes are small, the newcomers may compose a significant percentage of the combined deme. Therefore, the initial frequency of the rearrangement could already be high enough to exceed the unstable equilibrium frequency (see “Fixation Within a Deme” for discussion of this) and thus drive the rearrangement to fixation there as well. Hedrick and Levin called this “kin-founding”. They note that kin-founding is common in many groups of organisms. Anthropological data confirms this in humans (Fix, 1978). Since Lande’s model does not incorporate mechanisms that influence fixation such as meiotic drive and kin-founding, there is reason to believe that the overall fixation probability for human chromosome 2 may be higher than the base calculation for the deme.

In conclusion, this series of essays effectively counters every ID/C claim about the probability of fixation of human chromosome 2, using actual data from human populations as well as population models based on plausible assumptions. It is up to the ID/C community to present data that specifically supports their contentions.

Many thanks to Art Hunt for discussions and publishing help, and the PT crew for encouragement. Many thanks also to the ID proponent ‘Ilion’ (wherever he may be) for getting me mad enough to actually put it all together.

References

Fix AG (1978). The role of kin-structured migration in genetic microdifferentiation. Ann. Hum. Genet. Lond. 41: 329-339.

Hedrick PW & DA Levin (1984). Kin-founding and the fixation of chromosomal variants. Amer. Nat. 124(6): 789-787

Lande R (1979). Effective deme sizes during long-term evolution estimated from rates of chromosomal rearrangement. Evolution 33(1): 234-251.

52 Comments

Perhaps this is a really stupid question, but I’ve sometimes heard the argument from Creationists that the fusion event happened after God created man in the last 6,000 years, and quite simply man used to have the same number of chromosomes as Apes and it looked similar because we are relatively similar animals. I’m pretty sure that the mutational erosion on the telomeres basically proves this fusion happened millions of years ago, but can you properly/scientifically address this claim? I hate seeing Creationists throw it around without a complete answer.

The finding that the probability of fixation in the whole population is roughly equal to the probability of fixation in one deme does not seem intuitively obvious. Indeed, it seems counter intuitive.

If there is no advantage to having the fused chromosome over having two separate ones, and if only one deme out of a meta population of several demes has fixed the fused chromosome, then it seems to me the probability of fixation of the fused form in the whole population from that point on (i.e., once it became fixed in one deme) would be proportional to the 1/(# of demes).

Am I missing something?

Many thanks also to the ID proponent ‘Ilion’ (wherever he may be) for getting me mad enough to actually put it all together.

Sure Ilion wasn’t a ‘ras sock?

Ken Wrote:

I’m pretty sure that the mutational erosion on the telomeres basically proves this fusion happened millions of years ago, but can you properly/scientifically address this claim? I hate seeing Creationists throw it around without a complete answer.

While I hope others will give you a more detailed answer, I recall reading somewhere that the fusion event was ~1 MY ago, IOW, much closer to the present than to the chimp-human split (~6 MY ago). That would also be before the modern human - neanderthal split, but I don’t recall reading that specifically.

Don’t forget, though, that IDers (the DI and their followers) would either agree that the event occurred ~ 1 MY ago or, just play dumb about when it occurred. IOW nothing they would say about it would give any hope to YECs.

That brings up a question that I have for those with more knowledge of the topic. The fusion would be an ideal candidate for “designer intervention.” At least to please OECs who crave something special about H. sapiens. Has any IDer proposed that specifically, or are they still playing it safe suggesting that the “blessed events” occurred in the Cambrian or before?

Divalent -

If there were a “n” separate, non-interactive demes, and a probability “p” that each individual deme would fix the chromosome, the probability of fixation in the population would be p^n (p to the power n).

However, that is not what is being proposed -

They pointed out that the process of colonization often involves small groups of individuals, so genetic drift in the form of the “founder effect” plays an important role.

In a small population, a particular allele, broadly defined, may well reach a high frequency in the population due to random events. This can be proven mathematically and is also highly intuitive. This is traditionally called “random drift”. Here, the “allele” in question is actually the fused chromosome.

If that small population is also a founder population - the ancestors of a future much larger population the allele may well remain at the same very high frequency as the population grows. (The fact that they become a founder population implies that the high frequency allele is not very deleterious, or they would not be a successful founder population to begin with.)

Furthermore, the individuals in these founder groups may be closely related. So, if a chromosome rearrangement becomes established in a deme, then any founder groups derived from it will already have the rearrangement at a very high frequency to start with.

If a group branches off after the high frequency of the allele has become established, that branching group will carry a high frequency of the chromosome with them.

If the group establishes itself in an unoccupied area, fixation would be instantaneous for that deme.

Duh.

If the group joins another existing group, given that most demes are small, the newcomers may compose a significant percentage of the combined deme.

Those who branched off may encounter other groups of the same species, in whom the fused chromosome in question is not found or not at a high frequency. However, the newcomers themselves have it at a high frequency, and the newcomer group could be similar in size to or larger than the outside groups that they encounter.

Therefore, the initial frequency of the rearrangement could already be high enough to exceed the unstable equilibrium frequency (see “Fixation Within a Deme” for discussion of this) and thus drive the rearrangement to fixation there as well.

They could bring it in to the new group at such a high frequency that, based on a mathematical model, it would be the expectation that it would become fixed (present in virtually 100% of individuals) in the new group as well over time.

Notes -

1) All of this assumes that the human form of chromosome 2 was not even positively selected for to any serious degree in early populations.

2) Long before the discovery of chromosomes, let alone DNA, rational people noted that apes were especially related to humans. Such observations date back to ancient times.

The theory of evolution, even in its earliest form, has always predicted that, when the genetics of humans and chimpanzees could be studied, they would be very similar, possibly with actual evidence of population differentiation events at points in the past.

With each level of genetics that we understand, from early classical genetics to chromosome morphology studies to molecular biology, that prediction is fulfilled. This is devastating blow to any form of creationism.

The only form of creationism that can account for this is non-disprovable but irrelevant “last Thurdayism” - essentially, some variant of the idea that the Flying Spaghetti Monster or some other deity created the world last Thursday, giving us false memories and setting it up to look exactly as if humans and chimpanzees share extensive genetic similarity and a recent common ancestor. That’s what Casey Luskin-level squealing about “common design” amounts to - an admission that it “looks like” evolution and a claim that it “could have been” magic done to perfectly mimic evolution.

I’m thinking that this is all well and good, but of course IDers and out-and-out creationists are not really making scientific arguments, and will not be (at least for the vast majority) swayed by your discussion. They aren’t really arguing that this is a problem, they are just looking for anything that could possibly cast doubt on the science and reinforce their preconceptions.

Remember, they have already made up their minds. Their arguments are not really intended to start a dialogue (as would be the case if they actually did science). They are intended for the rubes to fall back on, and for them to point out later that they had (of course) devastatingly demolished the evolutionists’ arguments (even if their own arguments have been proven false).

Reprod Biol. 2007 Jul;7(2):143-62.(pdf) Links

Gonadal expression of aromatase and estrogen receptor alpha genes in two races of Tunisian mice and their hypofertile hybrids.Saïd L, Delalande C, Britton-Davidian J, Saïd K, Saad A, Carreau S. Biochemistry Laboratory, University, Caen, France.

House mice (Mus musculus domesticus) in Tunisia consists of two races, one carries the 40-acrocentric standard karyotypes and the other one is a robertsonian race (2n=22) homozygous for nine centric fusions (Rb). The F1 hybrids between the two chromosomal races showed a significant decrease in reproductive success and litter size. Such results can be related to the formation of meiotic trivalent in the hybrids leading to the production of viable aneuploid gametes and post-zygotic elimination of embryos due to chromosomal non disjunction events at meiosis. Moreover, testicular histology of F1 and backcross males showed in some cases a breakdown in spermatogenesis. In both females and males, androgens but also estrogens play an important role in gametogenesis. In this study, we have studied aromatase and estrogen receptor alpha (ERalpha) gene expression in the gonads of the two parental races and their chromosomal hybrids. The results showed that aromatase and ERalpha mRNAs are expressed in hybrid males of inter-racial crosses (female22Rb x male40Std and female40Std x male22Rb) and in hybrid females of inter-racial crosses (female22Rb x male40Std) as in the two parental races. However, in hybrid females of inter-racial crosses (female40Std x male22Rb) the amount of aromatase transcripts decreased sharply suggesting that this gene is involved in the breakdown of hybrid fertility in females, but not in males. However, in hybrid males, a putative post-translational modification of this enzyme, in terms of activity, should be verified.

Karyotype differences in a species are not unknown. Mice show this frequently with 6 different karyotype populations on Madeira, all from a founder event maybe 1,000 years ago.

Matings between different chromosomal numbers are still fertile. Having different chromosome numbers isn’t much of a reproductive barrier.

In Tunisia, one group has 40 chromosomes, another 22. A huge difference. They are still interfertile although fertility between groups is lower. This is thought to be beginning speciation.

GvlGeologist, FCD Wrote:

They aren’t really arguing that this is a problem, they are just looking for anything that could possibly cast doubt on the science and reinforce their preconceptions.

Ironically almost any evidence for evolution (or any scientific explanation) can be a source of another “weakness” if taken out of context. Not unlike how every transitional fossil turns one “gap” into two.

As you know, articles like these are for the benefit of those who truly want to learn. Occasionally that includes creationists who change their mind. For the rest it’s more fun to put the ball in their court and ask what the designer did, when and how. They can’t impress the rubes as well when they’re evading the very questions that the rubes want answered.

GvlGeologist FCD -

Remember, they have already made up their minds. Their arguments are not really intended to start a dialogue (as would be the case if they actually did science). They are intended for the rubes to fall back on, and for them to point out later that they had (of course) devastatingly demolished the evolutionists’ arguments (even if their own arguments have been proven false).

Yes, but -

1) Some small proportion of the rubes gradually begin to see the light when repeatedly exposed to it.

2) Third parties like Judge Jones and others are well-served by strong rebuttals of creationist/ID arguments.

No-one is arguing with Dembski or Luskin to convince Dembski or Luskin. Indeed, it’s debatable that words like “evidence”, “reasoning”, or “convince” even have meaning in the context of minds that biased by various factors.

The point is to prevent them from being the only voice that third parties hear.

GvlGeologist, FCD said: Their arguments…are intended for the rubes to fall back on, and for them to point out later that they had (of course) devastatingly demolished the evolutionists’ arguments (even if their own arguments have been proven false).

That’s why it’s important to emphasize to creationists that the “Signature In The Cell” that they claim proves intelligent design creationism cannot be proven to be solely that of Jehovah/Yahweh but can equally be “proven” to be the “signature” of Wotan/Odin, Jupiter/Zeus Pater, Amun-Ra, Brahma, Ometecuhtli, Chthulhu or the Flying Spaghetti Monster.

Great timing with this post as I’ve been debating Chromosome 2 with someone over at UD and this information is very helpful.

I have just one question you might be able to answer. If a hominid with 48 chromosomes mated with a hominid with only 46 chromosomes because two were fused, wouldn’t the two separate chromosomes from the first parent recombine with the fused chromosome of the second parent leaving the offspring with only 46 chromosomes? Would this lead to the change being fixed in the deme even faster than with other mutations?

I have a question similar to that of Camanintx but related to Equus.

Przewalski’s Horse (Equus ferus przewalskii), a.k.a. Wild Horses, have 66 chromosomes.

The Tarpan or Eurasian Wild Horse (Equus ferus ferus) have 64 Chromosomes.

The Domestic horse (Equus ferus caballus) has 64 chromosomes.

The donkey (Equus asinus) has 62 chromosomes.

The cross of a domestic horse and a donkey produces a mule or hinney with 63 chromosomes. Crossing a wild horse with a domestic horse produces a horse with 65 chromosomes.

Have the Equus genomes been studied in the same manner than human genomes have been studied in order to determine whether or not there are extra centromeres and/or telomeres?

Thanks,

Ed

Sorry if this is a duplicate, but my last comment seems to have not been posted.

Wild Horses (Equus ferus przewalskii) have 66 chromosomes. Domestic horses (Equus caballus) and have 64 chromosomes, and Donkeys (Equues asinus) have 62 chromosomes. The cross of a domestic horse and a donkey produces a mule or hinney with 63 chromosomes. Crossing a wild horse with a domestic horse produces a horse with 65 chromosomes.

Have any scientists studied the Equus genome in the same manner that scientists have studied the human and ape genome? Do we know whether or not there are extra centromeres and telomeres in the Equus species of lower chromosome number due to fusion of chromosome pairs?

camanintx said:

Great timing with this post as I’ve been debating Chromosome 2 with someone over at UD and this information is very helpful.

I have just one question you might be able to answer. If a hominid with 48 chromosomes mated with a hominid with only 46 chromosomes because two were fused, wouldn’t the two separate chromosomes from the first parent recombine with the fused chromosome of the second parent leaving the offspring with only 46 chromosomes? Would this lead to the change being fixed in the deme even faster than with other mutations?

A hybrid offspring between these two would have 47 chromosomes. The two free chromosomes would try to align with their homologous areas on the fused chromosome. l

INTELLIGENT DESIGN. YOU ARE STUPID IF YOU THINK THAT NATURE IS NOT INTELLIGENT.

Harold, I don’t think you see the problem. The statement “The overall fixation probability turned out to be the probability of fixation within that first deme …” implies (if I am interpreting it correctly) that once a new allele arises in one deme, then it is virtually certain to become fixed in the global population of all demes (since if the probability of A *then* B occurring is the same as the probability of A occurring, then the probability of B occurring once A has occurred is 1.0).

But if there is no advantage to having either form of the chromosome(s) (i.e., fused or separate), then this is not really different from the case of two equivalent alleles, and the deme aspect of it really doesn’t change the basic process.

All of the things you noted (the deme with the new allele could split and form a new deme by colonization, or a lot of them could migrate into an existing deme with few members, etc) would be just as likely to occur for demes with the old allele; and there are many more of them. Why couldn’t you look at any one (of the many) demes that have the old allele and use the same logic to say that it’s allele is virtually certain to become fixed in the population?

Divalent said:

Harold, I don’t think you see the problem. The statement “The overall fixation probability turned out to be the probability of fixation within that first deme …” implies (if I am interpreting it correctly) that once a new allele arises in one deme, then it is virtually certain to become fixed in the global population of all demes (since if the probability of A *then* B occurring is the same as the probability of A occurring, then the probability of B occurring once A has occurred is 1.0).

But if there is no advantage to having either form of the chromosome(s) (i.e., fused or separate), then this is not really different from the case of two equivalent alleles, and the deme aspect of it really doesn’t change the basic process.

All of the things you noted (the deme with the new allele could split and form a new deme by colonization, or a lot of them could migrate into an existing deme with few members, etc) would be just as likely to occur for demes with the old allele; and there are many more of them. Why couldn’t you look at any one (of the many) demes that have the old allele and use the same logic to say that it’s allele is virtually certain to become fixed in the population?

Divalent,

I wasn’t implying that the probability of eventual fixation within the entire population was certain if it became fixed within one deme. I was interpreting Lande to say the overall probability of fixation was essentially the same probability of the fusion becoming fixed in one deme (i.e. 4.5%-10%). Looking back through that paper, I may have been wrong. Your suggestion of 1/N sounds more reasonable for a purely random process, barring phenomena that skew fixation probability.

Paul Burnett Wrote:

That’s why it’s important to emphasize to creationists that the “Signature In The Cell” that they claim proves intelligent design creationism cannot be proven to be solely that of Jehovah/Yahweh but can equally be “proven” to be the “signature” of Wotan/Odin, Jupiter/Zeus Pater, Amun-Ra, Brahma, Ometecuhtli, Chthulhu or the Flying Spaghetti Monster.

You forgot Satan. What it does not provide, and wouldn’t even if it were legitimate evidence of a designer is nothing that would validate any of the mutually contradictory origins stories that Biblical literalists who desperately crave.

John Dingleberry said:

INTELLIGENT DESIGN. YOU ARE STUPID IF YOU THINK THAT NATURE IS NOT INTELLIGENT.

So close, yet so far.

ID - the DI’s particular version - argues that Nature is not intelligent enough (to build a flagellum, etc.). So that “something else” is needed.

So while we take the bait and argue among ourselves who they mean by the designer, or whether that designer exists, the rubes incorrectly think that ID validates their particular fairy tale.

harold said:

No-one is arguing with Dembski or Luskin to convince Dembski or Luskin. Indeed, it’s debatable that words like “evidence”, “reasoning”, or “convince” even have meaning in the context of minds that biased by various factors.

The point is to prevent them from being the only voice that third parties hear.

Amen to that brother! As an educator who deals daily with young people being brainwashed by fundamentalist nonsense, I can’t tell you how important it is to keep addressing the nonsense out loud and in public this way. If nothing else, it gives me a place to send the kids who have never even suspected that evolution is taken seriously by anyone.

Divalent -

I believe I must have misunderstood your question.

All of the things you noted (the deme with the new allele could split and form a new deme by colonization, or a lot of them could migrate into an existing deme with few members, etc) would be just as likely to occur for demes with the old allele;

That’s true; my understanding was not that the author was implying that as soon as it arose, it was inevitable that the fused chromosome would become fixed. To the best of my knowledge, if we’re assuming neutral selection pressure, that could never be true of any new allele in any population of greater than one or two organisms.

Rather, I thought the authors were showing how it was quite possible for it to become fixed, even if we don’t assume positive selective pressure.

and there are many more of them. Why couldn’t you look at any one (of the many) demes that have the old allele and use the same logic to say that it’s allele is virtually certain to become fixed in the population?

Again, my understanding is that in circumstances of near neutral selection and small population sizes, it is indeed true that any of multiple “competing” alleles might become fixed.

My understanding is that the point is not that “as soon a fusion chromosome arose, it was inevitable from that point forward that it would become fixed”.

Rather, my understanding is that the point is that “we see that it did become fixed, and we have a perfectly good explanation as to how that may have happened”.

If I roll a die, and it comes up 6, I have a good scientific explanation as to how that happened. That doesn’t mean that it was inevitable that it would come up 6 a priori.

John Dingleberry -

INTELLIGENT DESIGN. YOU ARE STUPID IF YOU THINK THAT NATURE IS NOT INTELLIGENT.

You’ve been told this once, but it’s worth repeating. The whole point of “intelligent design” is to claim that nature is insufficient to explain things like human evolution, and that magic is required.

Divalent -

Also,

Harold, I don’t think you see the problem. The statement “The overall fixation probability turned out to be the probability of fixation within that first deme …” implies (if I am interpreting it correctly) that once a new allele arises in one deme, then it is virtually certain to become fixed in the global population of all demes (since if the probability of A *then* B occurring is the same as the probability of A occurring, then the probability of B occurring once A has occurred is 1.0).

That is true if you interpret the statement that way.

That is not the same thing as saying that it was inevitable that it would become fixed in the first deme.

Another way of interpreting the statement might be that the mentioned probability of it being fixed in the population is the conditional probability of that, given it being fixed in the first deme in the first place.

Then the total a priori probability of fixation in the population would be p^2.

At this point, I’m going to stop here, because I haven’t read the original carefully enough to make very specific comments about the math. Sometimes population genetics can lead to surprising results.

My current take-away message is that it is NOT shocking that such a fusion could become fixed in a small population, and that no magic is required for such an event to have taken place.

Another take-away message is that it takes thought and effort to understand science

We’ve seen evolution of multiple karyotype demes in historical times. Mice on Madeira in 1,000 years have formed 6 different karyotype subpopulations. The authors claim it is not due to adaptation but genetic drift and geographic isolation.

Anyone who wants to say changes in chromosome number is unusual or hard during evolution is wrong. We see it happening often.

Nature 403, 158 (13 January 2000) | doi:10.1038/35003116

Janice Britton-Davidian1, Josette Catalan1, Maria da Graça Ramalhinho2, Guila Ganem1, Jean-Christophe Auffray1, Ruben Capela3, Manuel Biscoito4, Jeremy B. Searle5 & Maria da Luz Mathias6

AbstractMadeira is a small volcanic island in the Atlantic Ocean with steep mountains separating narrow valleys that are the only areas habitable by humans and their commensals. Here we show that house mice (Mus musculus domesticus ) on Madeira have an unexpected chromosomal diversity, the evolution of which is independent of adaptive processes, relying instead on geographic isolation and genetic drift.

Dave Wisker -

Thank you, by the way.

Raven -

It’s unbelievable how much we learn from mice.

actually the punishment for blasphemy is DEATH…

http://www.facebook.com/topic.php?t[…]=21080812784

see you deluded, fuc*.…

I am sending this via email to the entire university - faculty and stuff. Your blaspheming head will serve as example and warning to the whole place…

Sorry if this is a dumb question. I don’t have any experience in studying biology, so a lot of the “demes” and “negatively heterotic chromosome” terms are whizzing over my head. I’m having a tough time wrapping my head around this one. I get the notion that point mutations, even frame shifts can occur frequently. These seem like “easy” mutations. I get that these can potentially be passed on. But in order to get “fixed” in a population, unless for some reason the mutation is “dominant” (in some sense?) don’t such mutations have to occur more than once? Or, is it sufficient in a sufficiently small population that such a mutation only has to occur once, and with enough inbreeding you eventually get a pair with the right recessive mutation to pass on the full blown mutation?

But the merging or splitting of whole genes? That sounds too “difficult” in some sense to successfully occur very often. Even if a merge or split were to occur again, wouldn’t it have to occur at the same point in the same gene(s) in order to “match” in some sense? This difficulty in getting “matching” mutations (if my intuition is anywhere near accurate) would suggest that a given merge or split must have happened only a single (successful) time. This single mutation was then passed on, and through fortuitous inbreeding became “fixed” in some isolated lineage.

Without the jargon, am I on the right track here? And this essay is (perhaps) pointing out that the probabilities of “single” mutations (even “difficult” ones like gene fusion) getting “fixed” in a population are better than earlier predicted or popularly appreciated?

This really isn’t a troll question. I was hoping for some edification (I’ve been following PT for quite some time with great enjoyment), but I’m afraid I got bogged down in the jargon on this one, and didn’t get as much out of this essay as I had hoped to.

Thanks for your help and patience.

Scott said:

Sorry if this is a dumb question. I don’t have any experience in studying biology, so a lot of the “demes” and “negatively heterotic chromosome” terms are whizzing over my head. I’m having a tough time wrapping my head around this one.

Scott, it’s not a dumb question at all. I know some of the jargon can be daunting, especially to someone with little background in biology. Unfortunately, the issues in this particular case involve two esoteric disciplines: cytogenetics (the study of chromosomes), and population genetics, both of which are difficult to explain without the specialized jargon. But that is often the problem in rebutting the ID movement’s claims: they oversimplify and dismiss evolutionary issues, and it takes much more work (and technical explanations) to correct. I’ll try to answer or at least point you in the right direction, but I recommend– for better understanding– that you get a good background in basic biology and genetics first. You will be amazed at how many of the ID movement’s claims you will be able to see through.

To answer your basic question, the mutation we are talking about here most likely occurred only once–for the reasons you bring up– and through chance became fixed in a small subpopulation (deme) and eventually throughout the population. Some types of mutations do recur, but this particular one is extremely rare. The basic purpose of my essays was to come up with a probability estimate for the fixation.

Without the jargon, am I on the right track here? And this essay is (perhaps) pointing out that the probabilities of “single” mutations (even “difficult” ones like gene fusion) getting “fixed” in a population are better than earlier predicted or popularly appreciated?

I’d say you are right on track. The only thing I would add is, the probability of fixation is far better than ID’s simplistic claim that it was practically impossible.

I hope this helps.

scott:

But the merging or splitting of whole genes? That sounds too “difficult” in some sense to successfully occur very often.

You are confusing genes and chromosomes in your questions. Genes are carried on chromosomes which can contain hundreds or thousands of them. But spread and fixation of a novel chromosome fusion or new mutation in a gene uses the same principles and mathematics.

Who says it happens all that often? In the human case between chimps and humans, there is one case, fusion of chromosome 2. We are looking at 6 million years of evolution and billions of humans. Not common, not impossible.

I would add is, the probability of fixation is far better than ID’s simplistic claim that it was practically impossible.

Is that what the DI says, never pay any attention to them? The facts in the real world say they are wrong. We’ve seen it happen 6 times on one small island, Madeira, out in the Atlantic in less than 1,000 years. We know the timeline because Madeira is in the middle of nowhere way out in the ocean and was only discovered and colonized in ca. 1400’s.

There are other examples in mice and a commenter pointed out above that wild horses and domestic horses show chromosomal number differences.

The DI doesn’t let easily known facts get in the way of their lies.

It’s a bit like saying it is impossible to send robots to Mars. Well no, it happens because we’ve done it twice.

Scott said -

Sorry if this is a dumb question. I don’t have any experience in studying biology, so a lot of the “demes” and “negatively heterotic chromosome” terms are whizzing over my head. I’m having a tough time wrapping my head around this one.

It’s nice to see someone actually admit this.

There is a lot to learn for anyone who wants to study any branch of science in depth.

(If you really want to get into it, a course in genetics might be of interest to you. That might require you to take courses in general bio, general chem, physics, and math as pre-requisites, depending on your background. I’m not trying to be difficult or discouraging; on the contrary. But I am being realistic.)

I get the notion that point mutations, even frame shifts can occur frequently.

Correct.

These seem like “easy” mutations.

For biochemical reasons, these types of mutations are probably among the most frequent, yes.

I get that these can potentially be passed on. But in order to get “fixed” in a population, unless for some reason the mutation is “dominant” (in some sense?) don’t such mutations have to occur more than once?

No. A mutation doesn’t have to occur more than once, nor necessarily be “dominant” relative to some other given allele, to be fixed.

“Dominant” refers to how an allele impacts on the expressed phenotype of an organism, relative to other alleles that could occupy the same locus.

(The terms “dominant” and “recessive” oversimplify mildly but are very helpful. Most alleles are close enough to being exactly dominant or exactly recessive in most contexts. Typically, if the effects of an allele are to “make something happen” in the phenotype, it will be dominant. It its effects are to “make something not happen” it will be recessive to those alleles that “make something happen”.

But if two dominant, or two recessive, alleles occur together in a heterozygote, interesting things can happen.)

Or, is it sufficient in a sufficiently small population that such a mutation only has to occur once, and with enough inbreeding you eventually get a pair with the right recessive mutation to pass on the full blown mutation?

Yes. This is true whether or not the allele in question is recessive or dominant, relative to some other allele. We know this because we see it happen, and also because the highly mathematical field of population genetics shows that, under rational assumptions, it can happen.

However, this does assume relatively neutral selection. If the mutation is question is severely deleterious, for example if homozygote embryos invariably die during development (to take an extreme example), it can never be fixed.

It can be deleterious in terms of relative reproductive advantage of homozygotes and still be fixed by chance in some circumstances, but the worse the reproductive success of homozygotes the less likely this will usually be.

But the merging or splitting of whole genes? That sounds too “difficult” in some sense to successfully occur very often.

Why guess when you can do the experiment or make the observation?

Mutation events, broadly defined, that involve chromosome structure, gene duplication, etc, do indeed happen at an observable frequency. Several examples of this have been presented.

Even if a merge or split were to occur again, wouldn’t it have to occur at the same point in the same gene(s) in order to “match” in some sense?

It isn’t clear what you mean here, but perhaps this will information will be useful -

Successful sexual reproduction is frequently possible between individuals who have some differences, within limits, in chromosome number and structure, gene copy number, etc. Several examples of this have been presented.

Significant differences in chromosome number and structure CAN create situation where viable offspring cannot be born, or where offspring are viable but sterile, even if the mating organisms in question have similar phenotype. But that doesn’t ALWAYS happen.

This difficulty in getting “matching” mutations (if my intuition is anywhere near accurate) would suggest that a given merge or split must have happened only a single (successful) time. This single mutation was then passed on, and through fortuitous inbreeding became “fixed” in some isolated lineage.

That’s about right, I believe.

Without the jargon, am I on the right track here? And this essay is (perhaps) pointing out that the probabilities of “single” mutations (even “difficult” ones like gene fusion) getting “fixed” in a population are better than earlier predicted or popularly appreciated?

I’m not sure what you mean by “earlier predicted” or “popularly appreciated”. I’m not sure whether any serious scientist ever predicted that this kind of event was impossible. Nor do I think that there is much “popular” opinion on this subject.

I would say that the point here is that this type of fusion and fixation can be explained to a large degree by what we know about genetics. No magic is required.

A broader point is that rational people always knew that apes and humans are similar. The theory of evolution always predicted recent common ancestry, even before any genetic evidence was available. All genetic evidence has subsequently supported the theory of evolution.

This really isn’t a troll question. I was hoping for some edification (I’ve been following PT for quite some time with great enjoyment), but I’m afraid I got bogged down in the jargon on this one, and didn’t get as much out of this essay as I had hoped to.

Modern genetics involves both a lot of statistics and probability, and a lot of understanding of molecular biology. It is a very interesting field, but it takes an effort to gain serious expertise.

I’m by no means an expert on genetics myself. I have a strong enough background to follow fairly high level discussions, and to make use of high level references if I see something I’m not familiar with. To have a really high level of understanding of contemporary genetics, one would have to have done graduate work in the field, or at a minimum, have an undergraduate degree in a biology or a closely related field with a specific focus on coursework in genetics.

I would also add that there are living species (which I believe Dave knows about more than I) that maintain karyotypic polymorphism within their populations. The reproductive barrier that IDCs like to tos out is not what it is made out to be.

Make that ‘toss’ out…

The cartoon in your first post is wrong as it shows a Robertsonian fusion in which one arm of the centromere is lost. This happens in cancer cells but is not viable. The fusion of chromoosome 2 is a tandem telemere to telemere fusion in which no coding dna was lost. Robertson studied insects (grasshoppers) and insects don’t to my knowledge have these tandem fusions. Also, mice have acentric chromosome fusions since all their chromosome are acentric or acentric fusions. This confusion stems from many biology texts and several papers. Tandem fusions have been found in other mammals, especially mujac deer.

Looking at the dead centromere of chromosome 2 in a genome browser you see two things. One, is that it is very small. the other is that it contains 3 VSA primate retrotransposons of which two are human specific. As centromeres are very dynamic and retrotransposons can be very disruptive one could speculate on perhaps a single cause for the fusion. Just speculation.

The fusion event is currently said to be between 1 and 6 million years ago. Not very accurate. I don’t know of any method to date rearrangements as yet. Fusion seems to have stopped the active repeats, inserts and rearrangements of the fused telemeres because in chimps the regions are larger in one of the chromosomes.

Bill Beaver -

Your basic point seems to be correct.

However, Robertsonian translocations can also be germline in humans, and can be compatible with normal phenotype, except that descendants are at risk of chromosome imbalance. Without meaning to offend Wikipedia haters, Wikipedia does have a somewhat decent summary of this, with a small but good quality reference section.

It is certainly true that they can also be seen in cancer cells, which often have many somatic mutations and chromosome abnormalities.

Bill Beaver said:

The cartoon in your first post is wrong as it shows a Robertsonian fusion in which one arm of the centromere is lost.

The cartoon was merely an illustration of how a dicentric chromsome could form. That is why I said it was a ‘similar’ situation, not an exact replica of the fusion itself.

Unrelated but just wondering if anyone has read/heard about this book: http://www.amazon.com/When-Faith-Sc[…]r=1-1#reader

It was written by a professor in my Geology department (University of Mississippi). He is really adamant about his faith (He doesn’t preach in class or anything but he does say up front what he believes). I was thinking of getting it but then I read the ‘Tenets’ in the first chapter. Any comments/opinions would be appreciated.

Case in point: okapis routinely have either 44, 45 or 46 chromosomes. Most of them are heterozygous (I’m probably misusing the term?) with one fused chromosome (I looked it up, it’s rob(4;26)) matched to two unfused chromosomes.

Bring this up when creationists claim that chromosomes can’t fuse.

Bill Beaver Wrote:

The fusion event is currently said to be between 1 and 6 million years ago. Not very accurate. I don’t know of any method to date rearrangements as yet. Fusion seems to have stopped the active repeats, inserts and rearrangements of the fused telemeres because in chimps the regions are larger in one of the chromosomes.

Pardon another dumb question from a chemist with limited knowledge of molecular biology, but I would expect that comparing the rates of telomere changes to fused vs. non-fused chromosomes could provide some estimate of the timing of the fusion. Did that help determine the 1-6 MY window? Does any future research look promising at narrowing the window?

Y. Fan, E. Linardopoulou, C. Friedman, E. Williams, and B. J. Trask

Genomic Structure and Evolution of the Ancestral Chromosome Fusion Site in 2q13-2q14.1 and Paralogous Regions on Other Human Chromosomes Genome Res 2002 12:1651-1662

This and it’s companion are good papers and about all I’ve found so far that deals vaguely with timing. It does give a good idea of the complexity of the issue.

Just found another source from UC Davis using biased substitutions dated 2007:

2.94 - .74 million years

www.soe.ucsc.edu/research/compbio/ubcs/ubcs-supp1c.pdf

Thanks Bill!

Let me guess that you have not found any similar publications from the anti-evolution community. I don’t mean armchair “critiques” of the hard work of “evolutionists,” but papers describing their own hard work and supporting their alternate explanation independently of any problems they might have with evolution.

but papers describing their own hard work

It’s not the amount of work, it’s the match (or lack of it) between their premise and the reality they claim to be describing. Without a close match, no amount of hard work would help.

Dave, Harold, Raven: I appreciate your responses. It does sound like my understanding of the essay was on the right track.

1) I’m afraid that stats was one class I didn’t do very well in. And re-taking that doesn’t sound like fun. :-) One thing that I did learn was that, like quantum mechanics, statistics often runs counter to “intuition”.

2) I was at first surprised that changes in the shape or number of chromosomes (which I did confuse with “genes”) would allow for successful matings to pass on the change. Thinking further, it only makes sense that some such matings should still be possible, otherwise certain observed speciations (or differences in DNA) would not have been possible.

3) And yes, to be “certain”, rather than just speculating, I should run the experiments myself. However, I have neither the time (to change careers again), the expertise, nor the equipment to do such. I’m more than happy to defer to (and learn from) the expertise of those who know better, if I can understand what they’re saying. :-)

4) Upon further reflection, it also seems that changes in the shape or number of chromosomes may not be all that “difficult” for successful matings, in that such changes only effect how the genes within them are packaged, rather than in how they are expressed. (Except for any side effects produced by changing the proximity of certain stretches of DNA.) If the gene duplication mechanisms are sufficiently flexible, they should be able to “patch up” differences between the parents’ chromosome packaging, to some extent.

Thanks for the clarifications!

And, yes, #4 is still an oversimplification. There are other complications with changes in the shape or number of chromosomes, such as changes in how much of a particular gene product gets expressed. But, certain kinds of changes to the mere morphology of chromosomes should not be as “difficult” to pass on as I first supposed.

Dave Wisker said:

Bill Beaver said:

The cartoon in your first post is wrong as it shows a Robertsonian fusion in which one arm of the centromere is lost.

The cartoon was merely an illustration of how a dicentric chromsome could form. That is why I said it was a ‘similar’ situation, not an exact replica of the fusion itself.

Mr. Wisker, I dunno, but its looks like you may be having a Dawkins’ moment. Tell me it isn’t so.

Don’t thing the general public needs to be subjected to more ‘Weasel’expositions to defend your ‘merely an illustration of how a decentric chromosome ‘could’ form’.

Thanks for this series. I haven’t had time to follow PT, but I read the first post and then following this was a must.

Scott said:

I’m afraid that stats was one class I didn’t do very well in. And re-taking that doesn’t sound like fun. :-) One thing that I did learn was that, like quantum mechanics, statistics often runs counter to “intuition”.

Yes, about that: I had a similar experience, mostly because the university had a genius prodigy as teacher for both probability theory and statistics. Asking a question would return an “I don’t get why you don’t get this” and the resulting ‘shut up and calculate’ approach wasn’t really helpful. (o.O)

It was compounded with that probability theory contains both trivial and technically difficult proofs and methods, which is guaranteed to blow up any problems or insecurities you have. For comparison, distribution theory, a somewhat related subject in that probability distributions bend to its methods, is mostly only difficult, which helps a lot. :-o

But, which was supposed to be my main point if I hadn’t gotten emotional on the subject years after the fact, it turns out that if you need statistics in your work it is a great way to catch up on the methods if not the theory. I don’t claim to be good at it, but today I’m observationally able to use it when needed. So there’s a way around re-taking classes. And still have fun.

Btw, on your claim I was going to say that more than statistics, basic probability theory runs counter to some intuitions. For much the same reason as QM: a priori and a posteriori situations differ qualitatively with respect to what observables mean. The Monty Hall type of problem comes to mind. [In QM, more so of course; it inherits all the conceptual difficulties of probability and add some of its own, such as that observables have no meaning, no existence as such, before observation.]

This is contrary to intuition about property constancy of “classical” objects. I think this thread and its subject shows some of that, see for example harold’s comments.

But then I reminded myself that there are tricky difficulties in statistics as well, regression towards the mean comes to mind.

Steve P. said:

Dave Wisker said:

Bill Beaver said:

The cartoon in your first post is wrong as it shows a Robertsonian fusion in which one arm of the centromere is lost.

The cartoon was merely an illustration of how a dicentric chromsome could form. That is why I said it was a ‘similar’ situation, not an exact replica of the fusion itself.

Mr. Wisker, I dunno, but its looks like you may be having a Dawkins’ moment. Tell me it isn’t so.

Don’t thing the general public needs to be subjected to more ‘Weasel’expositions to defend your ‘merely an illustration of how a decentric chromosome ‘could’ form’.

By Gum, I think you’re right. Would a Dembskian notpology be more appropriate?

Steve P. said:

Don’t thing the general public needs to be subjected to more ‘Weasel’expositions to defend your ‘merely an illustration of how a decentric chromosome ‘could’ form’.

I don’t get this.

First, the “Weasel” example (not ‘Weasel’) is doing exactly what is claimed in the book. It is creationists that argue that it is a) example of something other b) as an example of this ‘other’ not doing what it is ‘claimed’ to do. (Well, duh!)

Second, I have learned that showing existence of pathways is an essential methodology in biology, as you often can’t predict exactly which pathway taken out of a set of likely ones. This is not exclusive to biology btw, similar situations adheres to QM and indeed path integrals are a great tool for its theorists as well.

So if a biologist says, “look, this trait somehow traveled from city A to city B, because it’s parents lived all their life in city A but itself is living in city B” and goes on to say “I know this is a possible explanation for seeing this trait in B because, for example, one can travel by train from city A through city B, a train we know goes regularly, see the time table here” it doesn’t mean that the trait couldn’t have taken the detour train travel from A through city C to B, say to visit a girl friend.

Equally, an illustration of train travel from city A to city B may as well point to a day train schedule, for example because it’s less filled with confusing exemption symbols, even if the night train was the one actually used. This is exposing the science in a truthful manner for easy digestion.

The exact pathway taken doesn’t matter if the prediction to test is merely that a pathway exists.

Steve P. said:

Don’t think the general public needs to be subjected to …

I’m laughing at the idea that the “general public” would do any more than produce a blank stare if this particular subject were mentioned.

This is the kind of “focused detail” issue that is only of interest to the professionals, enthusiastic amateurs, and evobasher cranks misrepresenting the facts to muddy the waters.

Surely Steve P wasn’t being serious.

Bob O’H said:

Many thanks also to the ID proponent ‘Ilion’ (wherever he may be) for getting me mad enough to actually put it all together.

Sure Ilion wasn’t a ‘ras sock?

Pretty Sure…

derwood said:

Pretty Sure …

Ah, takes me back to my Corporate days: The difference between God and a programmer is that God does not think he is a programmer.

The difference between God and a programmer is that God does not think he is a programmer.

I wonder if I should resemble that remark!!1111!!!

About this Entry

This page contains a single entry by Arthur Hunt published on September 14, 2009 6:37 PM.

US Bidding War for ‘Creation?’ was the previous entry in this blog.

Harvard Darwin150 Webcast tonight is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Categories

Archives

Author Archives

Powered by Movable Type 4.361

Site Meter