|
Author
|
Topic: John A. Davison: An Evolutionary Manifesto: A New Hypothesis For Organic Change
|
nosivad
Member
Member # 767
|
posted 20. June 2003 19:10
I cannot reconstruct the past. My opinion is based on the view of Robert Broom largely who claimed a new mammalian genus had not appeared in the last 2 million years. Julian Huxley had corresponded with Broom as early as 1932 and was convinced that Broom was correct. The details of that I spelled out in the Manifesto. Here was Huxley, the major spokesperson for the Modern Synthesis, claiming in no uncertain terms, that macroevolution was finished. This on page 571 of his magnus opus, 7 pages from the end. I regard that as on of the most mystifying statements in all of the evolutionary literature. Grasse also has noted that evolution has apparently slowed markedly or even stopped and suggested that in the past different mechanisms for change may have been in operation. Of course I agree completely since that is the major thrust of the semi-meiotic hypothesis. I have suggested, as you know, that a primary effect of obligatory sexual reproduction is to bring evolution to a virtual standstill. Thus evolution, like growth, differentiation and many other biological phenomena is an autoregulatory and self-limiting process. nosivad
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 20. June 2003 19:43
quote:
Nelson:
Charlie, you lifted those sentences out of context, and then you changed the subject, brought in a completely different quote, and then completely misinterepreted it.
I could hardly have quoted you out of context, since I ended up quoting your entire post. I could well have misunderstood you, in fact I am quite sure I did, so bizarre are some of your claims, but you haven't helped shedding any light on what you really mean. quote:
The difference between a monogenic trait and polygenic trait is that with a polygenic trait a single gene is insufficient, as it is with hybrid sterility.
And that's where you are mistaken in your definition. Monogenic traits can include things like partial penetrance, variable expressivity, epistatic interactions, etc. I understand that in the world of slide ruler, stick-and-ball imaginary biology things should not work that way, but in the real world of messy living things, they do. That's why according to your definition, retinoblastoma is not monogenic, while out here it's the prototypical monogenic, autosomal dominant hereditary cancer syndrome.
But again, you problem lies elsewhere. You think that retinoblastoma could not be monogenic because it's not expressed in fibroblasts quote:
With my example (Rb-null fibroblasts), it seems to indicate that retinoblastoma is not monogenic.
but of course retinoblastoma is a trait that is not expressed in fibroblasts, it's expressed in children (usually), and at the molecular level it's expressed in some (though not all) of their retinal cells. It is irrelevant that is not expressed in fibroblasts, just like albinism does not cease to be monogenic because it's not expressed in, say, liver cells.
Similarly, hybrid sterility is expressed in hybrids, i.e. in the context of a hybrid genome. Your demand that a bona fide gene causing hybrid sterility should cause sterility in non-hybrids as well is just, as I said, plain bizarre.
Equally weird is you claim that Darwin claimed natural selection to be the cause of phenotypes (or traits, or characters) - it of course does nothing of the sort, it only selects phenotypes. The origin of variation was a mistery to Darwin, so much so that to explain it he came up with his most-forgettable theory, pangenesis. One thing only he got right, he claimed that in order for his theory of evolution by natural selection to work efficiently, genetic inheritance had to be particulate rather than blending. He was of course proven right on that one by Mendel and a century-plus of genetics.
So you are confused about definitions of "monogenic" and of "phenotype", you mix up levels of analysis, and you seem to completely misunderstand one of the most basic tenets of darwinian theory.
Sorry most of our exchanges have to turn into a lecture on one basic biology issue or another, but I just can't help it (it must be in the genes! ![[Wink]](wink.gif) )
[ 20. June 2003, 19:44: Message edited by: charlie d. ]
IP: Logged
|
|
nosivad
Member
Member # 767
|
posted 20. June 2003 20:17
charlie d, speaking as an anti-darwinist, I would ask for a clarification of what you mean by "one of the most basic tenets of darwinian theory." nosivad
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 20. June 2003 20:41
Charlie wrote:
quote:
I could hardly have quoted you out of context, since I ended up quoting your entire post.
Charlie, you didn't quote my entire post. You quoted 4 statements lifted out of paragraphs from my 18. June 2003 23:51 post that was discussing why your examples do not point to retinoblastoma being polygenic. When I pointed out that you took me out of context, you then proceeded to quote a paragraph from my 17 June 2003 21:37 post that was discussing something entirely different, the difference between monogenic traits and hybrid sterility. And then you call my claims "bizarre", go figure.
charlie wrote:
quote:
And that's where you are mistaken in your definition. Monogenic traits can include things like partial penetrance, variable expressivity, epistatic interactions, etc.
None of this has to do with the major difference between a monogenic trait and a polygenic trait. Where the latter relies on interaction with at least one other gene. I realize that that sounds bizarre to you but thats how biologists define it.
charlie wrote:
quote:
but of course retinoblastoma is a trait that is not expressed in fibroblasts, it's expressed in children (usually), and at the molecular level it's expressed in some (though not all) of their retinal cells.
Speaking of lessons in Biology, retinoblastoma is derived from retinoblasts. The RB gene is expressed in all these cells. Mixing Rb-/- ES cells with Rb+/+ ES cells in an experiment, you will not see retinoblastoma. This suggests Rb null retinoblasts is not sufficient to generate retinoblastoma. Although this is negative evidence, many other cell types are also found to have Rb-null fibroblasts and nothing.
With respect to hybrid sterility, I never said that a gene should cause sterility in non-hybrids. The reason why this wouldn't happen is becuase it's not monogenic, Nup96 is incompatible specifically with another yet unidentified gene in the other species. What is weird is that you keep making stuff up that I said and then of course, you call it bizarre, but it's bizarre only because you made it up out of thin air.
Charlie wrote:
quote:
Equally weird is you claim that Darwin claimed natural selection to be the cause of phenotypes (or traits, or characters) - it of course does nothing of the sort, it only selects phenotypes.
Thats just being pedantic. Yes, natural selection only selects phenotypes, but it selects the change in the genes that is beneficial. So one can give natural selection the "blind watchmaker" metaphor in that phenotypic traits is built upon the last beneficial change, with each beneficial change selected for by natural selection. So, again, if changes to genes alone don't matter much, then as Frank Harold states, "selection has been ejected from its throne as the dominant creator of biological form".
Darwin himself did think that natural selection builds characters in this fashion, but he did give this caveat:
quote:
we may easily err in attributing importance to characters, and in believing that they have been developed through natural selection".
[ 20. June 2003, 22:17: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 20. June 2003 23:40
For cryin' out loud, Nelson, a retinoblast is certainly not a fibroblast! Also, you are now confusing the expression of the retinoblastoma gene product (the Rb protein), which is AFAIK ubiquitous, with the expression of retinoblastoma the trait, which is a tumor of retinal progenitors. How more messed up can you get. Just becasue a gene, its product and the trait its mutation causes bear the same name, it doesn't mean the three things are interchangeable.
But anyway, I realize you'll never relent. If you want to keep believing that dominant familial retinoblatoma must be a polygenic trait because it's incompletely penetrant (and not expressed in fibroblasts), that hybrid sterility is not a phenotype, that Nup96 is not sufficient to cause hybrid sterility because it only causes sterility in hybrids, and that Darwin claimed that natural selection was the origin of phenotypes, just suit yourself. I'm outta here.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 21. June 2003 00:19
Oh give me a break Charlie. Where do you get the idea that I wrote retinoblastoma gene product is retinoblastoma the trait? I said nothing like that. Retinoblastoma is thought to occur with the inactivation of Rb. So where it is expressed and if it is already inactivated and nothing happens then that is extremely relevant tot he question.
It's obvious you don't understand my posts, I noted quite specifically that other cell types that have Rb-null fibroblasts do not form tumors. And that specifically Rb null retinoblasts do not form retinoblastoma, retinoblasts and fibroblasts are kind of the same thing they are both precursor cells.
You went ahead and repeated the things you made up which I already responded to, no need to repeat myself.
Bye Charlie, see ya in another thread.
[ 21. June 2003, 00:20: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 21. June 2003 04:40
Well, that was a lot of frustration arising from what should have been an obvious and simple clarification about the distinction between a monogenic trait and the evolvability of a phenotype. But I'll explain at length.
If you start off with one genotype and corresponding genotype (e.g. brown hair, or round eyes), and split into two populations, you can get differing phenotypes in those two populations by mutations in a single gene in one population (e.g. giving blond hair, or bar eyes (in Drosophila)).
However, this is unlikely to be enough for speciation since it would require that heterozygotes were lethal but both homozygotes were viable. I'm not aware of any cases of this, but I suppose one might be able to construct a multimeric channel that had such a property.
Otherwise, you need two or more genes to change, so that if one population has genotype A;B / A;B and the other has a;b / a;b then A;a / B;b is lethal (e.g. via lethal interactions of a with B and/or A with b).
Nup96 appears to follow the second pattern; when you take an allele from one population (a) and stick it in the other population (A;a/B;B), it's lethal. In the context of the second population, Nup96 is sufficient for lethality, and thus is a monogenic trait (by some definitions).
However, to understand the evolution of this lethality, one must also understand what interacts with Nup96 to cause lethality in one population but not the other. This, so far, is missing and is being worked on.
[ 21. June 2003, 04:41: Message edited by: Rex Kerr ]
IP: Logged
|
|
nosivad
Member
Member # 767
|
posted 21. June 2003 06:34
Rex, there are plenty of cases where the heterozygote is lethal. If you fertilize Rana pipiens eggs with Rana catesbiana (bullfrog) sperm, development is normal up to gastrulation and then ceases when the heterozygote genome is turned on. I am sure there are countless other examples or do I misunderstand your meaning? nosivad
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 21. June 2003 08:12
Rex:
I do agree with you, but Nelson initially objected to my saying that Nup96 is sufficient for lethality in hybrids, which it is, and then proceeded to dig himself a hole by bringing up the concept of "monogenic" and butchering the definitions of phenotype, trait and a good deal of genetics and cell and molecular biology along the way.
He just can't seem to stop: quote:
I noted quite specifically that other cell types that have Rb-null fibroblasts do not form tumors.
Cell types with Rb-null fibroblasts? ![[Confused]](confused.gif) Maybe he meant other cell types with Rb-null mutations, but how can he expect to have anybody understand him when he uses biological terms this way? quote:
...retinoblasts and fibroblasts are kind of the same thing they are both precursor cells
Kind of the same thing? Fibroblasts are differentiated connective tissue cells, retinoblasts are precursors of retinal cells, of neural tissue origin.
I actually think Nelson probably understands more than one can gather from his posts, but his inability to express his thoughts properly, and to admit even his most obvious terminological and conceptual mistakes, always end up generating these ridiculous diatribes that add nothing to the discussion. Too bad.
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 21. June 2003 17:00
Nelson ponders:
quote:
So, again, if changes to genes alone don't matter much, then as Frank Harold states, "selection has been ejected from its throne as the dominant creator of biological form".
That makes totally no sense to me. Selection is an essential component to act on variation. That changes to some genes may have no impact on selectability seems somewhat irrelevant.
What is Nelson trying to communicate here? Certainly your 'responses' to Charlie do not seem to help furthering the discussion here.
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 21. June 2003 17:18
charlie wrote:
quote:
I do agree with you, but Nelson initially objected to my saying that Nup96 is sufficient for lethality in hybrids,
Nup96 isn't sufficient for lethality in hybrids. I don't know by what definition one would think that it is sufficient. If Nup96 caused inviability regardless of a second gene, then those changes would never have spread.
To relate this to Rex's post, lets start with an ancestor with a genotype of bbcc. In one species, we get a C mutation, giving bbCC (unidentified second gene). And in the other species we have a B mutation that results in BBcc (nup96). A hybrid of these two specices would form BbCc. That could result in hybrid inviability because the interaction could be lethal.
charlie wrote:
quote:
and then proceeded to dig himself a hole by bringing up the concept of "monogenic"
No. It was you that brought up the term monogenic in an attempt to rescue your botched interpretation of the paper that you yourself cited. I've been following your lead ever since. And ever since, there has been a flood of half-truths and outright deceit comming from your direction. I don't mind though, as you can see, I had no problem filling in the blanks.
But hey no hard feelings!
charlie writes:
quote:
Cell types with Rb-null fibroblasts? Maybe he meant other cell types with Rb-null mutations, but how can he expect to have anybody understand him when he uses biological terms this way?
See, this is what I mean. What exactly is wrong with the term Rb-null fibroblasts? It's actually used quite commonly in the literature even in conjuction with retinoblastoma,
quote:
One such example can be seen in the study of tumor suppressor gene function in general and the
retinoblastoma tumor suppressor in particular...We have used DNA microarray analysis to identify gene expression profiles in wild type and Rb null mouse embryo fibroblasts, as well as cells lacking other developing a more complete understanding of Rb function.
Distinct Gene Expression Phenotypes of Cells Lacking Rb and Rb Family Members Esther P. Black , Erich Huang Cancer Research (2003)
quote:
In addition, the relatively normal
transcriptional induction of the cell cycle-regulated genes c-fos, c-myc, and the cyclin D gene observed in pRb-deficient fibroblasts
suggests that these genes may be regulated by other non-pRb proteins."
Altered cell cycle kinetics, gene expression, and G1 restriction point regulation in Rb-deficient fibroblasts.Mol Cell Biol. 1996 May;16(5):2402-7.
Even with other types of genes:
quote:
whereas p53-null fibroblasts attain DNA
www.hawaii.edu/crch/Agarwal1_1998.pdf
and it accurately describes the situation I've been writing about.
[ 21. June 2003, 19:26: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 21. June 2003 17:22
Pim writes:
quote:
That makes totally no sense to me. Selection is an essential component to act on variation
That variation refers to random mutation of genes. But if changes to genes are only a very small part of the story (and it seems that they are) then selection acting on variation is irrelevant (well not completely, just not the main factor anymore). And so the phenotype can change or remain constant even against the wishes of the blind watchmaker.
[ 21. June 2003, 17:35: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Pim van Meurs
Member
Member # 541
|
posted 21. June 2003 18:21
Nelson suggests:
quote:
That variation refers to random mutation of genes.
Why? not necessarily. Was Darwin talking about genes when he proposed his theory of evolution? I doubt it.
quote:
But if changes to genes are only a very small part of the story (and it seems that they are) then selection acting on variation is irrelevant (well not completely, just not the main factor anymore).
How did you reach this conclusion? Irrelevant, not the main factor? Only a small part of thes story? It seems that there are some unsupported assertions that could benefit from some in depth support with actual data.
quote:
And so the phenotype can change or remain constant even against the wishes of the blind watchmaker
Again this does not make much sense. 'Wishes of the blind watchmaker'? What is Nelson refering to here? What wishes?
Btw you still seem to suggest that "Nup96 isn't sufficient for lethality in hybrids"
quote:
"... inviability has evolved between species as a by-product of adaptation of a protein with an essential function. The gene involved encodes a protein component of the nuclear pore complex that has evolved by positive natural selection in both species independently."
and
quote:
Presgraves et al. analyzed rescued crosses between D. melanogaster and its closest relative, D. simulans, and mapped a hybrid inviability gene to region 95AB on chromosome 3R. They analyzed all 12 loci within this region and identified a single-copy, dicistronic gene encoding nucleoporins Nup98 and Nup96, which function in RNA export and which contain a region surrounding a cleavage site that is perfectly conserved across many species, including humans. Analysis of the gene sequences suggested that positive selection accounted for the changes in both species and that the adaptive event was not recent.
and
quote:
In the June 12 Nature, Daven Presgraves and colleagues at the University of Rochester have adopted such an approach and describe a mechanism by which a single gene encoding two nuclear pore proteins can cause hybrid inviability (Nature, 423:715-719, June 12, 2003).
Source
From the press release
quote:
Presgraves and colleagues found 20 regions that differed on the chromosomes of two species of fruit flies that were estimated to have diverged in evolution 2.5 million years ago--fairly recently in evolutionary terms. He then needed to find a gene in one of those regions that was responsible for preventing successful reproduction between the two species. If the species could reproduce, then they could swap genes back and forth and thus would not be truly separate species. Something would have to prevent the transfer of genes, and in the case of Presgraves' fruit flies, that something was the proclivity for hybrid larvae to die before maturing into adults.
...
He found his gene, called Nup 96, that always prevented a hybrid of the two species from living to reproduce, and he sequenced its DNA.
....
Nup 96 turned out to code for a certain kind of protein that was part of an essential pore in the nucleus of every cell in the fly. If one member of one species of fly mated with a member of the other, this pore would not properly form and the hybrid fly would die. Even though this was an unprecedented finding in itself, it posed a new question that came wholly unexpectedly: Nuclear pores are found in every cell in the world that has a nucleus, so it was labeled as "highly conserved," meaning it was always assumed that it was so important to the survival of an organism that evolution would never tamper with it. Further study will be needed to understand why altering this pore was seen as useful in the evolution of this particular species of fruit fly.
Source
From the actual paper
quote:
Complementation tests show that hybrid lethality is caused by the D. simulans (Ds) Nup96 protein and suggest (but do not prove) that hybrid lethality involves its amino terminus.
and their conclusion
quote:
Our molecular and population genetic analyses ofNup96 allow us to address a number of issues in speciation genetics1,2. First, a single gene can explain the hybrid lethality of a small chromosomal region identified in our deficiency screen.
quote:
Third, Nup96 evolved by positive natural selection in both species’ lineages. As an incidental by-product of this adaptation, the D. simulans Nup96 protein is no longer compatible with an (unknown) interacting factor(s) encoded by the D. melanogaster X chromosome.
Nelson, you may also want to tone down your accusations
"And ever since, there has been a flood of half-truths and outright deceit comming from your direction. I don't mind though, as you can see, I had no problem filling in the blanks"
[ 21. June 2003, 18:51: Message edited by: Pim van Meurs ]
IP: Logged
|
|
Nel
Member
Member # 614
|
posted 21. June 2003 19:13
Darwin was not talking about genes but thats irrelevant, because he did not know they existed. The modern synthesis is talking about genes (neo-Darwinism).
I support this contention by the facts that I have references that ranged from nematoda to E. Coli. Please address those, all you are doing is asserting that I'm wrong again.
As far as Nup96, it's not sufficient to cause hybrid sterility and none of your quotes support that contention, which is why you didn't bother discussing them. In your first quote, you leave out this part:
quote:
Fine-scale molecular and evolutionary analysis of the first of these genes to have been identified shows species divergence in action.
First of these genes means there is more to follow.
In your second quote you repeat an error you made a few posts back which is kind of odd. You leave out this part:
quote:
The D. simulans Nup96 protein is no longer compatible with an (unknown) interacting factor(s) encoded by the D. melanogaster X chromosome," conclude the authors.
Thats at least two genes. This goes for your third quote as well. Your fourth quote simply discusses that they found Nup96. From the actual paper you leave out this tidbit:
quote:
Abundant evidence shows that these hybrid fitness problems are caused by incompatible interactions between loci: new alleles that become established in one species are sometimes functionally incompatible with alleles at interacting loci from another species.
Thats again, two genes.
Pim writes:
quote:
Nelson, you may also want to tone down your accusations
That was not an accusation, that was an observation.
[ 21. June 2003, 19:16: Message edited by: Nelson_Alonso ]
IP: Logged
|
|
Rex Kerr
Member
Member # 632
|
posted 21. June 2003 20:06
Nelson, if you have an organism with a given phenotype, and you change, remove, or add a gene A, and get a new phenotype, then we say that this modification is sufficient to cause the new phenotype.
Therefore, D. simulans Nup96 is sufficient to cause lethality in D. melanogaster. This is the standard way that the terms are used.
Changes in Nup96 alone give an incomplete picture of the evolution of hybrid inviability, but don't confuse that with the standard terminology for sufficiency.
Put another way, D. simulans Nup96 is sufficient for hybrid inviability with D. melanogaster, but changes in Nup96 are not sufficient to establish reproductive isolation from a single ancestral population.
(P.S. the reason "cell-types with Rb-null fibroblasts" is nonsensical is that it is analagous to "cars with broken-windshield Toyotas" .)
Now, I am completely puzzled by Nelson's apparent assertion that evidence from E. coli to Nematoda suggests that genes do not cause variation. There are thousands of known mutants between E. coli, S. cerevisae and S. pombe, D. melanogaster, and C. elegans that show phenotypic differences. Heritable phenotypic differences that map to changes in single genes.
Finally, nosviad, you do misunderstand me. I was talking about heterozygotes at a single locus. I agree that there are many examples where hybrid offspring are inviable; they're just heterozygous for more than a single gene.
IP: Logged
|
|
|
Printer-friendly view of this topic
