See also The Pond.
[mi1]
From Keith Robison: I ask you this. Humans and chimps have extremely similar genomes, and I doubt even you would attempt to claim they sprung to life independently.
JM: According to Senapathy, they probably did. If there are unique genes in either genome, then he would say that they must have.
Keith: Yet chimps and humans show a number of key differences which must be genetic, and are quite extreme. For example, chimpanzee females develop a large swelling on their buttocks during reproductive heat. Where did those novel genes/pathways come from? Are you going to claim that is "normal" variation?
JM: The Senapathy test is: are there unique body parts or organs (and thus unique genes)? If so, they are independent organisms.
Keith: Okay, so we agree -- by Senapathy's definition these must be independent species. Then why are humans and chimps so similar? Why, at the level of chromosome banding, do their chromosomes look nearly identical, save a few identifiable rearrangements? While we don't yet know the changes which make the difference between a chimp and human, we do they have an enormous amount in common. Do you find it more parsimonious to assume some sort of large-scale Senapathian borrowing, or that perhaps there are processes in the genome overlooked or inappropriately dismissed by Senapathy which might generate new genes? (e.g. jingwei)
Of course, Senapathy can't survive such an argument. If he gives in once, then the whole thing unravels. If you admit that chimps and humans are evolutionarily connectable, then the evidence for tacking on gorillas isn't much worse. Next it's the orangs. Progressive deterioration of the argument is now a gibbon (Ugh! :-) Lesser primates and monkeys join in, and before you know it all mammals have been connected (and so on). There's no obvious place to say "here the similarities stop."
[mi2]
From mike_f@delphi.com: How could one species with one number of chromosomes possibly evolve into another species with a different number of chromosomes?
[mi3]
From Roger Gary: I see [developmental timing] as a black box between genetics and comparative anatomy, where little is known. I think ultimately we are going to discover that most major morphological change springs from changes in regulatory genes that control developmental rate and timing.
Paleontologists have spent a lot of time studying the change from the reptilian quadrate/articular jaw joint to the mammalian squamosal/dentary joint. Tim Rowe (a paleontologist here at UT) gave us a great lecture on this illustrated with slides of possum embryos. Early in development, the quadrate, articular and angular begin ossifying at the posterior of the dentary. The back of the skull then expands rapidly and captures these bones. The quadrate and articular eventually end up in the inner ear as the incus and malleus, while the angular forms the tympanic. These developmental changes are mirrored in the fossil record of the synapsids.
The change in the rate of growth of the posterior bones of the skull seems to have been driven partly by expansion of the brain, and partly because synapsids were driven to small body size. The physical constraints imposed by housing sense organs which cannot be scaled down isometrically helped drive the changes in shape of the surrounding bones of the skull. This brings up the interesting question as to why the synapsids were driven to small body size during their phylogeny, while the dinosaurs never managed to get much smaller than a chicken. I think [heterochrony is a mechanism ... that works with selection to produce the patterns of macroevolution we see], and I think this is also evidence that Dr. Senapathy is wrong.
[mi4]
From Ingrid Jakobsen: But, forgive me for being stupid here, but I thought that was precisely what "natural selection" and "survival of the fittest" and all those catch phrases were meant to cover. Sure, there may be a long path of required mutations to get from species A to species B and most random mutations aren't on that path, but the individuals with those mutations don't contribute (as much) to the following generations, while those with the "useful" mutations do. I'm sure Richard Dawkins demonstrated the effect in The Blind Watchmaker.
Or is the assumption that only all of the 10% change at once is beneficial? That intermediate stages, with only some of the mutations, aren't viable? I'm confused.
While I'm here, could Dr. Senapathy or one of his supporters explain to me where independent genesis rather than microevolution kicks in?
- Between humans?
- Between humans and chimpanzees/bonobos/gorillas?
- Between the African apes and other apes?
- Between apes and Old World Monkeys?
- Between Old and New World Monkeys?
- Between monkeys and tarsiers? etc.... Where is the cutoff?
- How about among insects (e.g. Drosophila)?
- Plants? For example, does C4 metabolism require independent origin?
[mi5]
From Dave Oldridge: Senapathy simply does bad math. His error is identical to that of creationists who state that evolution could not happen because (for example) elephants are highly improbable.
JM: Senapathy says life could have originated independently more than once, but few evolutionists are prepared to debate that point, because then the question becomes: "how many times is too many?"
Dave: Simply: the evidence is that it either only happened once or that the other times were unsuccessful at establishing themselves. If Senapathy's theory is correct, for example, what happened to all the DNA that twists the other way? If Senapathy is correct, shouldn't there be two major branches of living organisms around, not just one?
[mi6]
From Peter H. Weis: The latest wrench thrown into the wheels of evolution are the results of two independent studies where researchers eliminated the genes for glucose assimilation in bacteria, then placed them into a glucose solution.
Although most bacteria died due to starvation as expected, the rate of acquisition of genes for sucrose assimilation was far in excess of what random mutation would allow. The astonishing recovery of the bacteria is still unexplained.
My hypothesis is that the electromagnetic field of the sucrose solution triggered a genetic response in the bacteria which led to the rapid acquisition of genes for sucrose assimilation. This would mean something like "focused accommodation" rather than random genetic mutation, genetic drift, or natural selection.
[mi7]
From Keith Robison: It is customary in science to look for simple models before the complex ones. There is quite a paper trail on "Cairnsian" mutation, and it is looking more-and-more like something quite ordinary. The explanation appears to be that the mutation rate increases in starving cells, and that cells which mutate so as to metabolize the sugar (lactose BTW) survive, while the others keep trying until they succeed or mutate themselves to death. The effect was inadvertently amplified in many of the published experiments by the experimental design (the mutated lactose-utilization gene was in a different genetic compartment, which probably has a much higher mutation rate under starvation conditions).
In summary, current evidence suggests that most of these experiments can be explained as a special case of Darwinian evolution acting at a molecular scale ("survival of the fittest chromosome"); it is looking less-and-less like a directed process, and certainly doesn't require bizarre "electromagnetic fields."
[mi8]
From Steve LaBonne: Well, I've also been very skeptical, and I would still say that few if any experiments have really had adequate controls to be able to say that unselected markers are not also mutating. But I was intrigued by a letter from Bryn A. Bridges in the 29 June issue of Nature. He proposes what strikes me as the first totally plausible and well-worked-out explanation of how purely Darwinian mechanisms could produce genuinely directed mutation. In brief, a non-bulky (hence invisible to mismatch repair) DNA lesion like 8-oxo-G could cause mispairing during TRANSCRIPTION (this is the key point). If such a lesion occurs in the selected gene, this transcriptional mispairing could give a transient wt phenotype. In turn, that transient wt phenotype would allow the cells with the "correct" lesions, and only those cells, to initiate a round of DNA replication- during which the mutation could become "fixed" by mispairing of 8-oxo-G with A (consistent with the fact that "adaptively" mutated alleles have frequent G-C to T-A transversions).
Of course this only means there is a plausible mechanism, not that the existence of "adaptive" mutations has been proven, much less that they are known to be a significant evolutionary phenomenon rather than an interesting laboratory artifact. But at least, this area has contributed to the study of stationary-phase physiology by motivating the discovery of a mutation pathway in starved cells that, whether potentially "directed" or not, is definitely novel. So even if the whole "directed" aspect evaporates, some genuine new knowledge will have come out of it.
[mi9]
From Ralph M. Bernstein: Jeff, how would you test Darwin's theory? In our time frame, it is really impossible to truly test it. Fossil records, if I remember correctly, you (for) Senapathy abjectly dismiss. ...
JM: I am trying to figure out how to test it. Certainly, the advances in molecular biology and knowledge about genetics will help, and a test may be just around the corner. A few people think the test can be done now. Senapathy does not dismiss the fossil record. Indeed, he likes the fossil record because it fits his theory much better than it does Darwin's.
Ralph: ... natural selection, natural selection, it all comes down to that. is plant growing/breeding or dog breeding natural selection per se?...
JM: No, that is clearly artificial selection. You are just modifying the animal with its own immutable framework.
Ralph: ...if so, why weren't/aren't these breeds of plant that we grow/grew through husbandry/farming already present? Isn't that what Senapathy-selection says?
JM: Senapathy never suggests that every possible genome came into being. He just says it is possible to find nearly every gene in the random DNA, if you use introns-early.
Ralph: ... "huge pool o' genes, primordial slush, they're all there!" I think that the original argument that you had has been "worked around" until you could find a topic you could "defend."
JM: No change that I know of. You seem to think that Senapathy predicts that every possible genome will come out of the pond. No, the pond just has the ability to produce every possible gene (with a few restrictions on exon length), and the genes that are produced may or may not ever become part of a viable genome. Most won't get that far.
Ralph: I am not being critical of you, I'm just weary of Senapathy and his musings. Honestly, I work on basic science, purely evolution, and nothing but. I work on the most basic stuff you can get-genes, genes and more genes. The group I work with actually described a phenomenon we call a "big bang" of gene duplication. But it is a relative thing. These genes weren't around in a big pool, floatin' for a putative crab or shark to 'come and get it'. Genes get distributed through out "evolution" by being present in an ancestor-a branching point if you will. This is the most common type of distribution. This can be seen in the distribution of immunoglobulin genes. these are really only present in vertebrates, gnathastome vertebrates to be precise. They are present in the most ancient ancestor, the shark. these are not present in the crab, which is what Senapathy's theory says should be true.
JM: Does he? Where? Please explain to me why his theory predicts this. In the early section about evolution he says "The immunoglobulins ... and the vertebrate immune system are totally absent in invertebrates. ... Those few substances which are capable of binding foreign materials in certain invertebrates have entirely different structures from the vertebrate counterparts. On the other hand, proteins very similar to the mammalian immunoglobulins are present even in the lower vertebrates such as the fish and amphibians. [snip] [This can] illustrate how it is impossible for the vertebrate immune system and its proteins to have originated from invertebrate creatures by organismal evolution." [page 134]
[mi10]
From Ralph M. Bernstein: (quoting JM) "No, that is clearly artificial selection. You are just modifying the animal with its own immutable framework."
Ralph: Actually, I think that this is natural selection, just speeded up. Anyway, immutable you say, I don't think that you mean immutable, since genomes are clearly shown to be "plastic" in many ways.
JM: Here, the term "immutable" means that no new genes could come about. But it does not mean that protein levels could not change. Existing genetic pathways could be turned on or off, but those pathways would have been there all the time. So, for example, you can get different size organisms of the same species, or different colors, and so on, but never get a new body part.
Ralph: "If you use intron early" what's that mean? Intron early is valid to me-mRNA world etc. But Senapathy's "work" on this is really just speculation. If you read his Science letters/comments from 4-5 weeks ago, you will see what I mean. Also, you may want to look up some of WF Doolittle's intron early/late papers -- for hard evidence The point is, if you believe in introns-early, you probably believe in an RNA world-early.
JM: Correct. All of Senapathy's first seed cells are eukaryotes, so the mRNA is there, too.
Ralph: This DNA world stuff just wouldn't work. RNA is catalytic and auto-catalytic, and DNA is almost never catalytic. In fact, some of the most important and ancient molecules in all of our bodies are RNA-zyme components. These are not DNA, and for DNA to "come into the pool" they must have been first. Hypothetically, DNA is really only around because of natural selection -- it was advantageous to have a storage device for the important material-RNA.
"The immunoglobulins ... and the vertebrate immune system are totally absent in invertebrates. ... Those few substances which are capable of binding foreign materials in certain invertebrates have entirely different structures from the vertebrate counterparts. On the other hand, proteins very similar to the mammalian immunoglobulins are present even in the lower vertebrates such as the fish and amphibians."
In fact, Jeff, they are almost identical! Fish and amphibians are vertebrates, and ALL jawed-vertebrates have all the major components of the vertebrate type immune response. And I would beg to differ that fish and amphibians are "lower" vertebrates, simply ancient (just my plug for the underrepresented -- hey I'm PC) now sharks and lampreys! Those are ancient.
OK, to the invertebrate part, invertebrate c-reactive protein and invertebrate ncam is pretty darn close to the "lower" vertebrates', even mammals'. Anyway, this is what may have given the vertebrates a great selective advantage in evolution, this ability to defend against non-self. This may have been one of the factors that allowed the vertebrate radiation to occur -- keep in mind that vertebrates "showed" up some where around only 500 million years ago. Not too long ago, actually.
"[This can] illustrate how it is impossible for the vertebrate immune system and its proteins to have originated from invertebrate creatures by organismal evolution." [page 134]
Well, they probably didn't evolve from an invertebrate, but an ancestor common to both vertebrates and invertebrates.
OK, well I just thought that I would put a couple-a-more cents in. I would welcome some comments on this from real-evolutionists out there, after all, I just study the evolution of the immune system through phylogeny.
[mi11]
From Keith Robison: We know of examples of new genes arising (e.g. jingwei), and know of many more mechanisms which could form genes.
JM: Can you please explain more about jingwei and the new genes that arose? And any mechanisms that could form new genes beyond those that Dr. Senapathy discusses in Chapter 4?
Keith: Look at "Natural selection and the origin of jingwei, a chimeric processed functional gene in Drosophila," Long M; Langley CH, Science 260: 91-5 (1993).
Senapathy pretty much covers all of the proposed mechanisms for gene formation -- not to say he covers them well. You'll note that his "disproofs" of almost every one rely on his flawed "genes can't evolve" "proof."
jingwei is a gene of unknown function found in 2 species of fruit flies. It formed through the retrotransposition of an alcohol dehydrogenase gene into another gene. The function of this second gene is not known. The result is a new gene with parts of its parents, and an expression pattern different from Adh. While its function is not known, population genetics analysis shows that it is under selection -- nucleotide substitutions which would alter the protein are less frequent than neutral substitutions. Hence, a new gene which has been retained by natural selection.
[mi12]
From Josh Hayes: (via e-mail): It's not simply that dogs and cats both have four legs, but that they have exactly the same bones in the legs -- and paws -- as we do, and as BIRDS do, as well. It's not just "a leg," it's about forty bones.
JM: This fits in with a theory of evolution, but it also fits Senapathy's theory, too, because of the re-use of genomes. If an organism was successful, it would contribute much more of its DNA to the pond and portions of that DNA and copies thereof would be more likely to be reused. If large parts of a genome were reused en masse, then such things as body structures and complex organs would be the same (or nearly so) in many organisms.
There is a subtle similarity between Darwin's evolution and Senapathy's reuse -- evolution is a form of reuse since each new generation is a copy of the previous, and the reuse of previous organisms' genes is a form of evolution since viable genomes would contribute to later-formed organisms. So, these two mechanisms could arguably produce similar results.
What we need is a definitive test that will determine which mechanism is correct. However, if you come up with a test, you must consider how it might be passed by both theories. Most of the "evidence" people say refutes Senapathy is, in fact, just evidence that supports evolution -- and support of either theory does not automatically rule out the other. Just to avoid confusion, "evolution" means long-term evolution (natural selection and descent with modification), not short-term (adaptation or artificial selection).
[mi13]
From Josh Hayes: The "cost" of new body parts involves some understanding of embryology: the cellular machinery for the manufacture of body parts is incredibly complex. We have only this year discovered a "master gene" for construction of eyes, and that gene cascades control through hundreds of other genes. The construction of a novel structure requires putting that kind of machinery in place, and that is no minor undertaking.
One of the things that Stephen J. Gould finds so compelling about macro- evolution is the sharing of characteristics across taxa, even to the point of making do with ridiculously poorly designed features (I think here of the human knee, but he'd probably look at flippers in cetaceans or "thumbs" in pandas). If new body parts could just be sprouted, one would think that design would be better than it is...as it is, a lot of animals make do with a lot of jury-rigged solutions to evolutionary problems. This strikes me as compelling evidence for the idea of common descent.
JM: And it strikes me as evidence for Senapathian independent births, assuming reuse is a playing a major part. In fact, if there are so many "ridiculously poorly-designed features," would that not tend to refute evolution because such features should be selected against (if something better came along) or never have come about in the first place due to little or no utility? In Senapathy's pond, useless/strange organs would be likely but, just as with most mutations, useless/strange organs might often be detrimental to (or too "costly" for) survival so they would be eliminated.
[mi14]
From Dave Oldridge: There would be no discernible cladistic tree and it would bear no discernible relationship to the tree derived from comparative anatomy if Senapathy was correct. Again, he ignores evidence that doesn't back up his theory.
JM: Then I have to assume you have not read his theory because anatomical similarities at all levels are predicted.
[mi15]
From Tim Ikeda: One other thing: Why the heck should the mitochondrial and chloroplast sequences produce the same trees as everything else if not by common descent? One might think that during eukaryotic cell "assembly" in Senapathy's "pool" these organelles could be added independently. Yet the evidence is that these organelles follow the same pattern as everything else and even root within specific bacterial groups.
[mi16]
JM: There is a subtle similarity between Darwin's evolution and Senapathy's reuse -- evolution is a form of reuse since each new generation is a copy of the previous, and the reuse of previous organisms' genes is a form of evolution since viable genomes would contribute to later-formed organisms. So, I'm not surprised that these two mechanisms could arguably produce similar results.
From Keith Robison:However the main result is something Senapathy "proved" is impossible! If we look at genes with analogous functions between different species, they frequently differ by >10%! So, if Senapathy's equations (along with Jeff's extensions) and their application are correct, then the reuse theory is just as invalid as evolution.
Of course, there is an obvious way out of this dilemma :-)
JM: What I'm looking for is a definitive test that will determine which mechanism is correct. But it is not as simple as some s.b.e. people think because, if you come up with a test, you must consider how it might be passed by both theories. Most of the "evidence" people say refutes Senapathy is, in fact, just evidence that supports evolution -- and support of either theory does not automatically rule out the other. Also, just to avoid confusion, by "evolution," I'm just talking about long-term evolution (natural selection and descent with modification), not short-term (adaptation or artificial selection).
Keith: So we are faced with a choice between a modest extrapolation from observed events which is consistent with most of the data, vs. a Goldbergian* scenario requiring enormous assumptions and which isn't consistent with any data (nor internally consistent)? I know which I'd choose.
* -- as in Rube
[mi17]
From Keith Robison: It's interesting to note that a multitude of quantitative disproofs of ... Senapathy's theory are available on the Web.
A sequence logo is an information-theoretic display of a sequence alignment. The display is based on the number of bits of information in the alignment, and is a visual statement of both the degree of conservation AND the probability of finding a sequence fitting that alignment in a random collection of sequence (i.e. it is statement about the properties of the ensemble).
The server http://www.blocks.fhcrc.org/ contains a large number of multiple alignments, and will give logos on request for those alignments. Buried in the last line of the logo PostScript code is the total information in the logo (area under the informational curve).
For example, the ntrC family is a group of bacterial regulatory proteins. There are 4 blocks of multiple alignment on the BLOCKS server, and they have 282, 122, 98, and 26 bits respectively, for a total of 318 bits (bits are a log2-scale, so adding bits is equivalent to multiplying probabilities). Therefore, the probability of finding an ntrC-class protein in random peptide sequence is approximately 10^-98.
One can repeat the exercise for the other alignments in the collection. While not all individual entries carry quite so many bits, the sum of all the information scores can be converted to the probability of finding all of these patterns in Senapathy's pond. (though you will quickly find yourself stuck in log-space, as the probability gets down below what most computers can represent internally).
For a full description of sequence logos: http://www-lmmb.ncifcrf.gov/~toms/sequencelogo.html
[mi18]
From Shane McKee: New anatomical complexity is an astonishingly rare thing to arise, so it's not surprising that we don't find much between "indisputably related organisms." For example, where is the new anatomical complexity distinguishing a human from a dog? All there seems to be is "reduction and distortion." We don't even need fossil sequences to show that.
I think in evolution, people get way too hung up on the rocks, and don't see that the evidence for the processes that drove/drive evolution is right here and now in the living beasties which populate the planet. We can extract DNA from them, but it's harder from rocks.
[mi19]
From William A. Brindley: We have many intellectual protections against being fooled by "circular reasoning." The objection of "circular reasoning" is no objection at all. Anyone who takes any trouble to dig into modern publications in science or who contributes them is well aware of this.
Evolution is not just the origin of complex organisms. Evolution is also changes in frequency of gene alleles in a population due to some kind of environmental pressure. In the case of complex organisms, evolution from one complex form to another appears to be accomplished by relatively modest changes in the body details or biochemistry of the previous organism. That being the case, it takes a geological time scale to go from a reptile to a mammal-like reptile to a mammal. Given that, the position of the fossils in the geological column will give one an idea of the directionality (not necessarily a good word to use in this context) of the "transition."
The issue raised of "reduction and distortion" is one I have never heard but it obviously contains a lot of individual value and definitional problems. An excellent example is the transition of the multiboned jaw of reptiles to the single boned jaw of mammals. Is this a reduction and distortion or is it an increase and elaboration since the former bones of the reptilian jaw were used to increase and elaborate the mammalian hearing apparatus?
There are also numerous examples of intermediate forms in limbs as the previous poster pointed out.
The assertion that reduction and distortion is the result of evolution from one closely related complex organism to another is also belied in the evolution of humans. There, transitions in the pelvis, skull, and brain are quite clear. John C. Eccles' Evolution of the Brain makes it very clear to me, at least, that reduction and distortion is a simplification beyond meaning.
The place to look for answers to evolutionary problems is to the stream of publications from the main-line scientists who have researched these issues. This includes, obviously, biologists but also geologists, some astronomers, and some cosmologists who have helped define the platforms or time scales upon which biological evolution operates.
[mi20]
From Shane McKee: The fundamental characteristics of an organism, phenotypically, are largely genetically determined. Let's take the differences between a seal and a dog. There is very little genetic difference. ... There are not many new genes. There are no "new body parts" worth speaking of.
JM: If there are new genes, then it's a new birth. However, all the common characteristics and similar genetic material fit perfectly and is predicted by the Senapathy theory. The "random" reuse does not mean the reused genes came out of a pond after being stirred up with an eggbeater. Large intact chucks, and even almost all of a previous genome, could be reused.
Shane: And don't use the old "flippers are not legs" line, because all that has changed is the strength of the relative embryological growth parameters. Looks like the concept of "fundamental immutable characteristics" needs some revision.
JM: No, no change, we just don't agree on the definition. If two organisms differ only because of different levels of proteins that change the "growth parameters," then those are related organisms (even under Senapathy) and they share the same "immutable characteristics." It is just that those characteristics are being expressed at different levels. The term "immutable" refers to the genetic makeup, not the potential or level of expression. Anyway, let's not get into a discussion about words, as in micro-macro-immutable-theory-fact. Concentrate on the examples of what is or is not a separate birth.
Shane: The case of the muntjac deer is one which delivers a substantial body-blow to the Senapathy case. Let me explain. ... the ORDER of the genes on the chromosomes (on the "macro" scale - there are exceptions...) is by-and-large irrelevant in generating a viable organism, in this case, a deer.
Now, this being the case, in Senapathy's gene-pond, the chromosomes for the seed-cells would have the genes mixed up essentially at random, while in standard evolution, we would expect some of the ancestral order to be preserved, and to show the same phylogenetic pattern as the nucleotide sequence of the genes themselves.
JM: Remember, there is no eggbeater in the pond. The amount of reused material is random, and precisely what material gets reused is random, but "random" does NOT require a complete mixing or shuffling of the genes. If it happens, fine, if it does not happen, that's predicted, too. It is obvious that when you say "mixed up essentially at random" you are assuming there is a primordial eggbeater.
Shane: Needless to say, the latter is indeed the case. Unless, of course, entire chromosomes are maintained in the gene-pond, in which case, no Senapathian "new genes" could be incorporated into the genomes anyway. Get out of that one!
JM: A challenge I can meet! Reprise: There is no eggbeater in the pond. It IS possible that entire chromosomes are maintained in the gene-pond, and it IS possible that new genes could be added (or genes removed) during the assembly of a new seed cell.
Shane: How? Evidence? All I see are assertions, with no back-up.
OK, so I haven't read the book yet. However, the book is not a
Bible for following blindly, nor is Dr S. some great guru, on whose
every quote we should hang. Nor is Charles Darwin. The theories
should rise or fall on their merits. I have demonstrated what I
see as holes in the hypothesis, and these have only been countered
by assertion so far. Big problems still remain:
Incidentally, I think it's very heroic of Jeff to be holding the corner for Dr. S., and he has generated some darn good stuff (albeit) logically flawed ;-) for being an engineer and all that. Is Dr. S. about so that we can hear from the man himself?
JM: Ah, some slight applause. Thank you, thank you. :-) (And remember, scientists figure out how things work, but it takes engineers to use that knowledge and build useful things.) And, yes, I have it on good authority that Dr. Senapathy is lurking here and is working on a submission. Finally, I'll see the relief troops! :-)
[mi21]
From Jeffrey Mattox:
To Keith:
I have combined a number of your recent comments here with my replies into one post. I'm trying to spare the tolerant but frustrated non-readers from seeing many separate messages.
Re: Assumptions:
Keith wrote: Senapathy's book is nothing but a stew of assumptions, with a seasoning of untruths disguised with a thin glaze of facts. And furthermore: In other words, assume the primordial broth had everything necessary, despite an utter lack of evidence. And you complain about our assumptions.
JM: You are mixing two very different kind of assumptions. I never complained about making logical scientific assumptions that for any theory may be based on limited data, but I certainly did complain about the non-scientific assumptions. In particular, I was referring to people saying Dr. Senapathy is "wrong because he must believe so-and-so" when those people had not read his book or asked what it was Senapathy wrote on the subject. Assuming Senapathy has a certain belief in order to criticize it is totally unfair, and I have seen plenty of that here. A recent example: assuming that "random" means mix up, as with an eggbeater.
Re: Bugs in the pond, eating it:
Keith: First, there are plenty of intron-bearing eukaryotes which would have a good time ruining a pond. Second, if you assume a world without decomposers all the available carbon and nitrogen will be used up. Again, as I have pointed out before, decomposers are inevitable and necessary, and will destroy the pond. There is absolutely no hope for Senapathy's pond to survive a century of life, let alone millions of years+. And so it's still true: Using the Senapathy example, the first bug will eat all the DNA.
JM: There are innumerable systems operating in equilibrium today -- chemical, biological, you name it. These systems are stable and "survive" because they are closed loop with negative feedback. Why could there not have been such a system (the details of which are unknown to us now) operating in the primordial pond that would produce excess DNA or inhibit the "bugs" so as to allow some of the DNA to survive long enough to participate in a Senapathy birth? Do you dismiss this as impossible, and if so, how do you know?
Repeating a little....
Keith: Also, as I have pointed out, even non-decomposer eukaryotes could quickly ruin an abiogenic mixture. All organisms shed nucleases and proteases, and even a dead critter falling in the pond would be disaster for it.
JM: You use strong words like "destroy," "ruin," and "disaster." You are certain in today's bio-world that this would be the result, and I accept that, but how can you be so certain of that same result in primordial days when you do not know all of the conditions that would exist back then?
Re: new genes:
Keith wrote: jingwei is a gene of unknown function found in 2 species of fruit flies. It formed through the retrotransposition of an alcohol dehydrogenase gene into another gene. The function of this second gene is not known. The result is a new gene with parts of its parents, and an expression pattern different from Adh. While its function is not known, population genetics analysis shows that it is under selection -- nucleotide substitutions which would alter the protein are less frequent than neutral substitutions. Hence, a new gene which has been retained by natural selection.
JM: Inquiry: how can the method (retrotransposition) be determined with certainty (and what does "retro-" mean in this bio context)?
You used the phrase "unknown function" three times. If its function is so unknown, how do you know it's beneficial? Because it is selected and retained? What if it was just a "parasite," as Senapathy discusses on page 120? To wit: "[T]he fields of genetics and molecular evolution are filled with arguments that transposons could be great contributors to evolutionary change ... despite the beliefs of some geneticists and molecular evolutionists that transposons are possibly selfish elements, parasites that function to maintain their survival within the genome[.]"
Re: ORFs and information content:
Keith wrote: In his book Information Theory and Molecular Biology, Hubert Yockey calculates the information content of the protein cytochrome c. ... The information content is directly convertible to the probability of finding a cytochrome c sequence at random from an ORF of similar length. ... This calculation is estimating the probability of finding a cytochrome c once you have generated a translatable mRNA.
JM: Then that is based on the spliced genes. You cannot apply that logic to Senapathy's theory because all of his probability calculations for independent births are based on the numbers for split genes.
ThatIsLikeComputingTheProbabilityForFindingThisPhrase, instead of just the separate words of this phrase in this order.
At your suggestion, I have obtained Yockey's book. Thus far I have only read a few parts, but I can find absolutely no reference to split-genes, introns, exons, or anything else that corresponds to Senapathy's theory that genes were initially formed in little pieces. You could help me by directing me to the particular section in Yockey's book that might be relevant, or by showing how his math or numbers would change if he had in mind finding an exon, not a complete gene.
Keith: Jeff, I'm surprised at you. You are always calling for rigorous estimates. That is exactly what the information theory approach tries to be -- a rigorous estimate of the probability of finding a functional cytochrome c sequence in a mountain of random peptide sequence.
JM: If you work with the information content of an exon, then I'll be more attentive....
Keith: Senapathy can splice and dice all he wants -- but unless you believe the splicing process can generate >10^100 possible messages we ain't going to see a cytochrome c (which would be a pretty good trick with 10^30 nucleotides!).
JM: ...or show me why your information theory numbers apply to the un-spliced gene.
Re: Sequence patterns:
Keith: It is not that we see one pattern -- yes, any data can be fit to a tree. What is absolutely critical is that we see essentially the same tree no matter what collection of data we use. This is quite a deviation from the random distribution you are arguing.
Arlin Stoltzfus countered this argument quite succinctly a long while back. The important "pattern" we see is that when we superimpose trees generated from different data (e.g. different genes, different morphological features), they are nearly always congruent. This is what Senapathy's "independent birth" theory cannot explain, except by the "special pleading" of genome reuse.
JM: If you made a comprehensive list of all the variables on which a tree could be based, including characteristics that are not commonly used, some would show the same pattern, and those will certainly support your view. But, how many will not? I bet there are many that don't support descent with modification, but do support Senapathy.
For example, Senapathy mentions the many different forms of eyes and how a tree based on those forms does not match conventional evolutionary trees. At page 80: "Even according to evolutionists' own account, it is very hard to correlate the gradations [(now quoting Darwin:) from a perfect and complex eye to one very imperfect and simple, each grade being useful to its possessor] within the supposed ladder of animal evolution. Even after freely buttressing their theme, they could come up with sixty or more separate 'independent' lines of eye evolution in the animal world. In fact, in my opinion, if all the buttressing is avoided, the number would go up to hundreds of different lines." (And at page 83): "Squids and octopuses, which are invertebrates, have highly perfect eyes similar to, and even more advanced than, those of the vertebrates. However, according to evolutionary theory, vertebrates are not descended from these animals. Instead vertebrates are supposed to have evolved from animals that had the compound, pinhole, or some other kind of eye of some other invertebrates. ... There are hundreds of different kinds of eyes in the living world, evolutionarily uncorrelatable; staunch evolutionists themselves say that if one has to construct an evolutionary tree of different organisms in the living world based on their eyes, it would be a monumental task. This, we can see, is because these eyes were not evolved from earlier ancestors by descent with modification."
Also, how do you account for all the unique structures and organs detailed on pages 456 through 487? Shouldn't there be more intermediates found?
Re: Reuse of genes:
I wrote: "There is a subtle similarity between Darwin's evolution and Senapathy's reuse -- evolution is a form of reuse since each new generation is a copy of the previous, and the reuse of previous organisms' genes is a form of evolution since viable genomes would contribute to later-formed organisms. So, I'm not surprised that these two mechanisms could arguably produce similar results.
Keith: However the main result is something Senapathy "proved" is impossible! If we look at genes with analogous functions between different species, they frequently differ by >10%! So, if Senapathy's equations (along with Jeff's extensions) and their application are correct,>then the reuse theory is just as invalid as evolution.
JM: My turn to say: "Shame on you." :-) You are referring to our old discussion about whether or not a 10% change in a gene by point mutation will result in a new gene. That has nothing to do with reuse of genetic material in the pond, and you know this because you have read Senapathy's book on this subject. They are totally different mechanisms, and there is no relevance to intermediate outcomes in the reuse case because no viable transitional forms of the gene are required.
I wrote: What I'm looking for is a definitive test that will determine which mechanism is correct. ...if you come up with a test, you must consider how it might be passed by both theories. Most of the "evidence" people say refutes Senapathy is, in fact, just evidence that supports evolution -- and support of either theory does not automatically rule out the other. Also, just to avoid confusion, by "evolution," I'm just talking about long-term evolution (natural selection and descent with modification), not short-term (adaptation or artificial selection).
Keith: So we are faced with a choice between a modest extrapolation from observed
events which is consistent with most of the data, vs. a Goldbergian*
scenario requiring enormous assumptions and which isn't consistent
with any data (nor internally consistent)? I know which I'd choose.
* -- as in Rube
JM: Evolution is only consistent with "most" of the data? If evolution is correct, why isn't it consistent with all of the data? It ought to be. Anything contrary would refute the accepted view of evolution, don't you think? If it was consistent, then I would not be here asking these questions about Senapathy's theory, and t.o. would (mercifully) be shut down, too.
And why do you say Senapathy's theory is not consistent with any of the data? Do you really believe that extreme is true? Don't any parts of Senapathy's theory does make sense to you? Can you honestly say that you've evaluated his work with an open mind? (Remember your first reaction? It was total rejection before you read the book -- although I complement you for having the book at all and for presenting your case based on Senapathy's theory rather than assumptions thereof. And now we have come full circle.)
[mi22]
From Don Cates:
Speaking of eyes. They are such useful things that I'm not surprised that there many different types independently evolved (in the pond or in the egg). Looking at the pond; from examination of the world today, I think you would agree that the genes for compound eyes would be quite well represented in the pond. Why is it then that there is not one case of compound eyes outside the insects and their close relatives?
JM: As I have been told here, evolution only works on the random mutations that happen to occur -- no particular mutation is necessarily expected one way or the other. The answer is the same for Senapathy: it only randomly happened once. There is no more reason for it to have occurred under Senapathy than under evolution. Also, the ponds were not homogeneous, and just because some genes worked once does not mean they have to be reused.
[mi23]
From Dave Oldridge: Anyone who has studied the anatomy of cephalopod eyes knows quite clearly the differences between those and vertebrate eyes. Yes, there are similarities, there are also large differences, enough to show that they stem from a very different evolution. And, should you examine the genes that cause them to form, you'll find that they bear little resemblance to one another. If Senapathy's mythical pond was churning out creatures for nearly a billion years, why NOT something like a cephalopod with a vertebrate eye controlled by genes for a vertebrate eye?
And I reiterate. If the pond could last that long in nature, Senapathy can create it in the lab.
It would be a large task [to construct an evolutionary tree of different organisms in the living world based on their eyes], but not really impossible. The trick is to recognize the fact that the eye is such a useful adaptation that it has evolved more than once, right from scratch. Your tree can only extend as far back and as the inception of each eye and can only be as broad as the number of descendants of each eye progenitor.
On the other hand, if Senapathy is right, we could either draw an entire tree, nor none at all, or it should just come out all jumbled. The latter is the most likely scenario, but that's just not what we observe in nature. Furthermore, if you DO that careful work with the eye and determine how many times it evolved, you will have a pretty neat set of subtrees that match closely with cladistic or comparative anatomy trees erected on far different criteria.
Obviously if you take a particular feature that has evolved at some point and build your tree on that, it will never be a full tree. For example, a tree based on notochords would, of necessity only involve phylum chordata.
Scientific theory is rarely EVER consistent with all of the data. Even
general relativity and quantum theory don't come up to that standard.
Here are the usual reasons:
The thing is, "and then a miracle happened," conveniently explains all
POSSIBLE data. It is not, however, a scientific theory. It makes no
general predictions about what can be found in nature or what
experimental results should be expected. There is no way to test it.
Senapathy's theory would explain just about any hodgepodge of data. But it DOES intimate that a pond of gene soup is viable over long periods of time. That leads to the obvious conclusion that such a pond ought to be fairly easy to emulate, if only in miniature. If Senapathy does this and has full-blown multi-celled creatures crawling out of it in the lab, in a repeatable experiment, his theory will gain TONS of respect. Until then, sitting around arguing that it OUGHT to be possible isn't getting it anywhere.
[