Home » Intelligent Design » The Sound of Taxonomy Exploding

The Sound of Taxonomy Exploding

“The phylogeny of the animals is currently incompletely resolved and has undergone major reorganisations over the past few years, mainly as a result of analyses of rRNA gene sequences”…
Kaboom!

Large-scale sequencing and the new animal phylogeny

Although comparisons of gene sequences have revolutionised our understanding of the animal phylogenetic tree, it has become clear that, to avoid errors in tree reconstruction, a large number of genes from many species must be considered: too few genes and stochastic errors predominate, too few taxa and systematic errors appear. We argue here that, to gather many sequences from many taxa, the best use of resources is to sequence a small number of expressed sequence tags (1000–5000 per species) from as many taxa as possible. This approach counters both sources of error, gives the best hope of a well-resolved phylogeny of the animals and will act as a central resource for a carefully targeted genome sequencing programme.

12/23/06 Update: Here’s another article that describes the problem in even more stark terms:

Bushes in the Tree of Life

Quotes of note (my emphasis and brackets):

Here we discuss how and why certain critical parts of the TOL [Tree of Life] may be difficult to resolve, regardless of the quantity of conventional data available. We do not mean this essay to be a comprehensive review of molecular systematics. Rather, we have focused on the emerging evidence from genome-scale studies on several branches of the TOL that sharply contrasts with viewpoints—such as that in the opening quotation [a quote by Dawkins that implies we'll get the TOL correct eventually]—which imply that the assembly of all branches of the TOL will simply be a matter of data collection. We view this difficulty in obtaining full resolution of particular clades—when given substantial data—as both biologically informative and a pressing methodological challenge. The recurring discovery of persistently unresolved clades (bushes) should force a re-evaluation of several widely held assumptions of molecular systematics. Now, as the field is transformed from a data-limited to an analysis-limited discipline, it is an opportune time to do so.”

Three observations generally hold true across metazoan datasets that indicate the pervasive influence of homoplasy at these evolutionary depths. First, a large fraction of single genes produce phylogenies of poor quality. For example, Wolf and colleagues [9] omitted 35% of single genes from their data matrix, because those genes produced phylogenies at odds with conventional wisdom (Figure 2D). Second, in all studies, a large fraction of characters—genes, PICs or RGCs—disagree with the optimal phylogeny, indicating the existence of serious conflict in the DNA record. For example, the majority of PICs conflict with the optimal topology in the Dopazo and Dopazo study [10]. Third, the conflict among these and other studies in metazoan phylogenetics [11,12] is occurring at very “high” taxonomic levels—above or at the phylum level.

For instance, theory [34] and simulation analyses [8] predict that a small fraction of substitutions will be homoplastic by chance (about 2–5%, depending upon model assumptions and evolutionary distances). However, analysis of the elephant/sirenian/hyrax dataset and the coelacanth/lungfish/ tetrapod dataset indicates that the actual level of homoplasy is ~10% of amino acid substitutions in the first case (178 homoplastic/1,743 total substitutions) and ~15% in the second case (588 homoplastic/3,800 total substitutions), several times greater than expected [8,34]. Similar high levels of homoplasy exist in datasets from other bushy clades [35] (unpublished data) and hold irrespective of analytical methodology [8].

Although it may be heresy to say so, it could be argued that knowing that strikingly different groups form a clade and that the time spans between the branching of these groups must have been very short, makes the knowledge of the branching order among groups potentially a secondary concern.”

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

59 Responses to The Sound of Taxonomy Exploding

  1. When the tree gets reshuffled, I suspect the NDE crowd will say evolution predicted that too. I wonder what creative stories will be told to reassure everyone that science predicted this via some study or computer model.

  2. Other’s posit that even an increase in data will not resolve the problem.

    Can we realistically hope to resolve diversification events spanning a few or even tens of millions of years that occurred in deep time? It is widely accepted that nucleotide data are of limited use for resolving deep divergences because of mutational saturation and homoplasy. Until the recent expansion in available data, it has not been possible to fully explore what the limits of the protein record might be. Like others in the field, we also had expectations that scaling up dataset size would be sufficient to resolve interesting groups. The evidence presented here suggests that large amounts of conventional characters will not always suffice, even if analyzed by state-of-the-art methodology . Just as it would be futile to use radioisotopes with modest half lives to date ancient rocks, it appears unrealistic to expect conventional linear, homoplasy-sensitive sequences to reliably resolve series of events that transpired in a small fraction of deep time. Although we have known this from theory, we are now confronted with the actual pattern of molecular evolution.

    http://biology.plosjournals.or.....io.0040352

  3. I find this very interesting… As I read the various comments here I notice that most (if not all) of you commenting on these issues understand SO MUCH more about biology than I ever will. But recently as I’ve learned more and more about some of these issues, I’ve asked myself this basic question “Is it possible that our system for classifying and categorizing living organisms isn’t even close to being right, particular at the lower end of the hierarchy?”

    And what I gather from this, is that to some extent I was right to ask that question.

    I guess it makes sense that prior to our understanding of DNA and genetics (which is still advancing) we were forced to make certain distinctions between species, genera, order and phyla based on physical characteristics, but now we have a relatively new criteria which just might render part of the taxonomic system obsolete.

    I guess I’m wondering whether I’ve read that correctly and also, just how screwed up could our taxonomic system be? Would I be correct to say that the system for classifying organisms was built totally, or in large part, on a Darwinian presupposition and that a large part of it will have to be discarded once Darwinism is finally sent packing?

    Very interesting… kind of puts us in our place, doesn’t it?

    TRoutMac
    The Intelligent (Graphic) Designer

  4. TRoutMac
    “Would I be correct to say that the system for classifying organisms was built totally, or in large part, on a Darwinian presupposition and that a large part of it will have to be discarded nce Darwinism is finally sent packing?”

    The precursor (in a sense) to the phylogenetic tree were the groupings provided by Linnaean Taxonomy developed by Carolus Linnaeus in the mid 1700′s well before the TOE. Linnaean Taxonomy was (as you mentioned) based on gross morphology. Unlike the phylogenetic tree, the taxonomic groups were not based on a presumption of relatedness, only on form. Linnaean taxonomy was latter adopted as the basis for the first “tree of life” maps of relatedness.

    Phylogenetics is literally the study of evolutionary relatedness. If the TOE were to be “sent packing”, the phylogenetic tree would just be a glorified Linnaean taxonomy structure. No evolution, no phylogenetic tree, or at the very least you would need to change the name.

  5. Jehu, your quote has validated Lurker’s prediction. However, I strongly disagree with Rokas and Carroll in that if the tree cannot be resolved I would see that as a serious charge of error in the fundimental theory — it would cause one to question the simplicity of common descent.

    TRoutMac:

    Would I be correct to say that the system for classifying organisms was built totally, or in large part, on a Darwinian presupposition and that a large part of it will have to be discarded once Darwinism is finally sent packing?

    TRoutMac, it is my understanding that the basic taxonomical system preceeds Darwin. There is clearly something correct about it. However, there seems to be growing evidence that it is not as “perfect” as NDE would anticipate. The taxonomical tree will weather whatever storm ID throws.

    However, this phenomenon causes me to seriously consider what DNA is telling us. In truth, I am having a bit of a hard time understanding how this “bushing” could have happened in the genome. We’ve got to find a good explanation for it. The explanation must be better than Horizontal Gene Transfer, that’s too cliche.

  6. I’m not sure exactly what your point is here, DaveScot. This paper is one more in the growing literature on using molecular methods to reconstruct evolutionary trees (according to a Google Scholar search for molecular phylogeny, 7,130 papers in 2005 and 6,530 papers so far in 2006 – remind me, what is the running total for ID so far?). This particular one is a pretty technical discussion of the appropriate methods to use to get an accurate reconstruction of the deeper branches in the evolutionary tree of the animals. I missed the bit where it said taxonomy had exploded.

    While you’re quoting, how about the opening paragraph of the paper itself:

    “Knowledge of phylogeny is important to biologists for several reasons; most directly, it tells us about the pattern of evolutionary relationships, revealing the historical pattern of speciation and divergence and enabling us to classify life according to a logical, informative, evolutionary scheme. Equally significant, however, is the central place of phylogeny in comparative biology: knowledge of the phylogeny of animals, for example, also tells us about the pattern of evolution of the heritable characteristics of those animals and, therefore, how the great diversity of animals evolved.”

    Doesn’t sound like they think evolutionary biology is about to crumble.

  7. However, I strongly disagree with Rokas and Carroll in that if the tree cannot be resolved I would see that as a serious charge of error in the fundimental theory — it would cause one to question the simplicity of common descent.

    It appears from the article I cited that how a tree is resolved depends upon which features you use. In other words features contradict. So it appears that what happens is they resolve the tree by choosing the most “probable” order of descent based on what pattern agrees with the most features. I do not find that intellectually honest. Even worse, it appears certain features are rejected if they do not line up with the “convnetional wisdom.” Check this out

    Many recent studies have reported support for many alternative conflicting phylogenies [5,6,9,10]. For example, Wolf and colleagues [9] analyzed 507 genes by maximum likelihood, finding support for Coelomata—a clade that joins phyla possessing a true coelom, such as arthropods and chordates, to the exclusion of phyla without one, such as nematodes (left-most tree in Figure 2D). In contrast, Dopazo and Dopazo [10] analyzed 610 genes also by maximum likelihood and, after exclusion of genes evolving at a faster rate in nematodes, found support for Ecdysozoa (rightmost tree in Figure 2D).

    Three observations generally hold true across metazoan datasets that indicate the pervasive influence of homoplasy at these evolutionary depths. First, a large fraction of single genes produce phylogenies of poor quality. For example, Wolf and colleagues omitted 35% of single genes from their data matrix, because those genes produced phylogenies at odds with conventional wisdom (Figure 2D) . Second, in all studies, a large fraction of characters—genes, PICs or RGCs—disagree with the optimal phylogeny, indicating the existence of serious conflict in the DNA record. For example, the majority of PICs conflict with the optimal topology in the Dopazo and Dopazo study [10]. Third, the conflict among these and other studies in metazoan phylogenetics [11,12] is occurring at very “high” taxonomic levels—above or at the phylum level.

    http://biology.plosjournals.or.....io.0040352

  8. Jehu, this data is certainly contradictory to the NDE theory. I am still puzzling, however, as to how a designer would assemble the evidence found. When I think like a software developer, I do not get this result from my software, though for most evolutionary evidence I do. We need a specific ID theory to explain this phenomenon rather than just declaring that it doesn’t fit the current theory.

  9. bFast,

    Well if we use the pattern of things designed by humans we see that they have a mix and match of features. Think about cars or computers for example. Each unique item has many parts in common with other items but then a few parts unique and in different combinations. Imagine what it would be like to attempt to organize a phylogenic tree of computers. Would things get bushy?

  10. Lurker

    if (something_happened = true)
    {
    Darwin said it;
    {
    else
    {
    Darwin predicted it won’t;
    {

  11. My code has a syntax error by the way, Darwin predicted that too ;)

  12. IDist: “My code has a syntax error by the way, Darwin predicted that too ”

    Yea, you never closed any of your left brackets. (an example of IC? the code is useless without the closing right brackets)

  13. Or, rather, the left-brackets after “Darwin said it” and “Darwin predicted it won’t” should be right-brackets.

  14. IDist,
    There’s a problem with your code: It’s designed, thereby requiring a designer. Since this logic is flawed the code was not designed – it only appears so, and you do not exist. True code only has the appearance of design, but has no purpose. If I can remember my C, I think this logic is more fitting, since Darwinism has at most 1 “logical” check:

    While(1

  15. hmm…didn’t include my code, I’ll try it again?

    While(1

  16. While(1 less_than 2){
    Darwin = TRUE;
    }

  17. ahha! It was the less than symbol that was killing me

  18. “It’s designed, thereby requiring a designer. Since this logic is flawed the code was not designed – it only appears so, and you do not exist.” That’s funny!!

  19. bFast:
    “When I think like a software developer, I do not get this result from my software, though for most evolutionary evidence I do. We need a specific ID theory to explain this phenomenon rather than just declaring that it doesn’t fit the current theory. ”

    On another post, you mentioned, in response to something I posted, that molecular clocks were, let us say, calibrated with reference to the fossil record. Well, that is how the clocks begin, but what I was objecting to was the use of MC’s to precisely define divergence moments independent of the fossil record. The paper that Jehu linked to, in one of its introductory paragraphs states that biologists are trying to use the MC technique to move beyond what the fossil evidence presents (IOW, they’re trying to figure out the Cambrian explosion). And the paper makes it clear that they can’t do this. I bristle at their techniques—it’s no more than a guess–because of the very thing the paper points out; that is, there’s too much “noise” in the genome to be able to pick out particular events. But, of course, that should have been obvious from the get go.

    As to ID developing a particular theory, I think this is completely unneccessary. Why? Because what this study points out is that studying “genes” isn’t going to help you figure out evolution. Divergence events–true divergence events–are the result of, if you will, a “reprogramming” of the DNA (I’m not using the word “genome” because it is an equivocal term: it can mean the sequences of nucleotide bases that code for “genes”, or it can mean the entire DNA of an organism–just another example of the sloppiness of terminology that befuddles biology) of organism A, which results in a divergent organism B. This is fully consistent with–and actually predicted by–ID thinking. As to the question: Why is there this “noise”? Well, they’re beginning to find out that what heretofore were thought to be blind, or silent, mutations may, in the end, actually turn out not to be mutations but rather part of a different code, one that is superimposed on the straightforward genetic code we’re all familiar with. This, of course, suggests an even more highly refined overall genetic program, which in turn buttresses the design inference. So all the “homoplasy” plagueing the resolution of identifying branching events may really already be built in (the paper suggests that mathematical modeling points to a rate of “homoplasy” of 5%, yet they find it to be 15%. So the explanation may be that, lo and behold, we’re not dealing with “random events”).

  20. if (something_happened = true)
    Darwin said it;
    else
    Darwin predicted it won’t;

    In the C programming language a single equals sign (=) is an assignment operator. A double equals sign (==) performs a comparison. Thus, in this code example TRUE will automatically be assigned to something_happened, the “if” statement will always evaluate as true, and “Darwin said it;” will always be the result.

    Therefore, no matter what, if something happened, Darwin predicted it. That sounds about right to me.

  21. PaV, “Well, that is how the clocks begin, but what I was objecting to was the use of MC’s to precisely define divergence moments independent of the fossil record.”

    PaV, the consern I had the other time I posed was that the chatter was getting too black and white. Molecular clocks have their significant weaknesses, therefore the whole idea of a molecular clock holds no merit whatsoever. I think it is well understood in scientific circles that during divergence molecular clocks go out of whack. The term used, I believe, is the “neutral theory”. The “neutral theory” generally states that if a gene is not mutating for function, it will drift according to a clock.

    From my reading of the article, which I admit is more skimming than reading, they are not using molecular clocks, but specific mutations in common, to determine the phylogenic tree. If gorillas and humans both have a specific mutation, and chimps don’t, then humans and gorillas diverged after the chimp. If chimps and humans both have a specific mutation and gorillas don’t then gorillas diverged before the chimp/human split. If I understand correctly, this analysis of the genes show that the gorilla is a closer cousin than the chimp about 5% of the time.

    As to ID developing a particular theory, I think this is completely unneccessary. Why? Because what this study points out is that studying “genes” isn’t going to help you figure out evolution.

    Here you and I diverge drastically. Firstly, I generally hold respect for the scientific community’s frustration with the “God did it” position. If ID is to be respected as science, it must be more than just a “NDE didn’t happen” argument. It must present answers, that make sense out of the puzzles that we are seeing in gene studies. “Studying “genes” isn’t going to help you figure out evolution” isn’t right eather. Studying genes is helping us figure out evolution. It is helping us to see that NDE is full of it. However, when we abandon the NDE paradyme, and find a new paradyme to shift into, these very gene studies will have helped us enormously in figuring out evolution. Evolution happened, I rather sure of that. How it happened, well, not by a filtering of random chance events, at least is sure doesn’t seem so from where I sit.

  22. Oops. I should correct my next-to-last sentence: Therefore, something always happens, and Darwin always predicts it.

  23. ” In the C programming language a single equals sign (=) is an assignment operator. A double equals sign (==) performs a comparison. Thus, in this code example TRUE will automatically be assigned to something_happened, the “if” statement will always evaluate as true, and “Darwin said it;” will always be the result.”
    Brilliant :D

    “It’s designed, thereby requiring a designer. Since this logic is flawed the code was not designed – it only appears so, and you do not exist.”

    I do exist, I am THE BLIND C CODER.

  24. ” In the C programming language a single equals sign (=) is an assignment operator. A double equals sign (==) performs a comparison. Thus, in this code example TRUE will automatically be assigned to something_happened, the “if” statement will always evaluate as true, and “Darwin said it;” will always be the result.”

    But there is also else branch, for redundancy reasons.

    Isn’t evolution wonderful?

  25. bFast:

    PaV, the consern I had the other time I posed was that the chatter was getting too black and white. Molecular clocks have their significant weaknesses, therefore the whole idea of a molecular clock holds no merit whatsoever.

    I was questioning the seeming certitude with which the MCH has been used. I’m not convinced that it has much utility at all. Rokas and Carroll’s paper makes the MCH sound like it doesn’t have the fire-power needed to ferret out true divergences.

    From my reading of the article, which I admit is more skimming than reading, they are not using molecular clocks, but specific mutations in common, to determine the phylogenic tree. If gorillas and humans both have a specific mutation, and chimps don’t, then humans and gorillas diverged after the chimp. If chimps and humans both have a specific mutation and gorillas don’t then gorillas diverged before the chimp/human split.

    Here’s what Wikipedia has to say about the “molecular clock”:
    “The molecular clock (based on the molecular clock hypothesis (MCH)) is a technique in genetics, which researchers use to date when two species diverged. It deduces elapsed time from the number of minor differences between their DNA sequences. It is sometimes called a gene clock.”

    And, more to the point:
    “The molecular clock technique is an important tool in molecular systematics, the use of molecular genetics information to determine the correct scientific classification of organisms. Knowledge of approximately-constant rate of molecular evolution in particular sets of lineages also facilitates establishing the dates of phylogenetic events not documented by fossils, such as the divergence of living taxa and the formation of the phylogenetic tree.”

    The Rokas and Carroll write:
    “The discipline primarily responsible for assembling the TOL–molecular systematics–has produced many new insights……”

    The “noise” I was speaking of is, essentially, random drift. The authors seem to be saying that unless the stem part of the TOL (that is, the part of the tree that is between divergence events) is large relative to the branches, then “homoplasy” masks the genomic differences that would otherwise indicate divergence. They define “homoplasy” as: “Shared characteristics found in different branches of a phylogenetic tree not directly inherited from a common ancestor; these may arise by chance or selection.” If you’re talking about “chance”, then that is “neutral/genetic drift”.

    “Studying “genes” isn’t going to help you figure out evolution” isn’t right eather. Studying genes is helping us figure out evolution. It is helping us to see that NDE is full of it.

    Well I certainly agree that studying genes is knocking the feet out from underneath NDE. Nonetheless, I see the genes as simply being a “parts list” for the construction of some building. Every home in southern California where I live is made out of mostly the same materials; yet their sizes, shapes, designs, functionality, etc., are all different. So what drives divergence is not a “parts list” but “blueprints”. IOW, to figure out evolution, it seems to me that we have to have some idea of the regulatory mechanisms behind organismal differentiation, and the study of “genes” is simply missing the point. That isn’t to say, however, that individual studies aren’t important since, as you point out, they might prove to be very useful once a more reasonable paradigm becomes visible and accepted. So I think we’re in agreement on that point. Yet, I still think that ID is useful, even in the abstract. For example, why spend all this time and effort to use, more or less, molecular clock techniques to discover divergences if you believe that fundamental divergences are at a regulatory level rather than down at the “parts list” level? Think of all of the time, energy, money that has been flushed down the toilet trying to use “molecular systematics” to single out divergence events when, as the authors–almost sadly–attempt to point out: “Although it may be heresy to say so, it could be argued that knowing that strikingly different groups form a clade and that the time spans between the branching of these groups must have been very short, makes the knowledge of the branching order among groups potentially a secondary concern.”

    Suffice it to say that IDists would never have gone on this wild-goose chase. Only those convinced that cosmic rays, intersecting with DNA over eons of ages, has brought about the fecundity of life we see would have gone (and did go) down that blind alley.

  26. Jehu– about features contradicting:

    If you look at a single gene, you may get false positives. I’m blue-eyed, but my brother isn’t. If you looked only at the eye-color gene, you’d think that I’m far more closely related to some random Scandinavian than I am to my own brother. (Which, I think you’d agree, is contrary to conventional wisdom.)

    I believe that chimps have the same ABO blood type system that humans do; if you compared a type O human to a type A human and a type O chimp, you’d think the type A human was the outlier. This is why trying to build a phylogeny from a single gene, or even a small number of genes, is liable to give very odd results.

  27. grendelkhan,

    This is why trying to build a phylogeny from a single gene, or even a small number of genes, is liable to give very odd results.

    It is not that simple. Please read this review if you haven’t.
    http://biology.plosjournals.or.....io.0040352

    Pay attention to the conclusion that even large amounts of data will not resolve the problem.

  28. What is the problem, again, exactly? The fact that the tree of life may be shaped more like a rhododendron than an oak? If it is, it is. What does this have to do with design?

    Meanwhile, if this is a typical day, another 19 papers were published today using DNA sequences to piece togther more branches of thetree of life.

  29. grendelkhan

    I’m blue-eyed, but my brother isn’t. If you looked only at the eye-color gene, you’d think that I’m far more closely related to some random Scandinavian than I am to my own brother.

    Actually that’s not at all true. I just read an article on eye color and the difference between blue and brown is a few nucleotides on a single gene. If you were to compare that gene between blue-eyed you and your brown-eyed brother it would be a very close match including synonymous substitutions. If you compared it to the same gene in a distantly related human who is also blue-eyed you could expect to find a number of differences in synonymous substitutions. This is based on the neutral theory and is the basis of genetic clocks – the more distantly related the more synonymous substitutions. The fly in the ointment is that not all genes even in the same organism experience the same background rate of synonymous substitution and when you start comparing different species the rates become even more skewed. Thus every molecular clock requires “calibration” based upon some ostensibly more reliable method of dating. The calibrations can become extreme – orders of magnitude differences in background rates must be factored into the comparative equation in some cases.

    See here for more in depth discussion.

  30. Anthemis, when 5% of the genes studied say that the gorilla is a closer cousin to us than the chimp is, that does not fit the NDE theory. The issue is not the shape of the oak/aspen/peach, its the fact that genetic studies are not supporting a nice clean common ancestry map. The map that is being presented looks more like the life is the product of an active genetic engineer.

  31. The fact that 5% of genes support alternative phylogenetic relationships is certainly not unexpected in a common descent framework. Depending on historical parameters such as ancestral population sizes and divergence times, as well as levels of hybridization and introgression during the initial period of divergence, coalescent theory predicts that a substantial proportion of the genome can reveal incongruent phylogenies. For example, if the size of a common ancestral population (or of the two diverging populations) is fairly large, genetic drift will act much more slowly, and lineages will remain in the diverging populations that reflect the ancestral state rather than the actual population history. The same effect can appear if internodes are short. In other words, genes can show discordant patterns if branches are short and fat. The problem with traditional systematics is that it often ignores population genetics, and the process of speciation is inherently an population genetics problem. There has been a ton of work on this issue but for a good early review see Maddison 1997 (syst biol).

  32. I’m not sure that the situation is as desperate as you seem to believe, bFast. It makes sense to me that it would be harder to reconstruct either very recent divergences such as the gorilla/chimp/human one (because there has not been enough time for sequences to diverge much) or very ancient ones such as those between the major animal phyla (because there has been so much time that divergence has overwritten most of the original resemblances). But over vast tracts of the middle ground between very ancient and very recent divergences, there has been massive progress piecing together the relationships. Molecular methods don’t have any trouble establishing, say, that gorilla, chimps, and humans, as a clade, are more closely related to each other than they are to lemurs or hedgehogs. Or take a look at this paper that recovers large amounts of well-supported tree-like structure in the largest flowering plant family:
    http://www.mnh.si.edu/biodiver.....ertree.pdf
    As an added bonus in this one, there is lots of biogeographic structure that is consistent with the phylogeny, just as you would expect if different branches of the tree had diversified in different parts of the world.

    It may not all yet be “nice and clean” – real life isn’t usually. But there are certainly large amounts of hierarchical structure there that looks awfully like a genealogy, whether it’s oak-tree like or bush-like, and more of it is coming into focus every day. (This year’s total of papers for “molecular phylogeny” on Google Scholar is up to 8,310 today).

    I would be interested to know what you, or other posters here, predict about what the ancestry map would look like if life were “the product of an active genetic engineer”. Please be as specific as possible in explaining how you derive your predictions about this from your hypothesis.

  33. I would be interested to know what you, or other posters here, predict about what the ancestry map would look like if life were “the product of an active genetic engineer”. Please be as specific as possible in explaining how you derive your predictions about this from your hypothesi

    Just off the top of my head, you would except the following:

    1. Numerous features of common design similar to those in similar organisms but slightly fine tuned for the specific organism.

    2. Some unique features not found in similars orginisms.

    3. Some features not found in similar organisms but found in unsimilar organisms

    3. Features will follow more of a mosiac pattern than a tree pattern.

    4. Numerous features that confer a slight benefit but not a “selective advantage.”

  34. Please be as specific as possible in explaining how you derive your predictions about this from your hypothesis.

    I derive my predictions because it is the pattern I observe in designed things.

  35. The fact that 5% of genes support alternative phylogenetic relationships is certainly not unexpected in a common descent framework.

    With the human/chimp/gorilla clade it is not 5% that support alternative phylogenic relationships , it is 21% of parsimony informative characters. Is that expected from a model of common descent?

  36. With all those explosions nowadays, it sounds as if evolutionists are almost finished digging their own grave.

    With so much evidence against NDE being dug out by the evolutionists themselves all the time there hardly is much need for ID research.

    I have always been intrigued by this passage by Behe in DBB:

    “The result of these cumulative efforts to investigate the cell – to investigate life at the molecular level – is a loud, clear, piercing cry of “design” The result is so unambiguous and so significant that it must be ranked as one of the greatest achievements in the history of science. The discovery rivals those of Newton and Einstein, Lavoisier and Schrödinger, Pasteur and Darwin. The observation of the intelligent design of life is as momentous as the observation that the earth goes around the sun or that disease is caused by bacteria or that radiation is emitted in quanta. The magnitude of the victory, gained at such great cost through sustained effort over the course of decades, would be expected to send champagne corks flying in labs around the world. This triumph of science should evoke cries of “Eureka!” from ten thousand throats, should occasion much hand-slapping and high-fiving, and perhaps even be an excuse to take a day off.”

    Now, ten years later it seems that the time for science to wake up is just around the corner.

  37. “With the human/chimp/gorilla clade it is not 5% that support alternative phylogenic relationships , it is 21% of parsimony informative characters. Is that expected from a model of common descent?”

    Indeed this is also possible. Theory predicts some situations where, under the right conditions, >50% of the genome will actually support the WRONG topology. In these situations it is not a problem with common descent or with molecular phylogenetics, but simply an artifact of population demographic history. In other words, there may be problems that are difficult or unsolvable using nucleotide composition data alone, but this does not mean that speciation in a common descent framework did not occur.

  38. coalescent

    So you’re saying that no matter what NDE can explain it.

    We here already knew evolution can explain everything. It can’t predict future evolution at all but it can explain every finding ad hoc.

    That’s sure some powerful theory, huh?

  39. I’m saying that using coalescent theory (which is a straightforward and well accepted theory that traces a set of sampled alleles back through time probabilistically generation by generation to a common ancestor and can also be used to simulate data under a specified topology), demographic histories can be found that explain a vast range of observed data. The trick for a given sample of genes, is to find the history that most likely would have generated the observed data. In some cases using standard systematic/phylogenetic methods that don’t take these population processes into account will result in the wrong topology. It is somewhat analogous to long branch attraction with parsimony, where distant lineages will attract each other in tree reconstruction because of multiple hit mutations (homoplasy). Evolution is complicated, and there is no way we can solve all problems with one method, but at least we are working on it.

    We do not search for explanations ad hoc. we use existing theory to best explain the observations. And we recognize when a problem is unsolvable with the available data. Then we go out and get more data. I am not trying to ‘predict ‘ what is going to happen in the future in the sense that you mean. But by trying to reconstruct the past we might find evolutionary patterns that will help us predict what will happen to given species under changing climate or habitat fragmentation, for example, or id areas where we might expect more new species to arise, etc…

    i’d like to see how ID is more predictive … i guess “poof it was designed” is a pretty good ad hoc argument. but how does this PREDICT anything hmmm?

  40. Yeah, “poof an accident happened” is so much better.

  41. We know the pattern of design because we have things designed by humans all around us. Computers, cars, appliances, wirelss phones. They have designed elements and specific parts in common don’t they? If you wanted to you could arrange them into nested hierarchies couldn’t you? Yes, very easily. All of these statments are also true about biological organisms found in nature. The pattern of nature fits the pattern of design.

    The patterns also match if you attempt phylogenies. If you tried to arrange things designed by humans into phylogenies you would become frustrated by the mosiac pattern that the unique and shared features form. (just for the sake of argument because obviously such things don’t reproduce) Instead of a phylogenic tree you would get no further then a bush, where the branches don’t really connect. And this is exactly what happens with biological oranisms found in nature. You end up with branches that don’t really connect. So again, the pattern in nature reflects the pattern of design.

  42. On the contrary, it is not usually possible to arrange human-designed artifacts into an unambiguous nested hierarchy. Try it with cars – should you make the first level split according to engine size, country of origin, manufacturer, fuel source, transmission type, body style, passenger capacity, etc., etc.? Choosing any of these as the primary split will give you a different nested hierarchy, and there is obviously no particular reason to claim that one of these is the “right” or the “natural” one. Any attempt to use this to develop a phylogeny of cars would be futile, because cars have not evolved by a process of common descent with modification from a single ancestral car. They have been individually designed by engineers who could pick and choose design elements from the whole range of possibilities and combine them in arbitrary ways. You want diesel, you want front-wheel drive, you want it to be a convertible, you want manual transmission …? Yep, we can do that for you. Just about the whole matrix of possible combinations has been tried or could be tried.

    This is very, very, very, very far from being the case with living organisms. Any given species is put together using a tiny fraction of the possible design solutions available over the whole space of living organisms, and they are combined in a very limited and stereotyped number of ways. This is what makes it possible to do cladistic analysis.

    Take the springtails. This is a group of small arthropods, most of whom can jump vigorously. They do this using an organ called the furcula on the underside of the abdomen, which is flipped sharply away from the body to propel the animal off the substrate. Jumping is clearly a useful thing to be able to do to escape your enemies or attack your prey, and lots of other animal groups have specialized in it: fleas, flea beetles, salticid spiders, grasshoppers, frogs, gerbils, kangaroos, etc. There are about 8,000 known species of springtails. A few of these have no furcula, but are still clearly and recognizably springtails in every other way. So let’s suppose there are 7,500 springtail species with furculas. I don’t know how many other animal species are specialized jumpers, but it is clearly a lot. Let’s say there are 20,000 of them. EVERY springtail that jumps has a furcula. NO OTHER JUMPING ANIMAL SPECIES ON THE FACE OF THE EARTH has a furcula.

    The common-descent explanation of these observations requires one uncontroversial fact (that only animals with furculas will have offspring with furculas), and one hypothesis: that all living springtail species are descended from an ancestral springtail species that had a furcula. The design explanation requires 27,500 hypotheses, and it requires them to fall into a very specific pattern. It requires that when each of the 7,500 springtail species was designed, and the time came to implement the jumping function, the designer reached into the tool box and said “OK, let’s give this one a furcula.” When designing each of the 20,000 other jumping animals, he reached in and said “I think we’ll skip the furcula this time – let’s try something else. Maybe we’ll give it expanded muscles in the hind femur.”

    So in terms of observations explained per hypothesis, the evolutionary explanation in this case has 27,500 times the explanatory power of the design explanation. Note, too, that the common-descent explanation is entirely falsifiable, and therefore testable. This one would be falsified by a single observation of a beetle or a spider with an unambiguous springtail-type furcula. You can choose whichever you like, but general in science we choose single, simple, powerful, testable, explanations over multiple, ad-hoc, weak, untestable ones.

    There is nothing whatsoever unique or special about this example. The same account could be given for just about any typical character of any taxon. This is why the evolutionary hierarchy of common descent is now almost universally recognized as the fundamental organizing principal of biological taxonomy.

    Oh, and to forestall one possible complaint, I hear someone say “Your argument is circular because you are using the presence of a furcula to decide whether an animal is a springtail or not”. Sorry, not so. Springtails have plenty of other defining characters by which they can be recognized, and a springtail with the furcula chopped off is still very recognizably a springtail.

  43. anthemis

    it is not usually possible to arrange human-designed artifacts into an unambiguous nested hierarchy

    If you’d read the article you’d discover we are finding a lot of ambiguities in the tree of life now as we do it by genotype instead of phenotype.

    In regard to taxonomy by phenotype you’re failing to mention all the many instances of so-called convergent evolution. ToE has all these convenient cop-outs to explain the exceptions to the rules like the rule of nested hierarchies. Wings, for instance, are inherited so that all winged creatures fit into a nested hierarchy, except when they don’t then it’s convergent evolution so you can have bugs, bats, and birds all in separate groups yet all sharing a common design feature.

  44. Anthemis,

    On the contrary, it is not usually possible to arrange human-designed artifacts into an unambiguous nested hierarchy. Try it with cars – should you make the first level split according to engine size, country of origin, manufacturer, fuel source, transmission type, body style, passenger capacity, etc., etc.? Choosing any of these as the primary split will give you a different nested hierarchy, and there is obviously no particular reason to claim that one of these is the “right” or the “natural” one.

    You actually make my point. Designed things fit into nested hierarchies as well or as bad as natural things. In order to create a phylogenetic tree, things like horizontal gene transfer, convergent evolution, and homoplasy to make it work and even then specific branches cannot be resolved because genetic parsimony informative characters contradict.

    For example, let’s look at what the “tree of life” looks like for the eukaryote/bacteria/archea clade.

    http://bp0.blogger.com/_DZH2cm.....f_Life.jpg

  45. Sorry, that last post was botched.

    Anthemis,

    On the contrary, it is not usually possible to arrange human-designed artifacts into an unambiguous nested hierarchy. Try it with cars – should you make the first level split according to engine size, country of origin, manufacturer, fuel source, transmission type, body style, passenger capacity, etc., etc.? Choosing any of these as the primary split will give you a different nested hierarchy, and there is obviously no particular reason to claim that one of these is the “right” or the “natural” one.

    You actually make my point. Designed things fit into nested hierarchies as well or as bad as natural things. In order to create a phylogenetic tree, things like horizontal gene transfer, convergent evolution, and homoplasy to make it work and even then specific branches cannot be resolved because genetic parsimony informative characters contradict.

    For example, let’s look at what the “tree of life” looks like for the eukaryote/bacteria/archea clade.

    http://bp0.blogger.com/_DZH2cm.....f_Life.jpg

  46. It looks like humans have twice as many regulatory genes in common with sharks than fish, even though sharks supposedly diverged from the human lineage 530 million years ago .

    Cartilaginous fishes represent the living group of jawed vertebrates that diverged from the common ancestor of human and teleost fish lineages about 530 million years ago. We generated ~1.4x genome sequence coverage for a cartilaginous fish, the elephant shark (Callorhinchus milii), and compared this genome with the human genome to identify conserved noncoding elements (CNEs). The elephant shark sequence revealed twice as many CNEs as were identified by whole-genome comparisons between teleost fishes and human.

    Ancient Noncoding Elements Conserved in the Human Genome http://www.sciencemag.org/cgi/...../5807/1892

    What would be really cool is if they did a knock-out study of those NCE’s, like with the mouse, and found no significant difference.

  47. Replying to DaveScot’s comment#43: I’m not sure if you are referring to the Rokas and Carroll article or to Philippe and Telford. However, I have in fact read both of these articles, as well as many others. (No-one can read them all, when they come out at a rate of 7,000 or 8,000 per year.) Nowhere do they assert that the ongoing effort to reconstruct the tree of life using phylogenetic methods has been futile or unproductive. Nowhere do they assert that characters are distributed across the animal and plant kingdoms in the sort of random, matrix-filling way that would make it impossible to do cladistic reconstruction. Indeed, they point to the impressive progress that has been made in assembling large parts of the tree, such as recognizing the clade Afrotheria among the mammals, to pick one example out of thousands that they could have chosen. The general outlines of the flowering plant tree, for instance, are largely complete (look up the work of the Angiosperm Phylogeny Group).

    Because these are both review-type articles, they focus on challenging areas where new types of data and analysis may be needed to resolve questions that have not yet been settled. The type of problems that these papers address are generally to do with the order in which these large, already-assembled portions of the tree should be in turn connected at their bases, reflecting evolutionary events that took place in the very distant past. The Philippe article suggests a new type of data that could be used to clarify some of the points that have not been settled by the methods used to date. This is what scientists do all the time, but it takes time. I personally feel that a more fertile approach than theirs might be the use of so-called rare genomic changes such as large deletions, insertions or transpositions in the genome. These are analogous to whole-word insertions or deletions, and therefore have a much higher information content than the single-letter level nucleotide changes used in most phylogenetic reconstructions to date. (Rokas, A., and W.H. Holland. 2000. Rare genomic changes as a tool for phylogenetics. Trends in Ecology & Evolution 15: 454-459.)

    I am also well aware of convergent evolution, as are the authors of these articles. Wings are indeed the classic example. The evolutionary explanation is that true wings originated four times, in insects, birds, bats, and the extinct pterosaurs. (Of course, a detailed examination of wings from these four groups shows that, while alike in function, they are radically different in their underlying structure.) Thus, the evolutionary explanation requires 4 hypotheses. The design explanation requires about 761,600 hypotheses, one for each of the independent choices that the designer made to put wings on each species that has them (about 1,100 species of bats, 10,000 species of birds, maybe 750,000 species of insects, and a few hundred pterosaurs). Actually even this is an underestimate, as it leaves out many undescribed or extinct species, as well as all the design choices that were made not to put wings on countless thousands of other species that could have used them. Thus at a conservative estimate the evolutionary explanation is 190,400 times more powerful than the design one, in terms of observations explained per hypothesis.

    Biologists do not throw convergent evolution around at will as a get-out-of-jail-free card. The algorithms that are used to reconstruct phylogeny are explicitly based on the principle of maximum parsimony, which means, precisely, finding the smallest possible number of instances of convergent evolution that you have to hypothesize to explain the observed data. If you look in any of the dozens of papers published every week on phylogenetic reconstruction you will see examples of “maximally parsimonious trees”.

  48. In reply to Jehu, comment #45: with all due respect, I do not actually make your point, nor do I agree with it, as suggested by the subtle clue in my next sentence “This is very, very, very, very far from being the case with living organisms.” Living organisms overwhelmingly do fit into a nested hierarchy in the way that human-designed artefacts, as we both agree, do not.

  49. Anthemis

    The design explanation requires about 761,600 hypotheses, one for each of the independent choices that the designer made to put wings on each species that has them

    This is incorrect. ID has no argument with descent with modification. But that’s beside the point since the article made no mention of ID at all. It merely underscores the fact that genotype is often at odds with phenotype and has caused a major reorganization of the tree of life in recent years. Please stay on topic.

  50. DaveScot: I realize that the articles did not mention ID, but some of the commenters here – perhaps not yourself – appear to think that they support the ID position. For instance PaV says “Suffice it to say that IDists would never have gone on this wild-goose chase”, bFast says “The map that is being presented looks more like the life is the product of an active genetic engineer”, and Jehu says “You end up with branches that don’t really connect. So again, the pattern in nature reflects the pattern of design.”

    However, it is interesting to know that at least some ID proponents accept common descent. I still think that ID requires far more ad-hoc hypotheses than common descent does.

    Incidentally, I don’t think genotype versus phenotype is really the issue here. Both morphological and molecular data can be used in cladistic analysis and the approach is essentially similar for both types of data. For the reasons already discussed, it is possible for there to be apparent inconsistencies within either of these types of study, as well as between them. Many studies combine morphological with molecular data, the “total evidence” approach.

  51. Anthemis,

    Just becuase the theory of common descent can wave its hand and claim a simple explanation for common feature doesn’t mean the evidence actually fits the pattern of common descent. And although your theory sounds quite simple in a High School text book, in real life, after you are done with the special pleading horizontal gene transfer, homoplasy, and convergence your theory isn’t nearly so tidy and in fact still cannot determine where the branches of the tree of life actually connect.

    Have you even bothered to see what the “tree of life” looks like for the eukaryote/bacteria/archea clade? I posted the link.

    http://bp0.blogger.com/_DZH2cm.....f_Life.jpg

    This is an elegant theory?

  52. The design explanation requires about 761,600 hypotheses, one for each of the independent choices that the designer made to put wings on each species that has them (about 1,100 species of bats, 10,000 species of birds, maybe 750,000 species of insects, and a few hundred pterosaurs).

    That is simply not true.

  53. Jehu, re comment 51: I could not agree more that many things are more complex in real life than in high school textbooks: evolutionary biology is certainly one of them. I don’t have any particular preconception as to how tidy, simple, or elegant the real evolutionary relationships of living organisms and their ancestors will turn out to be – the important thing is to reconstruct them in a way that best accounts for the evidence with the smallest possible number of ad-hoc hypotheses.

    Regarding the eukaryote/bacteria/archaea clade, those interested can find your diagram in context (always a helpful thing to do) in a 2000 Scientific American article posted at http://shiva.msu.montana.edu/c.....ooting.pdf. The caption indicates that it is intended as a schematic or symbolic representation of the hypothesis that there was “rampant lateral gene transfer” between the various unicellular lineages that later gave rise to existing prokaryotes and eukaryotes. This is an article in a popular scientific magazine that proposes an idea of how the TOL will look when its roots are better known. To find out which of these links are currently supported by hard evidence, it is necessary to get into the primary research literature.

  54. Jehu, re comment 52: it would help to clarify the basis of our disagreement if you could indicate which component(s) of my assertion you believe are not true.

    (A) a statement like “the designer chose to provide the Asian tiger mosquito with wings” is a hypothesis.

    (B) a statement like “the designer chose to provide the monarch butterfly with wings” is a different hypothesis to the one mentioned in (A)

    (C ) these types of hypotheses are required by ID theory

    (D) the choices made by the designer to provide different species with wings were made independently of each other

    (E) there are exactly 761,600 species of living and extinct winged animals (of course there are not, but if you have a more accurate estimate of this number you are welcome to substitute it).

  55. Anthemis,

    Thanks for finding a copy of the Scientific American article. I was aware the diagram was from that article but I couldn’t find it on online.

    Note that from this article, Doolitte does not believe the evidence supports common descent from a single common ancestor. As the caption states,

    This “tree” also lacks a single cell at the root; the three major domains
    of life probably arose from a population of primitive cells that differed in their genes.

    The article notes that the diagram is actually “misleadingly simple.”

    Though complicated, even this revised picture would actually be misleadingly simple , a sort of shorthand cartoon, because the fusing of branches usually would not represent the joining of whole genomes, only the transfers of single or multiple genes. The full picture would have to display simultaneously the super-imposed genealogical patterns of thousands of different families of genes (the rRNA genes form just one such family).

    Unless you have drunk the cool-aid, you have to realize that there is no evidence of common descent in the eukaryote/bacteria/archaea clade. You have to come up with so many hypothesis and theories to explain how the mosiac of features arose by common descent that your claim that common descent is powerful because of its “observations explained per hypothesis” completely fails.

  56. Anthemis,

    In reply to Jehu, comment #45: with all due respect, I do not actually make your point, nor do I agree with it, as suggested by the subtle clue in my next sentence “This is very, very, very, very far from being the case with living organisms.” Living organisms overwhelmingly do fit into a nested hierarchy in the way that human-designed artefacts, as we both agree, do not.

    Actually we do not agree. You in fact made my point originally. Let’s look at what you actually said.

    On the contrary, it is not usually possible to arrange human-designed artifacts into an unambiguous nested hierarchy. Try it with cars – should you make the first level split according to engine size, country of origin, manufacturer, fuel source, transmission type, body style, passenger capacity, etc., etc.? Choosing any of these as the primary split will give you a different nested hierarchy, and there is obviously no particular reason to claim that one of these is the “right” or the “natural” one.

    To begin with, this is a problem with biological organisms as well. For example, according to the Rokas article, in the elephant/sirenian/hirax clade 64% of parsimony informative mitochondrial features support and alternative tree. In the chordate/arthopod/nematode clade, in a study of over 1000 parsimony informative characters, 57% supported an alternative tree. And as we have already seen, in the bacteria/eukaryote/archae clade there is not tree to speak of.

    Now with cars, I would wager that putting cars into makes, models, and years would probably create a nested hierarchy consistent with the highest percentage of parsimony informative characteristics. Many features will obviously contradict but hey – that is how it is with biology as well.

  57. Re comment 56: I apologize if I misunderstood you. You posted this statement on Christmas afternoon:
    “If you tried to arrange things designed by humans into phylogenies you would become frustrated by the mosiac pattern that the unique and shared features form” and this is the one I agree with. Your position now is that “putting cars into makes, models, and years would probably create a nested hierarchy consistent with the highest percentage of parsimony informative characteristics”, and this I do disagree with. My wager would be that this exercise would not be able to find any significant phylogenetic structure in the data, and that the resulting cladogram would be very close to an unresolved polytomy (i.e. a comb-like diagram with no two “species” more closely related to each other than to the rest).

    However, what you or I would wager is not really evidence. I think this would be a genuinely interesting and instructive exercise to try, and the results should be quite publishable. I hope that someone reading this will be inspired to try it.

  58. Anthemis,

    Okay I take the blame for not communicating well, I am trying to use two different definitions at once. Let me give you some background on nested hierarchies.

    Nest hierarchies were invented by Carolus Linnaeus who understood biological organisms to be designed. He wrote in the preface to a late edition of Systema Naturae:

    Creationis telluris est gloria Dei ex opere Naturae per Hominem solum — The Earth’s creation is the glory of God, as seen from the works of Nature by Man alone. The study of nature would reveal the Divine Order of God’s creation, and it was the naturalist’s task to construct a “natural classification” that would reveal this Order in the universe.

    Under Linneaus’ system not every feature had to fit because it was an “artificial classification.” You just needed to get things in a general groups. In this use of the term, there is no problem putting designed things into nested hierarchies.

    When nested hierarchies are used in the Darwinian sense, features need to line up much more stringently. If the theory is correct, you should be able to describe the exact branching of the tree. But in fact you can’t, you can’t get any closer than an approximation, which is no better than the system Linneaus divised under the assumption of design.

    The bacteria/eukaryote/archae is perfect example. It is easy to place organisms into one of the three groups, but good luck trying to find the tree from whence they came.

  59. Thank you for the primer on the history of biological classification systems. I am reasonably familiar with this topic, but hopefully the information will be useful to others who may be following this thread.

    Today is a little busy in the real world, but I look forward to continuing our illuminating discussion soon, and trust in your patience in the mean time.

    If you have a little time today, I would still be interested in your response to my comment #54, regarding hypotheses in ID explanations.

Leave a Reply