More Recent Comments

Friday, August 23, 2013

Reading the Entrails of Chickens

Dan Graur has a recent post on the phylogeny of placental mammals [The Root of the Placental Phylogenetic Tree: Are we Overlooking Something?]. He refers to a recent review in Molecular Biology and Evolution (MBE) that discusses various options. Graur believes that the question has been settled by examining transposon insertions.

But that's not the part that caught my attention. At the end of his post he says,
Finally, there is a small sentence in the Teeling and Hedges commentary that drove me up the wall: “The timing of the splitting event—approximately 100 Ma based on molecular clocks—is not in debate, at least among molecular evolutionists (Hedges et al. 1996…” Actually, dear Blair, it is. And whether you like it or not, both William Martin and I are fine molecular evolutionists.
The reference is to a paper by Dan Graur and Bill Martin—a formidable team that you want on your side because the alternative can be very embarrassing. You really, really don't want to mess with these guys.

We need more papers like this one.

Graur, D. & Martin, W. (2004) Reading the entrails of chickens: molecular timescales of evolution and the illusion of precision. TRENDS in Genetics 20:80-86 [doi: 10.1016/j.tig.2003.12.003] [PDF]

The Graur & Martin paper attacks molecular clock calibrations, pointing out that there are very few solid time points in the fossil record that can be used to calibrate the rate of molecular changes (in years). Here's the abstract ....
For almost a decade now, a team of molecular evolutionists has produced a plethora of seemingly precise molecular clock estimates for divergence events ranging from the speciation of cats and dogs to lineage separations that might have occurred, 4 billion years ago. Because the appearance of accuracy has an irresistible allure, non-specialists frequently treat these estimates as factual. In this article, we show that all of these divergence-time estimates were generated through improper methodology on the basis of a single calibration point that has been unjustly denuded of error. The illusion of precision was achieved mainly through the conversion of statistical estimates (which by definition possess standard errors, ranges and confidence intervals) into errorless numbers. By employing such techniques successively, the time estimates of even the most ancient divergence events were made to look deceptively precise. For example, on the basis of just 15 genes, the arthropod – nematode divergence event was ‘calculated’ to have occurred 1167 +/- 83 million years ago (i.e. within a 95% confidence interval of, 350 million years). Were calibration and derivation uncertainties taken into proper consideration, the 95% confidence interval would have turned out to be at least 40 times larger (~ 14.2 billion years).
They point out that the mammalian radiation is commonly assumed to be about 110 My but there's no justification for that assumption.

The fossil record indicates that mammals (placental and marsupial) were already a diverse group 175 My ago. Two recent papers in Nature document early mammal-like species. One of the papers suggests that mammals may have originated about 215 My ago see Palaeontology: Jurassic fossils and mammalian antiquity]. Like it or not, there's still disagreement between the fossil data and molecular clock estimates.


  1. The fossil record indicates that mammals (placental and marsupial) were already a diverse group 175 My ago. Two recent papers in Nature document early mammal-like species. One of the papers suggests that mammals may have originated about 215 My ago see Palaeontology: Jurassic fossils and mammalian antiquity]. Like it or not, there's still disagreement between the fossil data and molecular clock estimates.

    Wait a minute. Are you actually saying there are fossil members of Placentalia and Marsupialia from 175ma? I am unaware of this, or in fact of any Jurassic crown-group mammals. I have lately seen papers doubting that the placental crown group predates the K/T boundary, and that's a mighty big discrepancy.

    And the recent papers in Nature aren't about mammals but mammaliforms, a considerably earlier node.

    When talking about the ages of groups it's crucial to be clear on just what groups you're talking about. molecular dating can be applied only to crown groups, e.g. Mammalia, Theria, Placentalia. Among other problems, assigning fossils to crown groups can be difficult, which is a source of error for molecular and fossil estimates both.

  2. Did Graur remove the post for some reason?
    The URL you requested could not be found.

  3. @ShadiZ1
    Like you I can't see the blog post.

    The original paper (2004) is here, at no cost:

    I'm convinced.

  4. Perhaps all we can do is operate with a cached version of the post (the original might re-surface with changes I guess but this does get the point across in a brief but nicely barbed manner):

    1. I just started using Feedly to replace Google Reader. I just added his blog, and the article shows up in Feedly, but I cannot navigate to it.

      Feedly is here

  5. They point out that the mammalian radiation is commonly assumed to be about 110 My but there's no justification for that assumption.

    Mmm... no. The divergence between primates and rodents is commonly dated at 110 Mya, according to the article. There are lots of mammals (not to mention mammaliforms) other than Supraprimates (the MRCA of mice and men plus all its descendants).

    1. The more usual name for that clade would be Euarchontoglires

  6. Is there some confusion between stem and crown group going on here perhaps? Of course there were mammal-like reptiles around 175 Million years ago, but crown-group Mammals and Placentals?

    1. The monotreme/therian divergence (that is, the root of Mammalia as a crown clade) may well be that old. We already have Lower Cretaceous monotremes (Teinolophos) and stem-group metatherians (Sinodelphys), as well as a well-preserved Late Jurassic stem-group eutherian (Juramaia). This means that even the age of the crown group Theria must be at least ca. 160 My.

  7. A lot of interesting points here for creationism.
    First the poverty of investigation about conclusions in some aspects here is a creationist lament.
    These guys say it too.
    Then using the fossil record for biological ideas when without the geology the biology wouldn't work and no evidence is given of biology being done except eyeballing fossils.

    Finally it is a error to see the minor matters of marsupialism or placentalism as demanding to define creatures. Therefore also demanding a different evolutionary story.
    Marsupials in the past looked exactly, in many cases, like placentals and this beyond chance.
    They are the same creatures with minor reproductive differences and a few other things.
    They were too quick to conclude that was the big deal. Lack of imagination.

  8. So, robert, if your mom had a pouch and the other particulars of marsupials, would you say that it's all just minor reproductive differences and no big deal?

    When fossil bones, shells, beaks, plants, pollen, tracks, etc., are 'eyeballed' and carefully compared to other fossils or recent/extant specimens, is that biological, or geological?

    When DNA is extracted from fossils and carefully studied/compared, is that biological, or geological?

    Tell me about the biological studies that you or any other creationist have done on yhwh, satan, jesus (including his zombie form), the holy ghost, angels, demons, the flood, a talking snake, a magic tree of good and evil, a salt woman, a dirt man, a rib woman, the animals and people on the ark, the six days of creation, a man who lived inside a fish, etc., etc., etc.

  9. Comparing fossils is just comparing fossils.
    Where is the biology being done about ideas of descent or process of it??
    Its mere observation of biological data points. The inbetween thing , this being the reason for theories, is not observed.
    Therefore the inbetween is not from biological investigation. In using fossils for their confident conclusions is all is based on conclusions from geology. The geology is needed to establish that the segregated deposition events with the fossils are very spaced apart in time.
    the geology is doing the biological investigation. Not the biology.
    thats a grand flaw in evolutionism.

    DNA is a atomic thing. It is also based on presumptions of a steady rate of change in DNA.
    Without this its worthless for telling a genetic origins story.
    YEC says its not a trail but anyways its still just a line of reasoning. DNA conclusions are not based on DNA investigation. Just following the dots.
    Any other mechanism in genetic change will destroy genetics as evidence since it rely's on mere lines of reasoning.

  10. Hey Robert, I want to thank you for opening my eyes to the truth. Now that I really think about it I realize that you're right that the "in between" is just made up by atheists who don't have a clue. There obviously wasn't and isn't any "in between". Everything that has ever lived and happened lived and happened all at the same time!

    The geological features of the Earth indicate nothing about long periods of time except in the minds of blasphemous heathens. Shame on them for thinking and claiming that they can possibly understand the overwhelming creative power of He who shall not be named! How dare they say that there were long periods of time, and especially vastly long ones, "in between" the existence of any of God's creatures, plants, etc. Such a claim is simply arrogant and evil, and there's a fiery eternity waiting for anyone who makes such claims.

    I make the following statements with the utmost humility and respect:

    I sincerely hope that you've arranged for your brain to be donated to science after you die. Just think of the incredible advances science could make if they could study one of if not the finest mind ever specially created. The things that could be learned are too amazing for me to even comprehend. I only hope that science is up to the task of reaping the wondrous good that would undoubtedly be available from such a study. And just think, it would be biological, not geological, because no one could ever legitimately accuse you of having rocks in your head!

    1. Christians don't believe human intelligence is in the brain. Its in the soul. We think the same in heaven as on earth with no need for brains.
      Geology can try to talk about geology but don't say biological conclusions are shown by biology if in fact its geology doing the work.
      Thats my point.
      There is no evidence of the inbetweens between fossils.
      Yet this is what is claimed to be investigated by biology and how they assert their conclusions.
      Its a flaw of methodology.

    2. So there you have it: Byers has no need for a brain. He's the anti-scarecrow. You can't make this stuff up.

    3. I'll take Byers at his word when he claims that his non existent intelligence resides in his non existent soul.

  11. Both molecular clock calculations and creation science tend to rely heavily on not bothering to carefully examine the evidence and hoping that the audience won't know any better. The fossil record for some groups would provide ample calibrations, if one were to take the trouble to do a careful morphological phylogenetic analysis to determine how the fossils and the modern forms are related, to determine what node the calibration actually applies to. Also, you must carefully study the literature to determine the correct ages of the fossils. A search for whatever has been entered into a database does not constitute careful study of the literature. Then, you need to analyze your data and see if the clock assumptions appear to be valid, rather than throwing out any calibrations that don't match. Don't forget to critically examine the molecular data for possible errors. If, after doing all this, you come out with what seem to be reasonably well-supported dates (taking into account all sources of error to have a realistic error bar, not the clock program's calculation of the odds of getting the same results if you make the same assumptions again), then it's once again back t the literature to carefully determine a full range of possible geologic events in that time range that might be relevant.

    In short, practically no molecular clock calculation is trustworthy.

    It should perhaps likewise be noted that Christianity does not claim that intelligence resides in a non-material soul; some Christians hold that position but not others. Some in neuroscience claim to prove that our decisions are fully dictated by genetics and environment, but this is a fanciful extrapolation from the highly questionable premise that a decision that can sometimes be tracked by a brain scan quicker than a person can consciously express it is proof that every decision is determinate. Also, Christianity is incompatible with creation science, as Christianity holds that falsehoods are morally wrong.

    1. ... Christianity does not claim that intelligence resides in a non-material soul; some Christians hold that position but not others.

      Really? Does that mean that some Christians don't believe in a soul or does it mean that some Christians believe in a material soul? Can you show me a picture of the material soul?

      Or perhaps you just meant that because people have souls doesn't mean they are intelligent?

    2. Larry, Some Christian denominations got the concepts of "soul" and "spirit" mixed up. Soul=living, material things-people, animals etc. and spirit=power/energy supporting life. Don't ask me who believe in what though :)

  12. The problem is not so much that there is a shortage of good fossil calibration points, though for soft-bodied organisms that is often the case. The real problem is that molecular biologists rarely do the necessary work of researching what appropriate fossils are known, doing morphological phylogenetics to accurately place the fossil calibration on the tree, critically examining the sequences for possible errors, testing the validity of the clock assumptions used, calculating a realistic error bar incorporating all the sources of error, and carefully researching the full range of possibly relevant geologic factors during that time span. As a result, molecular clocks rarely deserve more credibility than creation science or new atheist history and philosophy.

    It should perhaps also be noted that Christianity does not inherently hold that the soul and intelligence are completely independent of the brain; there are a wide range of views on the exact soul-body-mind relationship. Conversely, the claims that neuroscience disproves the existence of free will or dualism are fanciful impositions of philosophical biases onto less than impressive results. All the neuroscience can say is that some hints of a decision are sometimes detectable by machine sooner than a person consciously articulates it.

    1. Conversely, the claims that neuroscience disproves the existence of free will or dualism are fanciful impositions of philosophical biases onto less than impressive results.
      That's an interesting claim. Suppose for sake of argument the mind-soul-thing is really an immaterial entity, and your decision making is not done by your material brain, but this immaterial soul thing. Information is coming in through your sensory perceptions, then broadcast to your immaterial soul in however fashion, which then makes a decision and sends back the "material response" to your physical body.

      In what way is this any more "free will" than if it's just the brain that does it?

      If the decisions your "soul" does on that sensory input are somehow due to "your immaterial nature"(whatever that nature might be), how is this then different and anymore "free" than if we replace your "immaterial soul and it's nature" with "a material brain and it's nature"?

      What possibilities exist other than chance or necessity here? How does decision making take place by the soul anyway? If you are some specific you, and that you, whatever it is, makes decisions according to it's nature, then you'd be, I assume you would agree, make the same decisions given the same input, whether that decisionmaking entity is immaterial or not?

      Surely there must be some properties of your soul that makes decision making non-random. Otherwise, they're not even decisions of your soul-entity, but the chance events of, well, chance, by definition.

      Tell me about your immaterial free will. Tell me what that even means, tell me how it's free without being randoom. Baffle me with bullshit.

  13. Getting rather off topic on souls, etc., which neither fossilize nor have DNA and so are not too relevant to molecular clocks, but just to clarify: Christianity asserts that we are more than just the physical, but does not have a set, uniform position on how the spiritual nature relates to the physical. Neuroscience, and for that matter anything else that I can think of, neither proves nor disproves dualism or free will. The possibilities are chance, necessity, or that a particular entity is able to weigh various options and select based on what it wants to do, which might not be dictated by chance nor by necessity. Such a position does not require the supernatural; the ability could be an emergent property of a sufficiently complex neural system. Deciding among these possibilities is a matter of philosophical and religious judgment; science doesn't help.