OK, I’ll bite


A creationist on twitter is pestering me to ‘debate’ her. Here’s a sampling of her arguments and style.

No way can we have a conversation about that in a 140 character limit. So I’m telling her to come here and expand on her ideas, if she can.

But first, I’d like her to answer a little quiz to see if she’s worth wasting time on.

  1. What is a typical mutation rate? (be specific about the units!)

  2. Are most mutations deleterious, neutral, or beneficial?

  3. What is the difference between phenotype and genotype?

  4. What is the difference between mutation and “deformity”?

  5. What is “elephant disease”?

That’s a start. I’m going to tell her on Twitter to come here and address these very simple questions. Anyone want to take bets on whether she bothers?

Comments

  1. kantalope says

    If a tiger bites your leg off. Is that an animal induced deformity? Do they think that would be a mutation?

  2. kurt1 says

    Invite her to a google+ discussion. Would be interesting to have some of the usual confused people on there, trying to defend whatever quaint idea is roaming their heads and being disproven live.

  3. Ogvorbis: broken and cynical says

    A tiger? In Africa?

    Well, to us evilutionsticistismists, a lion and tiger are the same thing. After all, it only takes, what, fifteen minutes for a lion to decide to evolve into tiger, right?

  4. unclefrogy says

    if all you had to do to get rid of them was invite them in to bad no one thought of it earlier.
    uncle frogy

  5. says

    I am sure that evolution is true, but I don’t really think I could score well on your quiz. A quick set of google searches leads to the following probably wrong answers.

    1. Looks line the number I keep finding is .003 mutation per generation.
    2. Most are neutral.
    3. Organisms are classified by scientists in different ways. One way is by a description of the DNA that the organism carries. The other is by the traits that the organism has, which would include both its behavior and its physical form. The genotype of an organism is the set to which it belongs when classified by its genes, the phenotype is the set to which it belongs when it is classified by its traits.
    4. As noted in 2 most mutations are neutral, some are positive and some are negative. Deformity is probably not a term used much in evolutionary biology, because it is too vague and ill defined.
    5. I think that “elephant disease” calls out for a clever definition.

    This has been an interesting exercise and I welcome feedback about where my answers went wrong.

  6. Justin Griffith says

    Perhaps she meant ‘elephant flu’? Is that a thing?

    Maybe she is confused about the flu mutations popularized in the news (‘bird’ ‘swine’ etc.) I wonder if she argues with her TV when those are reported on the news.

  7. says

    I see she is back on Twitter now…still waiting though. When is this “debate” going to happen? I haven’t seen a decent creationist around here in a while.

  8. susannebullo says

    Just letting you know (I said as much on Twitter), you might not want to engage @insectidyll. Her name is Dawn Gordon, she’s from Brampton, Ontario and several people on Twitter/Facebook/various social media outlets can tell you she’s a pretty nasty little harassing stalker – @kaimatai can tell you just how obsessed and creepy she is. Just a warning :)

  9. says

    Insectidyll is a freaking idiot. She has replied on twitter; I told her NO, I’m not going to debate her in 140 character chunks.

    Now she wants to know if she can write her answers in Word and email them to me. NO.

    She may be too stupid to appear here.

  10. says

    I did read up on her. Small potatoes.

    Seriously, you have no idea how much hate mail and whiny-ass shit people fling my way. One more causes me no trepidation at all.

  11. Janine: Hallucinating Liar says

  12. wormaldson says

    I’m afraid I’m not so well versed in stupid; would someone mind translating this halfwit-speak into proper English? She isn’t really trying to say that evolutionary biology claims that people with this “elephant disease” are that way because they evolved from elephants is she? What the shit…

  13. Crip Dyke, MQ, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    What is a typical mutation rate? (be specific about the units!)

    A mutation rate most commonly experienced or commonly experienced. We are now not talking about the mean or the median, but the mode – and, perhaps, those mutation rates that have rates of occurence within one standard deviation of the rate of occurrence for the modal mutation rate. Frankly, this is true no matter what armed force one chooses to join (if any) or what battalian/company/troop to which one is assigned.

    What are you talking about PZ?

    Are most mutations deleterious, neutral, or beneficial?

    Since most DNA is non functional and the DNA that is functional has a subset that cannot be mutated and still develop into a living individual of a species, most mutations are neutral. Some, however, have noticed a redistribution of enzymes within certain cells and cell types, not to mention resources like oxygen. A growing number of mutations are choosing to occupy one of the Wall Gyri in protest.

    What is the difference between phenotype and genotype?

    A geno type is your type at the moment god created you. Your pheno type is the type you approach, like Achilles chasing a rabbit, but never quite achieve. It’s a creationist parable for how impossible it is to truly change one’s kind.

    What is the difference between mutation and “deformity”?

    Duh – whether it occurs in me or in you (respectively).

    What is “elephant disease”?

    Symptomatic invasions of pachyderms by parasites, fungi, bacteria, or viruses.

    =========================================================
    I rock at this biology stuff. I should change careers.

  14. kraut says

    “Seriously, you have no idea how much hate mail and whiny-ass shit people fling my way. One more causes me no trepidation at all.”

    There is an old german proverb: Viel feind, viel Ehr’..

  15. Nerd of Redhead, Dances OM Trolls says

    Poor Dawn, if she comes here, she will have to demonstrate with conclusive physical evidence, physical evidence that would pass muster with scientists, magicians, and professional debunkers as being of divine, and not natural, origin.

    Should be interesting seeing the lies and evasions to avoid acknowledging her deity only exists between her delusional ears…

  16. Janine: Hallucinating Liar says

    Nerd, Dawn Gordon will not even try to have conclusive physical evidence. It is beneath her. She gets her truth from GOD and GOD talks to her.

  17. A. R says

    And that PZ is really a christian. And how he is defrauding every one around him.

    [A. R wonders how PZ could be defrauding himself, considering that we’re all sockpuppets of him.]

  18. Nerd of Redhead, Dances OM Trolls says

    She gets her truth from GOD and GOD talks to her.

    I don’t expect her to get it, any more than txpiper does. I do this for the lurkers, to make sure they understand godbots have nothing but their presuppositional delusions, which everybody should ignore and laugh at.

  19. A. R says

    MarkNS: Can we avoid the armchair diagnoses please? No offense, but that’s something that the commentariat here doesn’t really appreciate.

  20. Crip Dyke, MQ, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    @23:
    I’m really not sure about mutation rates – one of the complicating things about such rates is that they are very different for different portions of the genome. We would expect sonic hedgehog to be conserved at a lesser rate compared to Hox genes, but at a greater rate compared to, say, eye color genes. And even eye color genes would likely be conserved at a greater rate than the non-functional pseudogenes contained within the so-called “junk DNA” regions of our genome.

    BUT…
    You do have genotype and phenotype wrong. Genotype doesn’t refer to *genome*-type. Instead it refers to *gene*-type. In other words, it is the particular gene combination that gives rise to a trait. Phenotype is the trait itself. This is important because with dominant/recessive gene pairs (rather than interactive gene pairs) you could have DD, Dr, or rD pairs all leading to the same phenotype. On the other hand, only one pair, rr, leads to the recessive phenotype in question.

  21. lancefinney says

    BTW, I’m pretty sure Insectidyll is male. Here’s the profile:

    Dedicated poetry to my wife and soul mate and my offspring, who helped me love the benefits and beauty of the bug.

  22. susannebullo says

    @48 – Lancefinney – no, she’s not. She created that account (like her 20th or so on Twitter) and was pretending to be someone that was in love with her. She’s given up that pretense now.

  23. says

    I had a run-in with Dawn a long while back. Lasted an incredibly short time because I wasn’t her primary target.

    And while I’m as much against armchair diagnoses as anyone else…

    Come on…

    There’s something going on, here…

    As far as Insectidyll is concerned… if it really is Dawn, why would she say she has a “wife”? I could be remembering wrong, but I thought Dawn was Homophobic?

  24. Crip Dyke, MQ, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    @33lp – #5

    Yes. A tiger in Africa. Well, not precisely a tiger. More a tiger costume. But from Africa! – I got it from a bicycle shop in Cairo.

  25. IslandBrewer says

    Seriously, you have no idea how much hate mail and whiny-ass shit people fling my way. One more causes me no trepidation at all.

    I would like to!

    Sounds like a fun post. On a good hate mail day (or spread out over several days to get an average), come up with a half dozen or so categories of hate mail, and count the number in each (some may fall into more than one category). Give us an example of each one, because, you know, we like that shit.

  26. Thorne says

    PZ said in #16:

    I know. I open the door, I invite them in, they run away.

    Well, try putting some clothes on BEFORE you open the door! You know how sensitive those godbots are.

  27. aspidoscelis says

    @Crip Dyke, #47:

    Regarding mutation rates, when it comes to variation across different loci you’re thinking mostly of the actions of natural selection, rather than variation in the mutation rate per se. Mutations are more or less random; they happen at approximately equal rates across loci. However, they are more likely to be purged by natural selection if they happen to disrupt genes necessary for survival.

    If you look at something like hox genes, the mutation rate is about the same as elsewhere in the genome. The difference is that you don’t see most of those mutations, because they have been purged by selection… i.e., most zygotes carrying mutations in these genes probably die before we have a chance to look at them and don’t leave any descendants.

    That said, it’s not uncommon to see post-selection rates of nucleotide change referred to as “mutation rates”… despite the fact that what we’re actually seeing is just what’s left of the mutation rate after natural selection takes a bite out of it…

  28. Daniel Schealler says

    A mutation rate is the speed of a mutation.

    So if we consider the mutant ‘healing factor’ of the character Wolverine in X-Men, then our units would be ‘time’ and ‘amount regenerated’.

    ‘Amount healed’ is inherently problematic because it isn’t easily quantifiable. However, one technique could be to remove volumes of flesh from various parts of Wolverine’s body, presumably while he is held in an unconscious state.

    The volume, mass and ‘damaged’ surface area of the portion removed would be recorded, and then the vacant portion monitored to see how long it takes for that section of the body to be fully restored.

    Each of these may then be marked against time. Taking samples from various parts of the body and various types of tissue would reveal which among these measurements is the most consistent between tissue types.

    I suppose that mass/second and surface area would be a factor. I propose that the rate of mutation would be proportional to:

    r = km/ta

    r = mutation rate
    k = unitless constant – varies between individual mutant and their powers
    m = mass
    t = time
    a = effective surface area over which the mutation takes place

    Taking this approach, mutation rate could be meaningfully evaluated in units of mass per second per cubic cemtimeter. I name this unit a dan of mutation rate. The symbol for this unit is a suffix, lowercase dn.

  29. Daniel Schealler says

    Note that the dan unit of mutation rate may also use mass/energy equivalence to convert between energy-based mutations and physical ones.

    This would provide an interesting basis for comparing the relative powers between mutants that does not depend on the consequences of those powers.

    I’m not sure how to adjust the dan for characters such as nightcrawler whose powers are more often spatially based. I’ve heard from physicists like Lawrence Krauss that space and energy can be compared, but I’m not sure if that translates into a technique that is consistent with that of a dan.

    Suggestions greatly appreciated.

  30. Holms says

    As far as Insectidyll is concerned… if it really is Dawn, why would she say she has a “wife”?

    Because she is nuts / stupid is my guess.

    Regarding this ‘debate’: we already know this will be an utter fizzle. Regulation issue apologetics will ensue, followed by standard apologetics retreat and much ‘hurt’ in the ‘butt’. She will declare herself the victor because the naughty heathens ‘don’t get it’ and that we are biased against, or hate, Jesus and co.

    So on, so forth, ad nauseum, etc. etc. etc.

  31. says

    Another vote for elephant disease meaning diseases that only elephants get here. Makes for decent “is this an elephant?” criteria I suppose. At the very least you would have to express similar membrane proteins and perhaps have have similar histocompatibility complex genes- I’m not sure how much those tend to impact the diseases of entire genera or family levels of organisms.

  32. Ichthyic says

    Dennis Markuze? Is that you?

    indeed.

    I looked at the list of people harassed, and what they had to say about Dawn Gordon.

    whoever she is, I would bet money she has the same personality disorder that Markuze has.

    the style of writing, the way she thinks god is telling her who is right and wrong…

    so similar to markuze that it simply can’t be coincidence.

    and, like markuze, I hope someone gets her the help she needs sooner rather than later.

  33. says

    Thorne

    PZ … try putting some clothes on BEFORE you open the door! You know how sensitive those godbots are.

    It’s not so much the nudity. It’s the war paint, and cow skull on his head that does it.

  34. Usernames are smart says

    It’s not so much the nudity. It’s the war paint, and cow skull on his head that does it.
    — Hyperdeath #67

    Ahh, you’ve had to deal with a Mormon infestation also?

    After politely, then sternly, then angrily telling them not interested, the above* was the sure-fire cure to ending (for good) their “visits.”

    * Didn’t have time to apply paint or skull :(

  35. says

    I don’t even understand why you’re even considering wasting your time with somebody that is obviously a troll and will only lead to an extremely unproductive debate. It seems a bit sadistic to me.

  36. sonofrojblake says

    I don’t even understand why you’re even considering wasting your time

    Everyone needs a hobby.

  37. says

    @50

    And while I’m as much against armchair diagnoses as anyone else…

    Come on…

    There’s something going on, here…

    I don’t think pharyngulites will ever agree about when it is appropriate to say someone is obviously having a mental health issue. Most seem to think it is never appropriate, even in the face of behaviors that simply do not occur in healthy people. I don’t know how else to characterize her behavior though, especially her emphasis on “imposter” identities and god revealing them to her.

  38. says

    instectidyll just wants free medical advice. She wants to know if she has turned into elephant man and become male, whether that is a mutation or a deformity, and if she is now afraid of mice.

  39. Crip Dyke, MQ, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    @aspidoscelis #56

    If you look at something like hox genes, the mutation rate is about the same as elsewhere in the genome. The difference is that you don’t see most of those mutations, because they have been purged by selection…

    That said, it’s not uncommon to see post-selection rates of nucleotide change referred to as “mutation rates”… [but] we’re actually seeing is just what’s left of the mutation rate after natural selection takes a bite out of it…

    Okay, very fair point. But you can only measure mutation rates in organisms through comparisons to other living organisms. Genes mutated in such a way as to prevent a zygote from becoming an individual of a species won’t show up in the gene pool.

    It may not be correct, but it certainly has reasons for being used… and it’s mainly how I’ve heard it used as a non-biologist (e.g. “DNA clock” studies, etc.).

    Anyway, thanks for correcting me. It does make sense that in an era when we can (now) study DNA in random fertilized eggs that we can study mutation rates much more directly… and thus would want to distinguish mutation rate from, I don’t know, “post-selection mutation rate”.

    What would be the accurate term, anyway?

  40. coyotenose says

    I’m going to tell her on Twitter to come here and address these very simple questions. Anyone want to take bets on whether she bothers?

    After seeing that, I’m not willing to even bet on whether she can read and understand your invitation.

  41. says

    Usernames are smart:

    Didn’t have time to apply paint or skull :(

    Be prepared. If you don’t keep paint and a cow skull near your front door, you can hardly complain if they’re not there when needed.

  42. says

    I couldn’t figure out why she had such a thing for deformities, but after a bit of twitter-digging it became clear; she subscribes to the view that all the remains found of other human species were really H. sapiens all along, just with deformities and diseases.
    See, there never was any Homo erectus or neanderthals or any of those. They were all just deformed human beings. And that’s why evolutionary theory is evil; it claims that deformed people are really animals. Or something.

    It’s sort of interesting to try to uncover how a person gets to such a strange point of view. Like mental archaeology.

  43. Sastra says

    I’m going to tell her on Twitter to come here and address these very simple questions. Anyone want to take bets on whether she bothers? …I know. I open the door, I invite them in, they run away.

    Which is why you should make this your standard response — with a variation taken from Randi’s Million Dollar Challenge. When those who claim paranormal abilities first apply, they do not deal with the Amazing One himself, but are given a preliminary test by one of the affiliated skeptic groups. IF they pass that, then the serious test takes place. So far, nobody has passed the preliminary.

    Comments section is your preliminary. If they get through us, then you step in.

    Of course, sometimes you might want to address something they say, anyway. But I suppose you can always step in when you want. It’s just that you’re clear up front you don’t have to.

    Creationists will likely claim they “defeated” the Hoard. But at least we will teach them caution. Those who venture in, that is. My guess is that your guess that it will filter out most of them upfront, is correct.

  44. susannebullo says

    PZ – she’s attempting to answer on Twitter (not realizing you’ve blocked her) and those answers are exceedingly pathetic. My 13 year old son has a better grasp of science, biology, human DNA, and evolution.

  45. says

    Question #2 is a trick question! There is no one-dimensional scale known, against which we can define at least three unambiguous points: deleterious, neutral, or advantageous. The way I see it, there is only survival or death. For example, a gene that permits/causes sickle-cell anemia seems to be bad- unless you live in anophelese territory! Was that genetic change deleterious, neutral, or advantageous? No way to say.

  46. Daniel Schealler says

    @ivarhusa #81

    I’m no expert and I don’t remember the specifics.

    There are some amino acids that can be produced by different ‘words’ in the genetic code. Some of these words are the same.

    For example, say that amino acid Alpha is produced by both ACT and ACG.

    If the base organism has ACT, but then the daughter mutates at that particular location to become ACG, then nothing in the protien itself will change, and there will be no effect whatsoever on the phenotype of the animal.

    That’s just one horrendously oversimplified example of a neutral mutation. Note that I made up the names and sequences due to poor memory – but I recall from previous reading that this is a scenario that does happen often, and that variation these kinds of ‘free’ markers are often used to evaluate the degree of variation between species.

    Back me up here with specifics, guys.

  47. Daniel Schealler says

    * ‘Some of these words are the same’

    That is a sentence that I butchered with copy/paste and should have deleted before posting. Ooops.

  48. Amphiox says

    For example, say that amino acid Alpha is produced by both ACT and ACG.

    If the base organism has ACT, but then the daughter mutates at that particular location to become ACG, then nothing in the protien itself will change, and there will be no effect whatsoever on the phenotype of the animal.

    That’s just one horrendously oversimplified example of a neutral mutation. Note that I made up the names and sequences due to poor memory – but I recall from previous reading that this is a scenario that does happen often, and that variation these kinds of ‘free’ markers are often used to evaluate the degree of variation between species.

    Back me up here with specifics, guys.

    http://en.wikipedia.org/wiki/Genetic_code

    Check out the diagram at the end.

    Valine, alanine, glycine, serine, arginine, threonine are four-fold degenerate for the third base, ie, the third base could mutate to ANYTHING and the amino acid won’t change.

    And Leucine is not only four fold degenerate for the third base, it is two fold degenerate for the first base, giving it six potential mutations without changing.

  49. Ichthyic says

    Was that genetic change deleterious, neutral, or advantageous? No way to say.

    well, actually you DID say the way to gauge this:

    it’s relative to the selective pressures of any given environment.

    but then, this IS the way mutations are always judged; they are only relevant to the fitness of individuals within a given population, period.

    but then, knowing this would also be part of how PZ would judge the answer to the question.

  50. hamishcook says

    Actually, it appears she’s Canadian but likes to harass people in NZ and email every member of parliament. Wierd.

  51. Amphiox says

    Was that genetic change deleterious, neutral, or advantageous? No way to say.

    Sure there is. If individuals possessing the genetic change produce more offspring on average than individuals not possessing the genetic change, with the overall population subject to the same environment, then the genetic change is advantageous. If they produce less, it is deleterious. If they produce the same amount, it is neutral.

    We even have a formal term that quantifies this for the purposes of mathetmatics: selection (or fitness) coefficient.

    The only restriction is that you can’t tell right away when the genetic change first appears, but only after at least one generation has passed.

    But in real life you virtually never actually identify a genetic change the moment it first appears, and usually you won’t have found it until at least several generations have already passed, so this isn’t a problem.

  52. David Marjanović says

    After politely, then sternly, then angrily telling them not interested

    Did you try saying “why don’t you wait till I die and baptize me afterwards, when you’re going to do that anyway”?

    somebody that is obviously a troll

    That’s a diagnosis that should be made more carefully.

    In my experience, trolls sooner or later brag about being trolls: “I’m just trying to get you upset so I can laugh at you, and look how well it’s working! LOL!”

    Was that genetic change deleterious, neutral, or advantageous? No way to say.

    Sure there is – it just depends on the environment. No mutation is universally deleterious, neutral or advantageous. Sickle-cell anemia is a good example.

  53. vaiyt says

    @amphiox:

    Sure there is. If individuals possessing the genetic change produce more offspring on average than individuals not possessing the genetic change, with the overall population subject to the same environment, then the genetic change is advantageous. If they produce less, it is deleterious. If they produce the same amount, it is neutral.

    We even have a formal term that quantifies this for the purposes of mathetmatics: selection (or fitness) coefficient.

    The only restriction is that you can’t tell right away when the genetic change first appears, but only after at least one generation has passed.

    I think you missed the point a tiny bit. What ivarhusa was trying to convey, I think, is that blanket statements about the fitness value of a mutation can’t be made because they’re dependent on that specific population’s success on that specific environment. Hence the sickle cell example. Another example would be the altered jaws (correct me if I’m mistaken) that made cave fish blind as a side-effect. Deleterious outside of caves, advantageous inside of them.

  54. Jerry says

    There is an absolute mutation rate independent of the effects (deleterious, neutral, beneficial) on the organism or species. This is simply the DNA copy error rate inherent in the DNA replication mechanism within the organism, measured in # miscopied bases per total bases (usually counted as #/hundred-thousand or #/million bases. Viruses have very poor error detection and correction, so they have a high error and mutation rate. Mammals have multiple complicated DNA mismatch detection and repair pathways, so the mutation rate is much lower than the simple error rate in the DNA polymerase which copies the chromosome. I’m several years away from learning and working on this kind of research, but that should be the basic info. Corrections gratefully accepted (be polite, please).

  55. aspidoscelis says

    Re. #74, Crip Dyke:

    Okay, very fair point. But you can only measure mutation rates in organisms through comparisons to other living organisms. Genes mutated in such a way as to prevent a zygote from becoming an individual of a species won’t show up in the gene pool.

    Generally, what you do is measure the mutation rate in regions that are under very little or no selection pressure (e.g., non-coding regions). Deviations from that rate in coding regions can be attributed to selection.

    There are some further complications, mostly beyond me…

    It may not be correct, but it certainly has reasons for being used… and it’s mainly how I’ve heard it used as a non-biologist (e.g. “DNA clock” studies, etc.).

    Anyway, thanks for correcting me. It does make sense that in an era when we can (now) study DNA in random fertilized eggs that we can study mutation rates much more directly… and thus would want to distinguish mutation rate from, I don’t know, “post-selection mutation rate”.

    What would be the accurate term, anyway?

    Yeah, for most studies, you can’t separate mutation rates from purging by selection, so you’ve got just the one number. “Observed mutation rate” should work, but it’s important to keep in mind that some or most mutations may not be observed!

    (This is often a problem in phylogenetics, which is mostly what I know. One of the old bits of wisdom in the field is, “Well, if you’re studying deeper relationships, you pick a more slowly-evolving locus.” Uh… no… that “slowly-evolving” locus is just one in which more of the potentially informative mutations have been selectively purged. Not good. You want a locus under as little selective pressure as possible. “Slowly-evolving” probably means “biased”. This source of possible error may be compounded in model-based phylogenetic methods–the model needs an estimate of the mutation rate, but this is not a trivial problem! Give any model-based method a coding locus, and its model is likely to be severely out of whack.)

  56. aspidoscelis says

    Oh, bugger, I screwed up the blockquotes. Let’s try that again.

    Re. #74, Crip Dyke:

    Okay, very fair point. But you can only measure mutation rates in organisms through comparisons to other living organisms. Genes mutated in such a way as to prevent a zygote from becoming an individual of a species won’t show up in the gene pool.

    Generally, what you do is measure the mutation rate in regions that are under very little or no selection pressure (e.g., non-coding regions). Deviations from that rate in coding regions can be attributed to selection.

    There are some further complications, mostly beyond me…

    It may not be correct, but it certainly has reasons for being used… and it’s mainly how I’ve heard it used as a non-biologist (e.g. “DNA clock” studies, etc.).

    Anyway, thanks for correcting me. It does make sense that in an era when we can (now) study DNA in random fertilized eggs that we can study mutation rates much more directly… and thus would want to distinguish mutation rate from, I don’t know, “post-selection mutation rate”.

    What would be the accurate term, anyway?

    Yeah, for most studies, you can’t separate mutation rates from purging by selection, so you’ve got just the one number. “Observed mutation rate” should work, but it’s important to keep in mind that some or most mutations may not be observed!

    (This is often a problem in phylogenetics, which is mostly what I know. One of the old bits of wisdom in the field is, “Well, if you’re studying deeper relationships, you pick a more slowly-evolving locus.” Uh… no… that “slowly-evolving” locus is just one in which more of the potentially informative mutations have been selectively purged. Not good. You want a locus under as little selective pressure as possible. “Slowly-evolving” probably means “biased”. This source of possible error may be compounded in model-based phylogenetic methods–the model needs an estimate of the mutation rate, but this is not a trivial problem! Give any model-based method a coding locus, and its model is likely to be severely out of whack.)

  57. David Marjanović says

    Uh… no… that “slowly-evolving” locus is just one in which more of the potentially informative mutations have been selectively purged. Not good. You want a locus under as little selective pressure as possible.

    But those under too little selective pressure have already mutated beyond recognition; saturation, too, is a real problem in molecular phylogenetics. Loci under strong selection will still accumulate neutral mutations that are phylogenetically informative.

    Anyway, what’s usually a good idea is a total-evidence approach: take everything and throw it into the same matrix. The signal will add up, the noise will cancel itself out.