subscribe to the RSS Feed

Thursday, September 18, 2014

Now This is Interesting…

Posted by jlwile on May 5, 2010

Answers research journal has just published a very interesting study called Baraminological Analysis Places Homo habilis, Homo rudolfensis, and Australopithecus sediba in the Human Holobaramin. If you are not familiar with the term “holobaramin,” it refers to a group of animals that are all related to one another through common descent.

Remember, in the creationist view, God created individual KINDS of organisms, and the genomes of those organisms were created so that they could adjust and adapt to changing conditions. As a result, the organisms that God created could change significantly, but not infinitely. Their amount of change is bound by the level of information in their genomes.

Wolves and dogs, for example, are a part of the same holobaramin. Even though a Chihuahua and a timber wolf might look and behave very differently, they both descended from the same kind of animal created by God. So when a creationist says that certain creatures belong to the same holobaramin, he or she means they both descended from the same created kind of creature.

The analysis presented in the peer-reviewed paper linked above says that Australopithecus sediba should be placed in the human holobaramin, which means it descended from people. If you recall, this is quite different from my analysis of A. sediba.

While the paper does do a solid baraminological analysis, I am not sure I agree with that conclusion. After all, this paper focused only on craniodental features. While I don’t have a problem with that in general, I do have a problem with including Australopithecus sediba in such an analysis. Remember that the majority of the cranial and dental bones recovered were from a juvenile, and it is difficult to compare juveniles with adults. Thus, I am afraid that making a definitive placement for Australopithecus sediba is a bit premature. Hopefully, more fossils will eventually be published so that a more thorough baraminological can take place.

The other conclusion of the paper was quite interesting:

Results indicate that hominins may be divided into as many as four different holobaramins: (1) the genus Homo (including Australopithecus sediba), (2) the genus Paranthropus, (3) Australopithecus africanus, and (4) Gorilla, Pan, Australopithecus afarensis, and Australopithecus garhi.

So Australopithecus africanus is not related to and A. afarensis or A. garhi. That surprises me. However, based on the craniodental data, it seems to be the correct conclusion.

Comments

12 Responses to “Now This is Interesting…”
  1. “Despite this widespread agreement that evolution cannot account for human origins, the question remains: What precisely is human (that is, descendants of Adam and Eve)?”

    By widespread agreement, the author means Adam Sedgwick (1860), Entomologist Erich Wasmann (1909), and William Jennings Bryan (1922). Wow, that’s impressive agreement.

    Going through this pile of baraminology is beyond my ability to hold my nose while reading, but I am interested in the analysis used. That is baraminic distance correlation (BDC).

    I found this from 2002:

    Space prohibits a detailed explanation of the baraminic distance method, but a short description of the metric is in order. The baraminic distance between two species is the percentage of characters in which the two species differ in their character states. The simplicity of this metric is very important, because most evolutionary phylogenetic methods make assumptions of common ancestry to calculate similarities and distances. With a percentage, no prior assumptions are made, so identifying both significant similarity between species (implying baraminic relationship) and significant differences between other species (implying discontinuity) should be straightforward. For a detailed discussion of the baraminic distance method, consult Robinson and Cavanaugh’s original paper.6

    That paper is from CRSQ in 1998.

    And this from CRSQ 2006, The Current Status of Baraminology

    The collaboration of BSG members Robinson and Cavanaugh in the late 1990s introduced new statistical methods to baraminology. Adapting phenetic concepts of Sokal and Sneath (1963), Robinson and Cavanaugh (1998a) defined the baraminic distance as a percentage of characteristics that differ between two taxa, while ignoring unknown characteristics. The baraminic distance is a modified simple matching coefficient (Cox and Cox, 1994). Because the characters used to calculate baraminic distance depend on the selectivity of a researcher, and because different created kinds seem to vary to different extents, the raw baraminic distance is not a measure that can be used directly to infer baraminic membership. Recognizing this problem, Robinson and Cavanaugh (1998a) proposed a correlation test to measure the relative similarities and differences between taxa.

    Cavanaugh also introduced a multivariate method called Analysis of Patterns (ANOPA) at the first BSG meeting in 1997. ANOPA treats each characteristic as a separate “dimension” in multidimensional character space and reduces the dimensionality to three dimensions. The taxa then appear as points in three-dimensional space and can be displayed using any 3D viewing software. The advantage of ANOPA is that it is not distance-based, and thus provides a method of evaluating a set of characters that is independent of baraminic distance correlation.

    More recently, Wood (2005b) introduced a standard multivariate technique called multidimensional scaling (MDS; see Cox and Cox, 1994) to baraminology. With MDS, a researcher can convert a matrix of baraminic distances into k-dimensional coordinates, where k is determined by the researcher. Typically, k=3 is used to maximize the information displayed and still be perceptible to human observers. Because MDS is distance-based and ANOPA is not, one method can serve as a test of the results of the other. When Wood (2005b) applied MDS to the baraminic distance matrix calculated from the Equidae dataset used by Cavanaugh et al. (2003), the 3D coordinates of ANOPA and MDS were strikingly similar, showing the same linear ordering of the taxa with both techniques.

    That’s lots of numbers and statistics! It must be relevant and rigorous!

    PS found an article about that “serpant” in Genesis 3:15 – spoiler alert – it’s a snake.

  2. Wow, I love me the intertubes! How’s this for an unbelievable coincidence, Panda’s Thumb posted on baraminology yesterday! That’s so unlikely, it must not have happened!

    And a quick link to PZ on Wood.

  3. Okay, I did it again, I linked to an article before I read it. Shame on me. But I wasn’t disappointed:

    “** “Barminology” is the creationist study of “bara mins” (“created kinds” in Hebrew). The key metric, “baraminic distance” basically seems to be a phenetic clustering method, where groupings are done by eye and outgroups are excluded if they make the ingroups seem too similar.”

  4. Now this is REALLY interesting:

    Skull Caps and Genomes, Carl Zimmer.
    Neandertals Live!, John Hawks.

  5. jlwile says:

    Shooter, the baraminic distance correlation (BDC) approach is quite relevant and rigorous. If you read the original paper (as your own quote indicates you should) you will find just how rigorous and relevant it is:

    Robinson, D.A. and D.P. Cavanaugh, “A quantitative approach to baraminology with examples from the catarrhine primates,” Creation Research Society Quarterly 34:196-208, 1998

    and, of course the full paper of the added bootstrap method is available on the internet here.

    Imagine that. You didn’t read anything about the process, and yet you still try to ridicule it. Once again, it is great to read your comments, as they show so clearly how evidence means nothing to you!

    Also, you seem to be missing some people in your “widespread agreement” analysis. People like Dr. Cornelius Hunter (Biophysics and Computational Biology), Dr. André Eggen (genetics), Dr. Maciej Giertych (genetics), Dr. Georgia Purdom (molecular genetics), Dr. J. C. Sanford (genetics), Dr. Arlton C. Murray (paleontology), Dr. John Whitmore (paleontology), just to name a tiny fraction of the modern scientists who all agree on that point. Yes, you are right. That is “impressive agreement.” Thanks for pointing out how impressive it really is.

    Your PZ link is yet another example of how you use character assassination instead of citing evidence. Nice job showing that Ben is exactly right about you.

    And, of course, your Panda’s Thumb link is completely wrong. If you would have bothered to educate yourself on what baraminic distance correlation is, you would see it is completely mathematical. It is not done “by eye.” There is also no way to exclude things a priori, as it is all based on the math. Indeed, had you even bothered to read the paper I linked in this article, you would have found that out. Once again, then, you try to ridicule something you know nothing about. It is so fun to read you make such a fool of yourself!

    Then, in an attempt to outdo yourself, you add another comment with two unrelated links. It is amazing how you fall into the same pattern over and over again:

    1. Try to make fun of something about which you are ignorant.

    2. Assassinate the character of people who disagree with you.

    3. Distract from (1) and (2) by adding unrelated links.

    Thank you SO MUCH for your comments. They are an excellent illustration of how irrational people who genuflect at the altar of evolution can be!

    By the way, while your John Hawks link was irrelevant to the discussion, it did serve to illustrate how once again the evolutionists were wrong and the creationists were right. Evolutionists have long tried to portray Neanderthal Man as sub-human, but from the beginning, creationists always said he was fully human. The fact that evolutionists now claim Neanderthals and humans interbred shows they have finally caught up with the creationists!

  6. My last two links were to show what a truly interesting scientific publication is. How can regression analysis of selected studies of selected characters of selected skull fragments compare to the publishing of the Neandertal genome? And that 1 to 4 % of the current human genome is from Neandertals? It’s fascinating.

    As for your last comment, Neandertals have always been classified as Homo. In fact some claim they are a sub-species of Homo sapiens. As this post shows, creationists don’t have an answer as to what “fully human” means. So how could they be right?

  7. jlwile says:

    Shooter, your last two comments were given to distract from the fact that you didn’t even understand the paper this post discusses, but you nevertheless tried to make fun of it. The fact that you couldn’t understand the paper doesn’t mean it’s not “a truly interesting scientific publication.” It just continues to display your ignorance of basic science. Indeed, to those who understand this paper, it is SIGNIFICANTLY more interesting than the links you gave.

    I agree that Neanderthals have always been classified as being in genus Homo, but they were clearly cast as sub-human by evolutionists up until just recently. The first reconstruction made him look very apelike. In Linguistic Inquiry, In 1971, Lieberman and Crelin argued that they couldn’t speak like human beings. In 1998, Science News ran a story about recent data that were uncovering all the similarities between Neanderthal man and humans, and it said these data were leading some to believe that Neanderthal man was just a subspecies of human beings. However, they note:

    Many, however, see them as a separate species and doubt that they shared H. sapiens’ newly sophisticated behaviors. Although Neandertals’ brains were roughly the same size as those of modern people, they often have been portrayed as lacking the language skills, foresight, creativity, and other cognitive abilities of modern humans.

    (Brainard, J., “Giving Neandertals their due—similarities with modern humans shift the image of the caveman brute,” Science News 154:72–74, 1998)

    It is nice to know that such nonsense has been eliminated. Too bad the evolutionists took about 140 years to catch up with the creationists, who never were taken in by any such nonsense.

  8. I understood it well enough to know it’s nonsense. For instance, on the MDS figures, why is there no units on the graphs? Why is there no table with all of the MDS data points? What is the mathematical formula for the ovals drawn on some MDS figures? What is the point of producing a 3D data set in 2D when how far away two points are is the relevant “measurement”? What is stress? What is bootstrap?

    This line is great: “Typically, k=3 is used to maximize the information displayed and still be perceptible to human observers.” Well, since “stress” goes down will more dimensions added, it’s more likely k=3 is used to get the best results without calling too much attention to the silliness of arbitrary dimensions.

    Those are just for starters. If you can answer those, I’ll have more.

  9. In case you didn’t notice, your first link and my first link are the same paper, Wood 2008 on bootstrapping. . Thank you for reminding me to quote it’s first abstract sentence here:

    “Although baraminic distance correlation (BDC) has been used to define more holobaramins than any other method, the technique remains largely untested.”

    This is from 2 years ago, 20 years since baraminology was proposed and 10 years after the “definitive” 1998 CSRQ paper (only the abstract is online). That’s pretty long for the most used baraminic method to remain untested.

    2002 paper link.

  10. jlwile says:

    If you think its nonsense, then you clearly didn’t understand it. Of course, your questions make it VERY clear that you didn’t understand it.

    “On the MDS figures, why is there no units on the graphs?”
    I assume you mean why ARE there no units on the graphs. If you would bother to learn the method (Wood, T. C. “Baraminic distance, bootstraps, and BDISTMDS,” Occasional Papers of the BSG 12:1–17, 2008) you would learn that this is because the numbers that are used by BDISTMDS are DIMENSIONLESS. Remember, this is a SCALING program. It SCALES data.

    “Why is there no table with all of the MDS data points?”
    Once again, the “S” stands for SCALING. Thus, this allows you to “blow up” or “shrink down” the relative distances so as to make a readable graph. Thus, the numbers themselves (being dimensionless) are not important. Only the grouping of data points is important.

    “What is the mathematical formula for the ovals drawn on some MDS figures?”
    There is none. Like the H-R diagram, the idea is to look at the GROUPINGS.

    “What is the point of producing a 3D data set in 2D when how far away two points are is the relevant “measurement”?”
    This, more than anything else, shows that you can’t grasp the paper. There are multiple dimensions in the analysis. That’s what “k” stands for. The the graphs ARE 3D (note the 3D axes) because it is very hard to draw graphs of higher dimensionality.

    “What is stress?”
    Once again, you simply need to take the time to understand what you are talking about. I know you don’t like to do that, as it interferes with your preconceived notions. Stress is a measure of the “goodness of fit” between the scaled data and the baraminic distances. BDISTMDS calculates stress at multiple dimensions and produces a stress graph that displays how well the scaled data is being represented for each number of dimensions.

    “What is bootstrap?”
    Once again, just take the time to learn the method. It would make you look a LOT less ignorant. Bootstrapping is a method by which the robustness of the distance calculation is tested. In bootstrapping, pseudoreplicates of an original dataset are created by selecting with replacement random characters from the original dataset. Trees inferred from 100-1000 pseudoreplicates are then examined for the occurrence of groups of interest. Groups that occur often in the pseudoreplicate trees are considered robust, as they are not sensitive to individual characteristics.

    Please note that bootstrap is a common term in phylogenetic analysis. Thus, this question demonstrates your ignorance of evolutionary research in general. Oh…I forgot – the research is irrelevant to you. Only the dogma from the high priests matters to you.

    “Well, since ‘stress’ goes down will more dimensions added, it’s more likely k=3 is used to get the best results without calling too much attention to the silliness of arbitrary dimensions.”
    The value of “k” is the number of dimensions. Since 3D is generally the limit of human perception, that is what is graphed. Of course, as the paper says, most of the graphs show a very small stress at k=3, so it is reasonable to use 3 dimensions.

    Your misunderstanding of the quote, “Although baraminic distance correlation (BDC) has been used to define more holobaramins than any other method, the technique remains largely untested” also shows you have no understanding of what you are reading. By “tested,” it means there was no bootstrapping. The paper discusses how bootstrapping has been added so that each BDC can be thoroughly tested. This, of course, is what the originally-linked paper (which you clearly can’t understand) did.

    It’s too bad you won’t bother to learn about science before you try to comment on it. Nevertheless, I will educate you where I can, whether you like it or not.

  11. jlwile says:

    I am not surprised you want to end this conversation. You have demonstrated that you are not only unable to understand the article, but that you don’t even understand the basics of phylogenetic analysis.

    You really are a delight! It is rare an evolutionist is willing to demonstrate so effectively and so often how ignorant most evolutionists are about science in general and evolution in particular.

home | top