Today is the 5th Anniversary of Kitzmiller Trial

On December 20th, 2005, Judge John E. Jones III, ruled that it was unconstitutional to read a statement that discredits evolutionary theory and barred the teaching of intelligent design there.  This was the first, but probably not last, trial that involved the teaching of intelligent design as an alternative to evolution.

Its hard to believe that it has only been 5 years since the details and motives of the Intelligent Design (ID) movement were brought out to the public in a public trial.  We learned from staunch ID supporter Michael Behe that there are no peer reviewed published articles supporting ID.  We learned how the Discovery Institute’s”wedge document” essentially points to the idea that ID was brought up as a way to replace the scientific method with “a science consonant with Christian and theistic convictions”  We also saw how the book in question, Of Pandas and People,  had earlier versions where they simply replaced the word “creation” with “intelligent design.”  Of course there was more damming evidence against the ID proponents in the trial, but these three facts alone provide sufficient evidence that intelligent design is not science and should never be taught in the classroom.

Intelligent Design does not make predictions and is not science.

Since it has been so long since I last wrote a blog post, I thought that I ease back into blogging by attacking the lowest of the low hanging fruit from the Discovery Institute. Namely, the argument that Intelligent Design (ID) is real science.

Recently, Casey Luskin wrote a post discussing how ID proponents test their theory in real world situations.  Luskin provides a short list of four items (is that the most he could come up with?) that are supposed predictions of ID.  Lets take them one at a time:

(1) Natural structures will be found that contain many parts arranged in intricate patterns that perform a specific function (e.g. complex and specified information).

This is not a prediction. Life has already been seen to have “intricate patterns that perform specific functions.”  In fact, isn’t the complexity of life what made people believe ID in the first place?

(2) Forms containing large amounts of novel information will appear in the fossil record suddenly and without similar precursors.

It is rare to find any facet of an organism that doesn’t already have a precursor, let alone “large amounts of novel information.”  The prediction of ID should state that there will only be the instant appearance of new structures. Any evidence for a slow, gradual development of a structure would refute ID and prove evolution.  For example, you might see a fully formed tetrapod without any precursors if ID was true.  However, we see in the fossil record myriad examples of transitional forms from fish to tetrapod (Tiktaalik, Panderichthys, etc.).

(3) Convergence will occur routinely. That is, genes and other functional parts will be re-used in different and unrelated organisms.

This prediction seems odd to me. Convergence is a prediction of evolution. Take for example the convergence of the ability to fly. Evolution would predict that different organisms would gain the ability to fly, but they would achieve this ability by slightly different methods.  Luskin is predicting that structures would be re-used by different organisms.  Why do all wings look so different? Does a fly’s wing resemble a bird’s or a bat’s? Not at all.

(4) Much so-called “junk DNA” will turn out to perform valuable functions.

4) “Much?” Luskin should have said “all.” if we are to believe in intelligent design.  Why waste even a single nucleotide if you were to design an organism’s DNA? This sort of ambiguity is what makes these predictions worthless.

I think it is clear to see that the predictions put forth by Luskin are basically worthless for providing support that ID is real science.  What do you expect when you can give the “designer” any attribute, power, or foresight that you desire?

Did Darwin make faulty predictions on evolution?

David Klinghoffer has been interviewing Intelligent Design (ID) proponent and Discovery Institute fellow, Cornelius Hunter. They discuss Hunter’s new website, Darwin’s Predictions. The idea of the website is to illustrate how some of Darwin’s predictions have turned up false.

Before we get into the interview, let’s review a few of Darwin’s predictions and see how they fared:

As can be clearly seen by these overarching predictions, Darwin got a lot right.  In fact, he got enough right that any subsequent theory needs to take into account his basic premises.  Unfortunately, many ID proponents choose to ignore these proven ideas in favor of the smaller, more complex details that have yet to be proven.  Details which have not been worked out yet or might not ever be due to the constantly changing earth.  That is exactly what Hunter is doing with his website.  He is actually exploiting the process of science, which involves bad predictions and failed ideas, to illustrate his point.  Therefore, not much stock should be put into these so-called failed predictions.

So, Hunter’s premise is flawed, but let’s go ahead and get into a few examples of what Hunter is calling Darwin’s failed predictions.  One example is where Hunter describes the similarity of the squid eye and the mammalian eye as being a failed Darwin prediction:

dramatic similarities are sometimes found in otherwise distant species. The eye of the squid and the human, for example, are incredibly similar. Such design convergence is rampant in biology, in spite of the evolutionary expectation.

In such a situation, evolutionary theory predicts that they have common ancestor with genes necessary for basic eye development and the final shape will be a result of subsequent modifications.  This doesn’t seem like a failed prediction to me.

Anyway, are the eyes of squid and mammals really that similar?  So similar that they preclude an evolutionary origin?  On the surface, they do seem very similar, but when you delve deeper, the differences become clear and obvious. The most obvious is that the mammalian eyes have the light sensing layers of the retinal inverted compared to the squid eye.  Furthermore, the two eyes develop completely differently with the squid eye arising from a series of invaginations, while the mammalian eye forms from cell signaling.  Besides, how else are you going to make an effective eye?

Another example that Hunter gives of Darwin’s failed prediction comes from the relatedness of very conserved genes:

the finding of long stretches of identical DNA in distant species is a good one. Evolutionists have worked hard to figure out how this could be

Did I miss something? When did conserved stretches of DNA falsify evolution? Of course Hunter is really talking about how they are too conserved to be explained by evolutionary theory.  Since this is the exception rather than the rule, I don’t see how this is a failed prediction.

His next example:

Then there is the evolution of contradictory behavior patterns, such as altruism. Evolution has undergone a big makeover in the past fifty years in trying to explain such behaviors.

Altruism is pretty straight forward.  A social group of organisms helping each other out will survive longer than those that don’t.  Plus, there is also the punishment of those that don’t play by the societies rules.  You can even see altruism in one of the simplest of organisms, Dictyostelium. Some of these amoeba actually kill themselves so that others will be able to live.

These examples that Hunter provided are not very convincing.  None of them really falsify evolutionary theory.  They do show how some ideas of evolution were wrong, but they don’t come anywhere near falsifying the theory.

Since I have been on a roll of calling out the Discovery Institute for its hypocrisy, I might as well end up with another example. At one point, Klinghoffer says:

Darwinists are compelled to mold their interpretations of data to match the preconceived theory.

This comes from the same group that says that once wrote:

Design theory promises to reverse the stifling dominance of the materialist worldview, and to replace it with a science consonant with Christian and theistic convictions.

Besides, not all evolutionists are atheists. Far from it.  Ever heard of Ken Miller?

Luskin can’t see the reality of evolution for the trees (part 3)

Part 3 – Analysis and rebuttal (cont)

This post is the third part of a three-post series aimed at clearing up the misinformation written by the Discovery Institute’s Casey Luskin in regard to phylogenetic trees and  the idea of a complete tree of life (TOL).  In parts 3, 4, and 5 of Luskin’s series, he continues his bastardization of the evidence for evolution.

In part 3, Luskin discusses what he calls “extreme genetic convergence.”  The problem is that he confuses genetic convergence with heredity.

One data-point that might suggest common design rather than common descent is the gene “pax-6.” Pax-6 is one of those pesky instances where extreme genetic similarity popped up in a place totally unexpected and unpredicted by evolutionary biology. In short, scientists have discovered that organisms as diverse as jellyfish, arthropods, mollusks, and vertebrates all use pax-6 to control development of their very distinct types of eyes.

Having the same gene controlling eye development in different organisms is completely compatible with evolutionary theory.  This gene (pax-6) could have controlled the development of a very simple eye, perhaps a patch of photosensitive cells, in a common ancestor.  Subsequent organisms would use the foundation laid by pax-6 and add their own specific modifications to yield different eye types.  Luskin’s assertion here that pax-6 argues against evolution makes no sense, except for someone who is actively looking to twist data to their preconceived notion.

Luskin is also outright wrong when he says that pax-6 is used to “control development of their very distinct types of eyes.”  Pax-6 is necessary for eye development, but it does not influence the type of eye made.  For example, if you take the mouse pax-6 gene and put it into a fruit fly, the fruit fly makes fly eyes, not mouse eyes.  It is clear Luskin is either confused or is misrepresenting the facts.

Homology is evidence against evolution?

In part 4, he argues that homology between animals, both at the molecular level and at the physiological level, is a problem for evolution.  Luskin doesn’t really do any of his own work, but instead quotes from the Explore Evolution “textbook”:

To summarize, biologists have made two discoveries that challenge the argument from anatomical homology. The first is that the development of homologous structures can be governed by different genes and can follow different developmental pathways. The second discovery, conversely, is that sometimes the same gene plays a role in producing different adult structures. Both of these discoveries seem to contradict neo-Darwinian expectations

Neither discovery contradicts the theory of evolution.  It doesn’t matter the path that a gene or structure takes to be effective, it just matters that it is effective.  Explore Evolution is trying to take an interesting facet of biology and say it disproves evolution without really showing how common descent precludes these features.

Let’s turn the tables and ask what do these two discoveries mean for intelligent design? Well it means that here is another example of stupid design. I say stupid because what designer would use “different genes” and “different pathways” to come up with the same structure. That would be a monumental waste of time and effort for the designer. How about using the same gene for different functions? Well, that is better design, but it goes completely against the first point.

Morphological vs. phlyogenetic trees

In part 5 of Luskin’s series of posts, he claims that morphological data does not correlate with phylogenetic trees.   Maybe they don’t fit exactly, but the similarities are so common that it is ridiculous to think they are not due to common ancestry.  Like I mentioned in part 1, there is a lot of problems associated with the creation of mathematical models used to predict phylogenetic trees. Likewise, trees based on morphology are subject to their own problems.

To make his point, Luskin actually refers to the gene (cytochrome B) that I had picked in part 1 of my series of posts.  Using the sequence of this gene from different ape species,  I was able to produce the exact same phylogenetic tree as had been done using endogenous retroviruses.   What does Luskin say about cytochrome B?

pro-evolution textbooks often tout the Cytochrome C phylogenetic tree as allegedly matching and confirming the traditional phylogeny of many animal groups. This is said to bolster the case for common descent. However, evolutionists cherry pick this example and rarely talk about the Cytochrome B tree, which has striking differences from the classical animal phylogeny.

I didn’t look throughout all of “classical animal phylogeny,” but I was able to create evidence for common ancestry using cytochrome B that matched both morphological and molecular evidence.  Without common ancestry, this should not have been possible.

One final point. Whenever someone looks up scientific articles, it is best to look at the newest articles for obvious reasons. Yet Luskin did the exact opposite.  To help make his point here, Luskin quotes from several scientific papers that were published around the turn of the century.  One is even from 1993.  These papers came before the genomics era and before automated sequencing was common.  They do not really belong in this discussion.

Conclusion

Through his series of posts, Casey Luskin tries to portray the state of phylogenetic analysis as being counter to the theory of evolution.  I hope that I have showed that the opposite is true.  While we don’t have and may never have a complete tree of life, the data that we obtain creating trees or bushes is points squarely to common descent.

Luskin can’t see the reality of evolution for the trees (part 2)

This post is the second part of a three-post series aimed at clearing up the misinformation written by the Discovery Institute’s Casey Luskin.  In his recent posts, Luskin tries to persuade his readers that the idea of a tree of life (TOL) and the very idea of phylogenetic trees is erroneous and not evidence of common descent.  These trees are created by looking at genetic similarities between organisms to arrange them in terms of relatedness and common ancestry.  In my series of posts, I will expose the weaknesses in the arguments put forth by Luskin.

Part 2 – Analysis and rebuttal

In this second part of my three-part series on the realities of the TOL, I will provide rebuttals to Luskin’s points.  These points were written in Luskin’s part 1 and part 2.

One of Luskin’s points in his post is to question the motives and biases of scientists.  Here, Casey Luskin claims that scientists assume there is a tree of life so their findings will support their preconceived notions:

the first assumption that goes into tree-building is the basic assumption that similarity between different organisms is the result of inheritance from a common ancestor

Of course this is a ridiculous proposition.  I guess Luskin has completely forgotten all about Charles Darwin and all the study into evolution since that time.  Prior to Darwin, common ancestry was not an idea that had any credence.  Sine the time of Darwin, more and more evidence keeps adding to Darwin’s idea basic ideas of common descent.  Basing ideas on evidence is not the same thing as assuming.

Luskin also contents that scientists engage in ad hoc reasoning:

whenever data contradicts expectations of common descent, evolutionists resort to a variety of different ad hoc rationalizations to save common descent from being falsified

No. What scientists do is to take this new information and form new hypothesis and alter the details of evolution. Science is always changing.  Finding unexpected things is what makes science interesting and nothing is gained in science by keeping ideas that have been proven wrong.

As far as saving “common descent from being falsified,” evolution is easily falsifiable.  Find a rabbit in the precambrian and all of evolution will fall apart.  Find genes in humans that more resemble cockroach genes than any mammal.  However, one result like this would need to be critically analyzed to go against years of research and thousands of experiments.

I find it hilarious that he uses the idea of “ad hoc reasoning” to criticize evolution. The whole idea of intelligent design (ID) is ad hoc reasoning. Any result or any piece of data can be simply said to have been designed that way. There are no predictions or testable hypothesis in ID.

In his second post, Luskin draws heavily on the false idea that scientists are abandoning the tree of life.  A lot of his all comes from the dreaded New Scientist article, “Darwin was wrong.”  I am not going to go into the details as many others have shown that the article was inaccurate to say the least here, here, here , and here.

In addition to heavily quoting the New Scientist article, Luskin “quote mines” from several different scientific papers.  One of the more egregious examples comes from a 2005 science paper by Rokas et al.  Luskin says:

Other scientists agree with the conclusions of the New Scientist article. Looking higher up the tree, a recent study published in Science tried to construct a phylogeny of animal relationships but concluded that “[d]espite the amount of data and breadth of taxa analyzed, relationships among most [animal] phyla remained unresolved.”

Luskin neglects to mention that the next couple sentences:

In contrast, the same genes robustly resolved phylogenetic relationships within a major clade of Fungi of approximately the same age as the Metazoa. The differences in resolution within the two kingdoms suggest that the early history of metazoans was a radiation compressed in time, a finding that is in agreement with paleontological inferences.

Luskin fails to mention a few critical points in the article.  He ignores the fact that a well constructed tree based on Fungi can be made.  Also missing is the fact that the authors came up with a hypothesis to explain the previous data. Finally, Luskin fails to mention that the authors provide for a better way to look create phylogenetic trees when problems arise, rare genomic changes.

Luskin continues the quote mining throughout the post, but he never really says anything favors an intelligent design perspective.  He is just using the tried and true method of ID proponents, namely to find the currently unresolved issues in the scientific literature and omit the overwhelming number of successful findings.

Luskin can’t see the reality of evolution for the trees (part I)

Update: I just realized that wordpress truncated my DNA sequences at the bottom of this post.  If you would like to get the full sequences, you can get them by “view source” in your browser.  If that doesn’t work, let me know.

This post is the first part of a three-post series aimed at clearing up the misinformation written by the Discovery Institute’s Casey Luskin.  In his recent posts, he tries to persuade his readers that the idea of a tree of life (TOL) and the very idea of phylogenetic trees is erroneous and not evidence of common descent.  These trees are created by looking at genetic similarities between organisms to arrange them in terms of relatedness and common ancestry.  In my series of posts, I will expose the weaknesses in the arguments put forth by Luskin.

Part 1 – Why we know the tree is real

In the first part of my three-part series on the realities of the TOL, I will cover some of the reasons why we know it is real and why there are difficulties in creating a single tree that encompasses all organisms.  This post doesn’t directly address Luskin’s concerns, but is more of an overview.  Point by point analysis will come later.

First, I should point out that when I write “tree,” I am not referring to a perfect tree with a single trunk at the base and straightforward branching all the way to the top. I am referring to a tree that does have plenty of instances with a typical branching pattern, but also one that has tangled, intertwining branches.  This is a widely held view among scientists in the field and was written about nearly 10 years ago in Scientific American by Dr. Ford Doolittle.

Without further ado, lets look at why the idea of a phylogenetic tree of life is valid:

Trees can be created that match observations –  This simple fact alone should be enough to show the validity of phylogenetic  trees.  If common descent wasn’t true, then any phylogenetic tree should be meaningless, but they generally follow the organization of standard taxonomy that we have known for years.  Furthermore, these trees are very similar to each other when different genes or combination of genes are used.

Other new lines of evidence correlate with trees base on gene similarities – One good way to see if a scientific idea is valid is to see how well it correlates with other data.  The phylogenetic TOL correlates with taxonomic observations, but what about other lines of evidence for common descent?  For this post, I decided to perform a little experiment.  It has previously been shown (Lebedev et al. 2000)  that the introduction of endogenous retroviruses (ERVs) into genomes can be used to created phylogenetic trees.  For a discussion of this topic, see here. This technique was used in primates to create the tree shown below on the left.  I decided to see if a tree created by using the similarities in the gene for cytochrome B would yield the same tree (Details on exactly how I did this are at the bottom of this post).  As shown below, it did match exactly.

Comparison of trees made by ERV or sequence comparison

Comparison of trees made by ERV or sequence comparison

Two different techniques yielded the exact same tree.  Both trees created here match to a tree that is created by looking at physical similarities and differences.  This one example should be enough to show the validity of phylogenetic trees, but you know nothing would be sufficient to convince the Discovery Institute.

So I have shown a couple reasons why I think phylogenetic trees are valid.  Keep these in mind when you read about the supposed failures of phylogenetic trees from the ID proponents.  Next, I am going to describes some of the difficulties in creating a complete TOL.  Remember, these issues do not mean that common descent is wrong, it just means that our ability to decipher the tree of life is limited.  A weakness in data is not a weakness in a theory.

Difficulties in creating a tree of life

There are many problems with creating a complete tree of life that takes into account every piece of data available.  Think about all the information and computation required for the creation of an all-encompassing  phylogenetic tree. In addition, a TOL has to determine the relatedness of organisms that might not have a common ancestor for over half a billion years.  Below is an overview of some other problems in creating such a tree.

Lack of information – As much as we know about the world and its vast number of organisms, we don’t know nearly enough.  New species are discovered all the time.  New fossil finds occur frequently.  New functions for DNA sequences are being uncovered.  The list goes on and on.  What does this mean for a TOL? It shows that we are nowhere near knowing enough to fully understand life’s complexity.  Often, scientific ignorance such as this is exploited by ID proponents to allow for a god of the gaps argument.

Imperfect models – As with any other mathematical model used to predict the world, the algorithms used to create phylogenetic trees are imperfect.  In fact, with the exact same dataset, different models will yield different trees.  The differences are slight but real.  Think about the models that are used to predict hurricane paths.  Remember those different colored lines all showing slightly different paths?  They all start with the same data, but the models all yield different results.  The same issues are present in phylogenetic tree creation.

Confounding biology – The real world is never as simple as we want it to be.  In creating phylogenetic trees, we have to contend with biological phenomena that confuse and distort our understanding of evolutionary relationships.  This includes such factors as horizontal gene transfer, gene duplication, cross-breading, etc.  These factors lead to the crisscrossed and tangled branches that we now know make up TOL.

Gene evolution and speciation are independent – This factor is a subtle, yet important distinction.  Speciation can occur instantly (from a geological event or even a single mutation). The accumulation of mutation associated with speciation events could then lag behind what would normally be predicted from speciation.

I hope that I have shed some light on the validity of phylogenetic trees in spite of the numerous problems in creating them.  For Part 2, I will directly address Casey Luskin’s points from the Discovery Institute’s Evolution News and Views blog.

Creation of the phylogenetic tree

First, I downloaded the sequences for the cytochrome B genes from the listed species.  I picked cytochrome B because I could find it in all the species.  The sequences were found on the NCBI website (shown below in small font).  After obtaining all the sequences, I entered them into a program called Clustal W.  This program aligns DNA or protein sequences.  I chose the Kyoto University Bioinformatics Center because it has a nice interface.  After aligning my sequences, I chose the option to create a tree based on the N-J algorithm.  The result is the tree above.

>gorilla (gorilla gorilla)

ATGACCCCTATACGCAAAACTAACCCACTAGCAAAACTAATTAACCACTCATTCATTGACCTCCCTACCCCGTCCAACATCTCCACATGATGAAACTTCGGCTCACTCCTTGGTGCCTGCTTAATCCTTCAAATCACCACAGGGCTATTCCTAGCCATACACTACTCACCTGATGCCTCAACCGCCTTCTCATCAATCGCCCACATCACCCGAGATGTAAACTATGGCTGAACCATCCGCTACCTCCACGCTAACGGCGCCTCAATATTCTTCATTTGCCTCTTTCTACACATCGGCCGGGGCCTATACTACGGCTCATTTCTCCACCAAGAAACCTGAAACATCGGCATCATCCTCCTACTCACAACCATAGCAACAGCCTTCATAGGCTATGTCCTCCCATGAGGCCAAATATCCTTCTGAGGGGCCACAGTAATCACAAACTTGCTATCCGCCATCCCGTACATCGGAACAGATCTAGTCCAATGAGTTTGAGGTGGTTACTCAGTAGATAGCCCTACCCTTACACGATTCTTTACCTTCCACTTTATCCTACCCTTCATTATCACAGCCCTAACAACCCTCCATCTCCTATTTCTACACGAAACAGGATCAAACAACCCTCTAGGCATCCCCTCCCACTCTGACAAAATCACCTTCCACCCCTACTACACAATCAAAGACATCCTAGGCCTATTCTTCTTTCTCCTGACCTTGATAACATTAACACTATTCTCACCAGACCTCCTAGGAGACCCAGACAACTACACTTTAGCCAACCCCCTAAACACCCCACCCCACATCAAACCCGAATGATATTTCCTATTTGCCTACGCAATTCTCCGATCTGTCCCCAATAAACTAGGAGGCGTCTTAGCTCTATTACTATCCATTCTCATCCTAACAATAATTCCTATTCTCCACATATCCAAACAACAAAGCATAATATTCCGCCCATTAAGCCAACTACTCTACTGATTCCTAATCGCAAACCTCTTCACCCTAACCTGAATCGGAGGACAACCAGTAAGCTACCCCTTCATTACCATTGGGCAAGTAGCATCCGTACTATACTTCACGACAATCCTATTCCTGATACCAATCACATCCCTGATCGAAAACAAAATACTCAAATGAACCT

>orangutan (Pongo abelii)

ATGACCTCAACACGTAAAACCAACCCACTAATAAAATTAATCAACCACTCACTTATCGACCTCCCCACCCCATCAAACATCTCCGCATGATGGAACTTCGGCTCACTCCTAGGCGCCTGCTTAATCATCCAAATCACCACTGGACTATTCCTAGCTATACATTATTCACCAGACGCCTCCACTGCCTTTTCATCAATCGCCCACATCACTCGAGATGTAAACTACGGCTGAATAATTCGCCACCTCCACGCTAACGGCGCCTCAATATTCTTTATCTGCCTCTTCTTACATATCGGCCGAGGCCTATACTATGGCTCATTCACCCACCTAGAAACCTGAAACATCGGCATCATCCTACTATTTACAACTATAATAACAGCCTTCATAGGTTACGTCCTCCCATGAGGCCAAATATCCTTCTGAGGAGCCACAGTAATCACAAATCTACTGTCCGCCATCCCATACATTGGAACAGACCTGGTCCAATGAGTCTGAGGTGGCTACTCAGTAAATAGCCCCACTCTAACACGATTCTTCACCCTACACTTCATACTACCCTTCATTATTACAGCCCTAACAACTCTACACCTCTTATTCCTACACGAAACAGGATCAAATAACCCCCTGGGAATCCCCTCCCATTCCGACAAAATCACCTTCCACCCCTACTACACAATCAAAGACATCCTAGGCCTACTCCTTTTTCTCCTCGCCCTAATAACACTAACACTACTCTCACCAGACCTCCTAAGCGACCCAGACAACTACACCTTAGCTAACCCCCTAAGCACCCCACCCCACATTAAACCCGAATGATATTTCCTATTCGCCTACGCAATCCTACGATCCGTCCCCAACAAACTAGGAGGTGTAATAGCCCTCATACTATCCATCCTAATCCTAACAACAATCCCTGCCCTTCACATGTCCAAGCAACAGAGCATAACATTTCGCCCATTGAGCCAATTCCTATATTGACTTTTAATCGCCGACCTTCTAATTCTCACCTGAATTGGAGGGCAACCAGTAAGCTACCCCTTCATCACCATTAGCCAAGTAGCATCCACATTGTACTTCACTACTATCCTTCTACTTATACCAGCCTCTTCCCTGATCGAAAACCACATACTCAAATGAACCT

>human (homo sapiens)

ATGACCCCAATACGCAAAATTAACCCCCTAATAAAATTAATTAACCACTCATTCATCGACCTCCCCACCCCATCCAACATCTCCGCATGATGAAACTTCGGCTCACTCCTTGGCGCCTGCCTGATCCTCCAAATCACCACAGGACTATTCCTAGCCATACACTACTCACCAGACGCCTCAACCGCCTTTTCATCAATCGCCCACATCACTCGAGACGTAAATTATGGCTGAATCATCCGCTACCTTCACGCCAATGGCGCCTCAATATTCTTTATCTGCCTCTTCCTACACATCGGGCGAGGCCTATATTACGGATCATTTCTCTACTCAGAAACCTGAAACATCGGCATTATCCTCCTGCTTGCAACTATAGCAACAGCCTTCATAGGCTATGTCCTCCCGTGAGGCCAAATATCATTCTGAGGGGCCACAGTAATTACAAACTTACTATCCGCCATCCCATACATTGGGACAGACCTAGTTCAATGAATCTGAGGAGGCTACTCAGTAGACAGTCCCACCCTCACACGATTCTTTACCTTTCACTTCATCTTACCCTTCATTATTGCAGCCCTAGCAGCACTCCACCTCCTATTCTTGCACGAAACGGGATCAAACAACCCCCTAGGAATCACCTCCCATTCCGATAAAATCACCTTCCACCCTTACTACACAATCAAAGACGCCCTCGGCTTACTTCTCTTCCTTCTCTCCTTAATGACATTAACACTATTCTCACCAGACCTCCTAGGCGACCCAGACAATTATACCCTAGCCAACCCCTTAAACACCCCTCCCCACATCAAGCCCGAATGATATTTCCTATTCGCCTACACAATTCTCCGATCCGTCCCTAACAAACTAGGAGGCGTCCTTGCCCTATTACTATCCATCCTCATCCTAGCAATAATCCCCATCCTCCATATATCCAAACAACAAAGCATAATATTTCGCCCACTAAGCCAATCACTTTATTGACTCCTAGCCGCAGACCTCCTCATTCTAACCTGAATCGGAGGACAACCAGTAAGCTACCCTTTTACCATCATTGGACAAGTAGCATCCGTACTATACTTCACAACAATCCTAATCCTAATACCAACTATCTCCCTAATTGAAAACAAAATACTCAAAT

>Chimpanzee (Pan troglodytes)

ATGACCCCGACACGCAAAATTAACCCACTAATAAAATTAATTAATCACTCATTTATCGACCTCCCCACCCCATCCAACATTTCCGCATGATGGAACTTCGGCTCACTTCTCGGCGCCTGCCTAATCCTTCAAATTACCACAGGATTATTCCTAGCTATACACTACTCACCAGACGCCTCAACCGCCTTCTCGTCGATCGCCCACATCACCCGAGACGTAAACTATGGTTGGATCATCCGCTACCTCCACGCTAACGGCGCCTCAATATTTTTTATCTGCCTCTTCCTACACATCGGCCGAGGTCTATATTACGGCTCATTTCTCTACCTAGAAACCTGAAACATTGGCATTATCCTCTTGCTCACAACCATAGCAACAGCCTTTATGGGCTATGTCCTCCCATGAGGCCAAATATCCTTCTGAGGAGCCACAGTAATTACAAACCTACTGTCCGCTATCCCATACATCGGAACAGACCTGGTCCAGTGAGTCTGAGGAGGCTACTCAGTAGACAGCCCTACCCTTACACGATTCTTCACCTTCCACTTTATCTTACCCTTCATCATCACAGCCCTAACAACACTTCATCTCCTATTCTTACACGAAACAGGATCAAATAACCCCCTAGGAATCACCTCCCACTCCGACAAAATTACCTTCCACCCCTACTACACAATCAAAGATATCCTTGGCTTATTCCTTTTCCTCCTTATCCTAATGACATTAACACTATTCTCACCAGGCCTCCTAGGCGATCCAGACAACTATACCCTAGCTAACCCCCTAAACACCCCACCCCACATTAAACCCGAGTGATACTTTCTATTTGCCTACACAATCCTCCGATCCATCCCCAACAAACTAGGAGGCGTCCTCGCCCTACTACTATCTATCCTAATCCTAACAGCAATCCCTGTCCTCCACACATCCAAACAACAAAGCATAATATTTCGCCCACTAAGCCAACTGCTTTACTGACTCCTAGCCACAGACCTCCTCATCCTAACCTGAATCGGAGGACAACCAGTAAGCTACCCCTTCATCACCATCGGACAAATAGCATCCGTATTATACTTCACAACAATCCTAATCCTAATACCAATCGCCTCTCTAATCGAAAACAAAATACTTGAATGAACCT

>Gibbon (Hylobates lar)

ATGACCCCCCTGCGCAAAACTAACCCACTAATAAAACTAATCAACCACTCACTTATCGACCTTCCAGCCCCATCCAACATTTCTATATGATGAAACTTTGGTTCACTCCTAGGCGCCTGCTTGATCCTCCAGATCATCACAGGATTATTTTTAGCCATACACTACACACCAGATGCCTCCACAGCTTTCTCATCAGTAGCTCACATCACCCGAGACGTAAACTACGGCTGAATCATCCGCTACCTTCACGCCAACGGTGCCTCAATATTTTTTATCTGCCTATTCCTACACATCGGCCGAGGCCTATACTACGGTTCATTCCTTTACCTAGAAACCTGAAATATTGGCATTATCCTCCTACTCGCAACCATAGCAACAGCCTTCATGGGCTATGTCCTCCCATGAGGCCAAATATCCTTTTGAGGGGCCACAGTAATCACAAACCTACTATCCGCCGTCCCATACATCGGAACAGATCTAGTCCAATGGGTCTGAGGCGGCTACTCAGTAGATAACGCCACACTCACACGCTTTTTCACCTTTCACTTCATCCTACCTTTCATTATCACGGCCCTAGCAGCCCTGCACCTTCTATTCCTACACGAGACAGGATCAAACAATCCCTTAGGCATCTCCTCCCAACCAGACAAAATCGCCTTCCACCCCTACTATACAATCAAAGACATCCTAGGACTATTTCTCCTCCTCCTCATACTAATAAGCCTAGTACTATTCTCACCCGACCTCCTAGGCGACCCGAGCAACTATACCCAGGCTAATCCCCTAAACACCCCTCCCCACATCAAACCCGAATGATACTTTTTATTCGCATACGCAATTCTACGGTCCGTCCCTAATAAATTGGGAGGCGTACTAGCCCTCCTACTATCAATCCTCATCCTAGCAATAATCCCCGCACTCCACACAGCTAAACAGCAAAGCATGATATTTCGCCCACTAAGCCAGCTCACGTACTGACTCCTAGTAATAAACTTACTGATTCTCACATGAATCGGAGGACAACCGGTAAGCTACCCATTTATCACCATTGGACAAGTGGCATCCGCACTATACTTCACCACAATCCTAGTACTTATACCAGCCGCCTCCCTAATCGAAAACAAAATACTCAAATGAACCT

>Rhesus Macaque (Macaca mulatta)

ATGACTCCAATACGCAAATCCAACCCAATCCTAAAAATAATTAATCGCTCCTTCATCGATTTACCCGCCCCACCCAACCTCTCCATATGGTGAAACTTTGGCTCACTTCTTGCAGCCTGCCTAATTTTACAAATCATCACAGGCCTACTCCTAGCAATACACTACTCACCAGACACCTCCTCCGCCTTCTCCTCAATTGCACATATCACCCGAGATGTAAAGTACGGCTGAATCACTCGCTACCTCCACGCCAATGGTGCCTCTATACTCTTCATTTGCCTTTTCCTACACATCGGTCGGGGCCTTTACTACGGCTCATACCTCCTCCTAGAAACCTGAAACATTGGTATTATACTCCTTCTTATAACTATAACAACGGCTTTCATGGGTTATGTTCTCCCATGAGGCCAAATATCATTCTGGGGAGCAACAGTAATCACAAACCTGCTATCAGCAATCCCGTATATCGGAACCAATCTCGTCCAATGAATCTGAGGAGGATACGCCATCGACAGCCCTACTCTCACACGATTCTTCACCTTACACTTTATCCTACCCTTCATCATCATCGCCCTCACAACCGTGCACCTACTATTCCTGCACGAAACAGGATCAAACAACCCTTGCGGAATCTCCTCCGACTCAGACAAAATCGCCTTCCACCCCTACTACACAACCAAAGACATCCTGGGCCTAGTCCTCCTTCTCTTCATCCTAGCAACACTAACACTACTCTCACCCAACCTCCTAAACGACCCAGACAACTACATTCCAGCCGACCCATTAAACACTCCCCCACATATCAAACCAGAGTGATACTTCCTATTTGCATACACAATCCTACGATCCATCCCCAACAAACTGGGAGGCGTACTAGCACTCTTTCTATCGATCCTCATTCTAGCAGCCATCCCTATACTTCACAAATCCAAACAACAAAGCATAATATTCCGCCCACTCAGCCAATTTCTATTCTGACTCCTAATCACAATCCTATTGACCCTTACCTGAATTGGAAGCGAACCAGTAGTCCAACCCCTTACCACTATCGGCCAAGTAGCATCCATAATATACTTCATCACAATTCTAATCCTAATACCACTGGCCTCCCTAATCGAAAACAACCTACTCAAGTGAACTT

>White-fronted Capuchin (Cebus albifrons)

ATGACCTCTCCCCGCAAAACACACCCATTAATAAAAATTATTAATAGTTCATTTATTGATCTGCCCACACCATCCAACATCTCCTCCTGATGAAACTTCGGATCACTTCTAGGCGCCTGCCTAATAATTCAAATTACCACAGGCCTATTCTTAGCGATACACTATACGCCAGACACCTCAACCGCCTTCTCCTCAGTAGCACATATTACCCGAGATATTAATTACGGTTGAATAATCCGCCTCCTACACGCCAATGGTGCCTCCATATTTTTTGTGTGCTTATTTCTCCACACTGGCCGAGGCCTCTACTACGGATCTTTTCTCTTTCTAAACACCTGAAATATTGGTACAATCCTATTATTAATAACAATAGCCACAGCCTTTATAGGCTATGTCTTACCGTGAGGCCAAATATCATTCTGAGGAGCCACAGTTATTACAAATCTTCTATCAGCCATCCCCTATACCGGACATAACCTTGTACAATGAATCTGAGGTGGCTTTTCAGTAGATAAACCCACCCTCACACGATTCTTTACCTTTCACTTTATTTTACCTTTTATTATCACAGCCCTAGCAACTATTCACCTTTTATTTCTACATGAAACAGGCTCAAATAATCCATCAGGAATAGCATCTAGCCCCGATAAAATTATATTCCATCCCTACTACACAACCAAAGATATTTTTGGATTAACCCTTCTTCTCCTACTCCTTACAAGCCTAACCCTATTTACCCCCGACCTTTTAACTGACCCAGATAACTACACACTAGCCAACCCCCTTAATACTCCACCCCATATTAAGCCAGAGTGATACTTTCTATTCGCATACACAATTTTACGATCTATTCCAAATAAACTAGGAGGTGTTCTAGCTCTTCTATTATCTATTATAATCCTAACAATTATCCCTGCCACTCACCTATCCAAACAACAAAGTATAATATTCCGACCAATCACCCAAATCCTATTCTGAACCCTAGCAGCCGATCTACTTACACTTACATGAATTGGAGGCCAACCAGTAGAATACCCCTTTGAAGCCATTGGCCAAACCGCATCTATTGCTTACTTCCTTATTATTACTCTAATTCCTCTATCAGCCCTAACTGAAAATAAGCTACTTAAATGATAA

Two different techniques yielded the exact same tree.  Both trees created here match to a tree that is created by looking at physical similarities and differences.  This one example should be enough to show the validity of phylogenetic trees, but you know nothing would be sufficient to convince the Discovery Institute.

So I have shown a couple reasons why I think phylogenetic trees are valid.  Keep these in mind when you read about the supposed failures of phylogenetic trees from the ID proponents.  Next, I am going to describes some of the difficulties in creating a complete TOL.  Remember, these issues do not mean that common descent is wrong, it just means that our ability to decipher the tree of life is limited.  A weakness in data is not a weakness in a theory.

Difficulties in creating a tree of life

There are many problems with creating a complete tree of life that takes into account every piece of data available.  Think about all the information and computation required for the creation of an all-encompassing  phylogenetic tree. In addition, a TOL has to determine the relatedness of organisms that might not have a common ancestor for over half a billion years.  Below is an overview of some other problems in creating such a tree.

A response to the Discovery Institute’s criticism of Wetherington’s expert testimony

In a recent post over at the Discovery Institute’s blog, Evolution News and Views,David Klinghoffer writes about the expert testimony of Ronald Wetherington. Ronald Wetherington is an anthropology professor at Southern Methodist University.  In his post, Klinghoeffer claims that Wetherington was “sloppy with his facts.”  Unfortunately, I can’t find a transcript of Wetherington’s testimony, but a recording can be found here.

Some of the criticisms focus on hominid evolution, the subject of Wetherington’s expertise.  This is not my expertise and I will leave it up to the readers to decide. However, I find it hard to believe someone from the Discovery Institute over an expert.

Lets look at some of Klinghoeffer’s specific claims:

Klinghoffer criticizes evolution of the mammalian eye. He describes Wetherington’s discussion of the subject as “laughably simplifies what eye-evolution would entail.” However, he never gives any real criticism of the ideas except to say that it is absurd. He never makes specific criticisms of current models of eye evolution, but he does lend support to the idea of eye evolution by quoting Sean B. Carroll.  Carroll warns us about “simple” eyespots (believed to be precursors to modern eyes): “But do not be fooled by these eyes’ simple construction and appearance. They are built with and use many of the ingredients used in fancier eyes.”  This is a good point and shows the reducible components of the eye.

Klinghoffer then goes on to attack Wetherington’s use of genetics as support for evolutionary theory. Klinghoffer’s arguments are really just an exercise in quote mining to support the logical fallacy of personal disbelief.  He quotes from a 2000 article in Annual Review of Genomics and Human Genetics where the authors describe the “mystery” of how mutation and natural selection resulted in the complexity of life today. He also quotes Frank Harold as saying that “there are presently no detailed Darwinian accounts of the evolution of any biochemical or cellular system, only a variety of wishful speculations.” Neither reference is evidence against evolution, but are an acknowledgment of our incomplete understanding of evolution.

Klinghoffer says:

The more we know about genetics, the more we must, if we are honest with ourselves, doubt Darwin.

This is simply ridiculous. Genetics provides excellent support for evolution. For example, all living animals on earth have the same genetic code in their DNA. Dead viruses found in genomes can be traced back to when they were inserted along an evolutionary tree. Gene similarities between closely related species are more similar than those same genes between less related species. The list goes on, but it is clear that genetics provides excellent support for evolution, not the other way around.

Klinghoffer then goes on to regurgitate the deeply flawed observation by Michael Behe about the rate of mutations in humans by looking at a pair of mutation in the organism that causes malaria (Plasmodium falciparum).  I discuss Behe’s flawed reasoning here.

Klinghoffer has a few other nit-picky complaints towards the end:

-Wetherignton says that the term missing links isn’t used too much anymore.  Klinghoffer says that “ it is used in places like Science, Nature, Paleobiology, the American Journal of Physical Anthropology, and elsewhere.  My own experience tells me that Wetherington is correct, but just to be sure, I did a quick pubmed search with “missing link” and evolution. Only 96 results.  Wetherignton makes the point here that the reason we don’t use the term anymore is because it is inaccurate.  Once a fossil is found, it is no longer missing.  Besides, once one missing link fills in a gap, there are now two gaps to fill.

-Wetherington says Cambrian explosion lasted “at least 25 million years.” Klinghoffer says it was “under 10 million.”  Geez, what a terrible mistake of Wetherington, I guess evolution is completely false!  Seriously though, the length of the Cambrian explosion seems to vary dependent on who you ask and how you define it.  For example, Charles Marshall of Harvard writes:

Depending on when exactly one thinks the Cambrian “explosion” began, it is clear that there is a considerable temporal anatomy to the radiation. From the first appearance of heavily skeletonized animals to the first body fossils of trilobites, the radiation took some 20 million years. If one starts with the first abundant trace fossils through to the end of the Cambrian, then the radiation ran for some 65 million years.

-Wetherington says the explosion was “dominated by two phyla.” Klinghoffer says that Wetherington is wrong and actually  “19 of 28 phyla appeared.”   “Dominated by” is not the same as “appeared.”  Klinghoffer is trying to pull a fast one here.

-Klinghoffer complains that Wetherington confused a taxonomic class with a taxonomic order. Talk about grasping at straws

-Wetherington discusses research showing that hox genes can be interchanged between species. Klinghoffer denies this possibility  and claims this is “a piece of information that would startle Darwinian biologists.” Yet,  it has been shown that you can replace a Drosophilia Hox gene with a mouse Hox gene.  The switch leads to legs instead of wings, but illustrates overlap of function that Wetherington was talking about.  There are many other examples in the literature.

Finally, Klinghoffer has an excellent quote in this post that describes the work of ID proponents:

it seems obvious that men and women who invest themselves in their work over a lifetime may come to tell lies to themselves without ever knowing it, in order to maintain crucial fictions on which their life’s work depends. It’s human nature.

Only, he was talking about research scientists instead of ID proponents. However, evolution has facts and experimental data to support it, while intelligent design has only human intuition and logical fallacies to back it up.