Discovery Institute, primate limbs, and “junk DNA”

In his September 24, 2008 post entitled “Study Challenges Two Icons of Evolution: Functional Junk DNA Shows “Surprising” Genetic Differences Between Humans and Apes,” Casey Luskin shows us just how much he really is out of touch with molecular biology and modern evolutionary theory.  He refers to an interesting study that seems to have found at least one of the reasons why there is such a difference in human limbs compared to other primates.  This article was published in the September 5th issue of Science.  Luskin somehow thinks these findings go against evolutionary theory.  They are perfectly consistent with the theory, as you will see below.

Luskin promotes the idea that scientists still believe that all the sequence that does not directly code for a protein is “junk.” The article of which he refers made a discovery in a region which may have once been considered “junk,” but not for a long time.  Scientists have known that DNA that is far away from a gene (thousands of base pairs) can directly regulate its function for decades. Other regions are believed to provide structural functions or regulate a gene that is nowhere nearby (see siRNA). This stuff widely known and Luskin is either lying or is willfully ignorant.  I also want to put some blame on the people who are responsible for the Yale press release. They are helping perpetuate this myth of “junk DNA,” but Luskin should know better as a person heavily involved in the evolution/ intelligent design debate.

To make his point about how “junk DNA” is still used as an “icon” for evolution, Luskin quotes a human physiology textbook by Willam McArdle et al. (emphasis mine):

“junk DNA” “is considered defective” and are “inherited sequences [that] perform no currently known ‘genetically useful’ purpose, yet they remain part of the chromosomes.”

Forget that the book is not an evolutionary biology textbook or even a molecular biology textbook. Did you notice the qualifier “not currently known?” Once a stretch of DNA is found to have a function, then it isn’t considered “junk” anymore.  Yet, Luskin still takes the stand that finding anything important in what he calls “junk DNA” discredits evolution.  Nothing could be further from the truth.

Here is the meat of Luskin’s argument:

Most studies that have claimed that humans and apes have nearly identical genomes have primarily looked at the gene-coding portions of the genome, not the non-coding DNA (formerly claimed to be “junk”). Perhaps as biologists study the non-coding portions of our genome, they will find evidence that challenges two Darwinists icons: Not only does “junk” DNA have function, but humans aren’t as genetically similar to apes as was once thought.

We have already convered that “junk DNA” is a misnomer and he even says that it was “formerly claimed.”  So this being an “icon” of evolution doesn’t really make any sense.  But what about his argument that we “aren’t as genetically similar to apes?”

The article in question seems to promote the idea (or at least Luskin’s take on it) that humans and chimpanzees aren’t as similar as once thought due to the press release saying that the finding“was especially surprising, as the human and chimpanzee genomes are extremely similar overall.”

Lets do some overly simplified math to see if these findings contradict evolution. The region in question (HACNS1) is 546 bases long. They found 16 differences between humans and chimpanzees. So if we do the math (16/546), we find that there is approximately a 3% difference between the human and chimpanzee HACNS1 region. This percentage is exactly the same as the overall difference between the two organisms and matches nicely with the rate of mutation predicted by evolutionary theory since the time of human-chimpanzee split (~6 million years ago).

If the math all lines up, why is this finding surprising? It turns out that there are certain regions of DNA that are so critical that mutations in these regions would be devastating for their function. Only rare and beneficial mutations would be allowed to persist. In fact, this region is highly conserved throughout land animals and it has 4X as any mutations as would be expected based on this evolutionary constraint.

Tracer showing human expression in mouse

Tracer showing human expression by HACNS1 in mouse limb

The end result of these mutation seems to effect the way humans thumbs, wrists, and other limb characteristics develop. In fact, when the scientists put the human DNA region in question in mice with a tracer (see image) and compared it to chimpanzee region, they saw a stark contrast in the levels of the tracer between the two organism’s DNA. The regions where the tracer was observed corresponds to areas that seem to be responsible for both the greater flexibility of the human thumb and wrist. Of course more study needs to be done on the subject, but It does show that this small region could be important.

Everything in the report is consistent with evolutionary theory, despite what Luskin says. If anything it strengthens the argument that small changes (12 changes in 3 billion) between humans and chimpanzees can go a long way. Remember that the “icons” that Luskin speaks of are really strawmen created to divert unsuspecting readers away from the real arguments and evidences.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to FurlAdd to Newsvine

Functions of endogenous retroviruses does nothing for intelligent design

Scanning electron micrograph of HIV-1 (in green) budding from cultured lymphocyte.

On August 21, 2008, Casey Luskin wrote Large Scale Function for Endogenous Retroviruses: Intelligent Design Prediction Fulfilled While Another Darwinist Argument Bites the Dust. In this post, Luskin uses a recent article in Bioinformatics by Conley et al. to attack a piece of evidence brought up by Douglas Theobald in his 29+ Evidences for Macroevolution. This article can be found at TalkOrigins, and is a must read for anyone who is interested in the evolution debate.

Short response:

Luskin is not really addressing the evidence provided by Theobald. Luskin says that new evidence shows that endogenous retroviruses have function. Theobald never addressed this issue. Therefore, Luskin’s point is moot and should be disregarded as such.

Full Response:

The piece of evidence in question here concerns the presence of endogenous retroviruses (ERVs) found throughout genomes. When a retrovirus (such as HIV or HPV) infects a person, the virus gets incorporated into the person’s DNA. If a germ cell gets infected by one of these viruses, then it will become part of the offspring’s genome. Once it has become part of the genome, it is endogenous and will be passed down to successive generations.

Because these ERVs are passed down to offspring, one should be able to trace the introduction of different ERVs through evolutionary history by examining phylogenetic trees. As pointed out by Theobald, this is in fact what we see when we look at the ape family or the felines. We never see an ERV in the same position in two different distantly related apes without seeing it in the intervening species. So this provides strong support to evolutionary theory. An excellent overview of this can be found in episode 113 of the Evolution 101 podcast.

So how does Luskin try to disprove this evidence? He discusses the Bioinformatics article that points to ERVs affecting transcription of genes. Luskin states:

Douglas Theobald claims that “Endogenous retroviruses provide yet another example of molecular sequence evidence for universal common descent.” The presumption behind his argument is that endogenous retroviruses (ERVs) are functionless stretches of “junk” DNA that persist because they are “selfish”-but they have no function for the organism… The force of Theobald’s argument thus depends upon the premise that ERVs are selfish genetic “junk” that do not necessarily perform any useful function for their host.

However, this is not at all the point of Theobald’s argument. He says nothing about whether the ERVs are able to affect the function of genes. This is clearly a strawman that Luskin created for his uninformed readers.

The focus here is not if the ERVs have a function, it is do their positions in the genome match what a phylogenetic tree would predict.  The fact that ERVs affect gene expression is not unknown or unexpected. For one thing, retroviruses utilize their host cell’s machinery to express their own genes. So if the retrovirus gets put into the genome near another gene, why wouldn’t it affect the gene?

The real thing to ask yourself, or someone from the Discovery Institute, is why a Designer would need to put the skeletons of retroviruses by genes to effect its expression?  Other genes do not have these remnants of retroviruses and seem to function just fine.  This should really be the focus of someone interested in reality, but we know proponents of intelligent design are not interested in what is going on.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to FurlAdd to Newsvine

One step closer to a complete Tree of Life

The inability of scientists to create a complete Tree of Life (TOL) has been a talking point of ID proponents for some time. Indeed, a complete TOL based on DNA sequence alignments has yet to be completed. Instead of a single tree, usually ‘bushes’ of different clades are about as close as we have gotten. I don’t believe this is because it is impossible to make a complete tree that encompasses all known living organisms, it is simply another example of where the science is a work in progress. Molecular biology is in its infancy and the era of bioinformatics has just started. Nonetheless, IDers take our incomplete understanding as evidence against evolution (For the Discovery Institute’s take, see here and here).


These phylogenetic trees are made by taking DNA sequences from different organisms and aligning them with each other. Organisms that are closer evolutionarily will have more DNA in common and therefore the sequences will mostly match and be easily aligned. Organisms that are distantly related will have less similarities in their sequences. This is not an easy task and is fraught with error due to mutations that lead to an insertion of a piece of DNA, deletion of pieces of DNA, substitutions of DNA, etc.


In the June 20, 2008 issue of Science, Ari Löytynoja and Nick Goldman report a new and better way to align DNA sequences thereby creating better phylogenetic trees. Aligning DNA sequences is a mathematically complex process with several different algorithms designed to take into account small changes in the sequence. Don’t be scared, I am not going to bore you with the mathematical details (mostly because I don’t understand it).

According to the authors, the biggest problem with the current algorithms is that:

Traditional multiple sequence alignment methods disregard the phylogenetic implications of gap patterns that they create and infer systematically biased alignments with excess deletions and substitutions, too few insertions, and implausible insertion deletion–event histories”

Surprisingly, they found that simply adding more and more sequences from similar organisms did not increase the accuracy due to every additional sequence adding evolutionary time to the analysis and therefore more DNA deletions, insertions, or substitutions. This is where the authors say that their new method really shines due to its ability to utilize the phylogenetic relationship of sequences.


What does this all mean for the theory of evolution? Hopefully this will lead to better phylogenetic trees and bring us one step closer to that all important TOL. On the more technical side, the authors think that their new method will allow for a better understanding of the frequency of DNA insertions, deletions, and substitutions.

When its good to keep quiet

This post covers the May 27 and 28, 2008 posts entitled: “MSNBC’s Alan Boyle and Sean B. Carroll Argue Scientists Should Keep “Quiet” about Support for Intelligent Design”.  This post is broken down into two parts with the first Luskin’s answer to Carroll’s opinion and the second is Luskin attacking Carroll’s facts.  To be honest, I completely agree with Sean B. Carroll.  We should be teaching our children the best theories of the day, not unsupported ones.  Educators should keep quiet about Intelligent Design and any other unsupported idea. 

Luskin writes (emphasis mine):

The implication is clear: Boyle and Carroll think that there should be no academic freedom for scientists or educators to speak in favor of intelligent design. In Boyle and Carroll’s world, if you have real doubts about evolution, then like Newton, you should just keep “quiet.”

I have to hand it to Luskin here. He shows us what the “academic freedom” bills really are: Intelligent Design bills.  The rest is Luskin rehashing how all ideas should be given equal credence.

In the second part of the post, Luskin tries to poke holes in evolutionary theory by pointing to the unresolved issue of a complete tree of life (TOL). As I have written before, ID proponents love to point to unresolved issues in science (evolution, geology, etc) as somehow giving validity to ID. In this case, Luskin is using the imperfect science of creating phylogenetic trees. These may be imperfect due to mathmatical constraints, lateral gene transfer, high frequency of mutation, etc, but they generally produce a tree that is in agreement with evolutionary theory.  However, problems do arise when trying to work out the details.

The inability to construct robust phylogenetic trees is not due to evolutionists being wrong. If common descent was wrong, then you would never be able to produce trees that match up so well with evolutionary relationships. Luskin is taking small problems in evolutionary biology as reasons to discount it, while ignoring the complete lack of data supporting ID.

Luskin then writes about the relationships between genes of different species: 

“Since their DNA might be similar due to functional requirements and not inheritance from a common ancestor…”

This statement goes a long way to illustrate the lack of understanding that Luskin has in regard to conserved sequences. Genes are quite complex and are made of different regions. Some of these regions are the parts that are critical for the genes function, while other parts are unimportant. The less important parts are still similar between species, but since they are not important for function, they are definitely not similar due to simply functional requirements.

I want to finish by showing a quote of Luskins that illustrates a new technique used by IDers:

If the loss of function by turning off genes, and the usage of the same genes to build organs in vastly diverse organisms—a fact cited by design-proponents as supporting common design—are the best facts [Carroll] can muster against design, then it would appear that ID has very little to fear from the discoveries of evo-devo.

So IDers are now going to take evidence that really supports evolution and say that it supports intelligent design? Although intellectually dishonest, it is a brilliant move to persuade people to a side that has no evidence of its own.