Uncommon Descent Serving The Intelligent Design Community

Retroviral promoters in the human genome


The paper whose abstract lies below the fold has been cited as supportive of intelligent design here by my friend Casey Luskin of the Discovery Institute . I’m afraid I disagree with Casey’s analysis but I don’t have access to the full paper and would welcome review of my take on it from someone with access and expertise in virology. I’ve never agreed that ID, per se, predicts that “junk DNA” isn’t really junk. That’s a prediction based on young earth creationism. If an omnipotent designer created a perfect human genome 6,000 years ago then we might reasonably expect most of it today would still be functional. Design detection in and of itself does not predict any specific state of perfection or decay in the design. Thus the common assertion that “ID” predicts junk DNA will have function is not strictly an ID prediction at all but rather a young earth creation science prediction. Failure to make the origin of the predictions clear in these cases is a big reason why we keep getting slapped down in courts. It’s too transparent that design detection alone doesn’t predict things about junk DNA. You have to add in some young earth creationism to get there.

Retroviral promoters in the human genome
Andrew B. Conley , Jittima Piriyapongsa and I. King Jordan
School of Biology, Georgia Institute of Technology, 310 Ferst Drive, Atlanta, GA 30306, USA

Motivation: Endogenous retrovirus (ERV) elements have been shown to contribute promoter sequences that can initiate transcription of adjacent human genes. However, the extent to which retroviral sequences initiate transcription within the human genome is currently unknown. We analyzed genome sequence and high-throughput expression data to systematically evaluate the presence of retroviral promoters in the human genome.

Results: We report the existence of 51,197 ERV-derived promoter sequences that initiate transcription within the human genome, including 1743 cases where transcription is initiated from ERV sequences that are located in gene proximal promoter or 5′ untranslated regions (UTRs). A total of 114 of the ERV-derived transcription start sites can be demonstrated to drive transcription of 97 human genes, producing chimeric transcripts that are initiated within ERV long terminal repeat (LTR) sequences and read-through into known gene sequences. ERV promoters drive tissue-specific and lineage-specific patterns of gene expression and contribute to expression divergence between paralogs. These data illustrate the potential of retroviral sequences to regulate human transcription on a large scale consistent with a substantial effect of ERVs on the function and evolution of the human genome.

First of all one must understand that retroviruses all incorporate a promoter region. They must. Their modus operandi in reproduction is to commandeer the DNA transcription machinery in a host cell so that it manufactures copies of the viral particle. Basically the virus uses reverse transcriptase to insert a copy of its own code into the host DNA. The insertion includes a so-called “promoter region” which causes the victimized cell to execute the foreign instructions.

Sometimes the viral promoter region will also promote adjacent non-foreign code to be transcribed as well as the foreign code. Indeed, this is a big problem for “gene therapy” in medicine. In one method of gene therapy an engineered retrovirus is used to deliver a good copy of a defective gene (along with a promoter region) into the DNA of someone with a genetic disease. As of yet, unfortunately, the insertion point of the good gene cannot be controlled and this lack of control sometimes results in very undesirable side effects. The most common undesirable side effect AFAIK is cancer.

So the way I read the referenced paper is that it simply identified 114 ERV promoter regions in the human genome which happen to be promoting a non-foreign human gene. Given that it was only 114 out of nearly 60,000 ERV promoter cites this doesn’t seem to have any bearing on intelligent design but is merely a predictable consequence of so many ERV promoters embedded in the genome. Some of these 114 may actually have some beneficial effect but that’s doubtful as random changes in finely tuned machines rarely have any benefit. I think it’s more a matter of these 114 not having any really negative effect which makes them invisible to natural selection and thus they can persist in the genome. If there are a larger number of these among the 60,000 ERV promoter regions I don’t find that something that should be unexpected. After all, the human genes these ERV promoters are promoting obviously have some purpose else they wouldn’t be human genes in the first place. These genes must have had their own native promoters at some time in the past and it may simply be that the ERV promoter made the original promoter redundant.

I don’t see at all how this either supports or detracts from the intelligent design hypothesis. The only way I’ve ever seen the existence of retroviruses as supportive of intelligent design is to propose them as a good candidate for a tool that a hypothetical designer could use to effect evolutionary change. A highly infectious airborne or waterborne retrovirus that targets a specific species could insert heritable genetic material into the germ line of that species. The inserted material could cause any kind of change in the species the designer desired including saltation into a new species. As our own abilities as “intelligent designers” increases we’ll likely be altering the course of evolution ourselves in this manner. Obviously it doesn’t require anywhere near omnipotence to effectively accomplish any wholesale tinkering with the evolution of life.

Thanks Dave for your remarks. I have to say, as I am not a YEC, I was puzzled as to why you seem to imply that YEC thinking inspired my “junk DNA” arguments. Dembski is not a YEC and yet as I observed in my post, he said: “If, on the other hand, organisms are designed, we expect DNA, as much as possible, to exhibit function.” So ID’s prediction of function for “junk” DNA does not derive from issues relating to the timescale. Rather, such predictions stem from the fact that intelligent agents design things for a purpose. Sure, that intended functions can be lost over time via natural processes. But overall, there’s an expectation of function that Dembski correctly identifies. Regarding the paper, the excerpt you include above fails to quote the crucial interpretation of the data by the authors that I quoted in my post: “These data suggest that ERVs may regulate human transcription on a large scale.” The question is, exactly what are they regulating the transcription of? As you correctly state in the abstract, they found “51,197 ERV-derived promoter sequences that initiate transcription within the human genome.” The paper reports that “most of the ERV promoters identified with CAGE map to intergenic regions.” However, using the CAGE method, 9292 promoters were found to be "gene-associated." The paper says that “these ERV-derived promoters are likely to be responsible for generating alternative transcripts of human genes.” That’s very interesting, to say the least. In the end, total functionality for ERVs may turn largely on the question, what do these intergenic regions do? As we know, in recent years, geneticists have become increasingly convinced that intergenic regions of DNA have important regulatory functions. This is ongoing research but the point is that ERVs are playing a major role in initiating transcription of huge portoins of the genome–and that sounds suspiciously function-like to me. A lot of it is intergenic but according to this paper, at least 20% is associated with genes. And if more than our genic DNA plays a role in our cells (like geneticists are beginning to believe), then ERVs may have massive importance. I think that’s significant. Casey Luskin
Dave, I agree with some of your remarks about the paper, but not with your general point that the prediction of function for "junk" DNA is linked to YEC. I am no YE creationist at all, but still I find very natural to anticipate function for an entity which represents more than 98% of the human genome. Darwinian theory has affirmed for a long time that most of the information in the human genome is "junk", and I must confess that I find that notion rather puzzling even from the point of view of a theory vastly based on random variation. The idea that humans bear such a big burden of errors from their evolutionary history is rather uncomfortable for the supposed efficiency of NS. But we know how darwinists are always ready to exploit anything, and I mean really anything, to sponsor their theory, and that explains why junk DNA was immediately coopted as evidence of the mysterious workings of evolution. From the point of view of ID, instead, although I certainly agree with you that "design detection in and of itself does not predict any specific state of perfection or decay in the design", still I think that a state where only 1.5% of the written information is useful, and the rest is at best non functional, does not seem to be a very credible model of design. One thing is to affirm that design needs not be perfect (I absolutely agree on that) or that it can be subject to some decay, another thing is to envision a complex and highly efficient machine like the human body, certainly based on a very high level of design, being reproduced and maintained by a bulk of information, all of it preciously stored and conserved and defended, 98% of which would be completely useless. Frankly, if that scenario is already not so credible from the point of view of darwinian evolution, it is almost inconceivable from the point of view of ID. I think that's what IDists mean when they say that ID predicts some function for non coding DNA. That does not imply YEC, nor a perfect design by an omnipotent God. It is just the logical consequence of believing that, if a structure is carefully designed, then it is very unlikely that it carries in itself 98% of wrong information. Indeed, my personal view is that non coding DNA will be shown to play a fundamental role in transcription regulation. I think both introns and transposons are probably deeply involved in that, although we still have to understand how (but obviously, some aspects are already beginning to be understood). Regarding ERVs, I really don't know. There are some hints of a function for them, including the paper mentioned above, but your objections are certainly potentially valid. The issue is very technical, and so I think we should wait for further information before drawing conclusions. IMO, ERVs are not necessarily the most interesting part of non coding DNA, and they have been often the center of interest because they remain the most objective argument for universal common descent. Not being too much interested in the discussion about common descent, I think that I will keep an open mind about ERVs, at least until the evidence is less controversial. gpuccio
"These genes must have had their own native promoters at some time in the past and it may simply be that the ERV promoter made the original promoter redundant." It's not clear to me from the abstract that the genes were not retrotransposed pseudogenes lacking promoters. This could be a mechanism, or an important part of a process, turning pseudogenes into functional genes. Liz Lizard
While I generally agree with Dave that the ID prediction of "no junk" is rather spurious, not a necessity of ID, I do not concur with his view that it is a pure product of YEC thought. In Nature's Destiny, Denton clearly predicts a lack of junk. He, in fact, suggests that proof that a majority of human DNA is junk would falsify his theory. I don't follow his logic, but he's hardly YEC. That said, the flip side seems clear. Darwinists have calculated that only a small portion of human DNA can be functional or there would be too many mutations per generation to manage. (I wish I had a citation.) My calculations say that if there is more than one phenotype affecting mutation per generation, natural selection falls apart. In the known protein coding DNA, the average human has, I think, 3. This already presents a serous problem. If the amount of DNA that renders in the phenotype were 10 times the size of the identified active DNA, then we would be the product of 30 phenotype affecting mutations per generation. That would be a serous problem. My understanding is that about 30% of DNA is somewhat conserved between man and mouse. If natural selection, which only acts on the phenotype, is responsible for this conservation then at least 30% of human DNA renders in the phenotype. This is a serious problem for the darwinian theory. The other alternative is that there is another preservative in DNA, something other than the broad-based error correction technology that is know. This would suggest that a major discovery is called for. bFast
I think junk DNA is a given. There can be arguments over how much of a specific genome is junk and how much is useful. One cannot just look at humans to make this observation. Some very low species on the totem pole of life have very large genomes. There is a measure of genome size and it is called the C value and there is something called the C value paradox. From Wikipedia "The C-value enigma or C-value paradox is a term used to describe the complex puzzle surrounding the extensive variation in nuclear genome size among eukaryotic species. At the center of the C-value enigma is the observation that genome size does not correlate with organismal complexity. For example, many plant species and some single-celled protists, have genomes much larger than that of humans." So a lot of these large genomes is most likely non coding or not functional or junk DNA. jerry
Would you agree that Darwinism "predicts junk DNA", or perhaps "Darwinists have predicted junk DNA", but that ID does not necessarily predict purpose in said "junk"? Intelligent agents do plenty of things that make no sense to any outside observer. We can't attack Darwinists for their "God wouldn't do it that way" arguments, then say "God wouldn't make junk DNA". russ

Leave a Reply