What the Literature Says about Chromosomal Fusion and Why It Says It
Critics of our new book Science and Human Origins have been trying to manufacture a small ruckus over a comment I made in Chapter 4 about chromosomal fusion. David Klinghoffer has reviewed the whole episode here, but it boils down to Carl Zimmer refusing (for a time) to read the book to see what it said, and then asking the following question:
"An article on Evolution News & Views stated that the evidence for the fusion of human chromosome two "appears in a 'degenerate,' 'highly diverged' form that should not be the case if the joining happened in the recent past, circa 6 million years ago, as the Darwinian interpretation holds." Where is the scientific evidence for this?"Zimmer and others apparently want to make people believe I made those statements up. But I didn't. All Zimmer had to do to find out where they came from was to read the book and look up my citations. Apparently he's unwilling to do this. I'm not here to play games, but to have a serious scientific discussion. Here's the citation for those statements:
Yuxin Fan, Elena Linardopoulou, Cynthia Friedman, Eleanor Williams, and Barbara J. Trask, "Genomic Structure and Evolution of the Ancestral Chromosome Fusion Site in 2q13-2q14.1 and Paralogous Regions on Other Human Chromosomes," Genome Research, 12 (2002): 1651-62.The paper is available free online here. Why it was too difficult for Carl Zimmer to buy the book and/or look up a reference online is beyond me. Though I have purchased (and have enjoyed reading) a number of books by Carl Zimmer, David Klinghoffer even offered to send Zimmer a free copy of our book. Zimmer would have found that the statements I quoted from that paper are right there, in the paper, precisely as I quoted them, in context, as we will see below.
My chapter on chromosomal fusion in Science and Human Origins cited three quotes from that paper. Klinghoffer has reproduced that whole section here, but here's what I wrote:
Additionally, a paper in Genome Research found that the alleged telomeric sequences we do have are "degenerated significantly" and "highly diverged from the prototypic telomeric repeats." The paper is surprised at this finding, because the fusion event supposedly happened recently -- much too recent for such dramatic divergence of sequence. Thus the paper asks: "If the fusion occurred within the telomeric repeat arrays less than ~6 mya [million years ago], why are the arrays at the fusion site so degenerate?"47 The conclusion is this: If two chromosomes were fused end-to-end in humans, then a huge amount of alleged telomeric DNA is missing or garbled.Let's look at those three quotations one at a time.
- Quote 1: "degenerated significantly": The paper states: "The head-to-head arrays of repeats at the fusion site in RP11-395L14 have degenerated significantly (14%) from the near perfect arrays of (TTAGGG)n found at telomeres." So my comment there appears to be perfectly accurate.
- Quote 2: "highly diverged from the prototypic telomeric repeats": This quote requires a bit more explanation but it is also accurate and in-context. The paper states:
Two short arrays of degenerate telomere repeats, in addition to the arrays marking the fusion site, are found within 2qFus. They are 181 bp and 248 bp long, and 17 kb and 21 kb distal of the fusion site, respectively. Interstitial arrays of degenerate telomere arrays are common in the human genome, particularly in subtelomeric regions (Riethman et al. 2001). Like the array at the fusion site, these arrays are highly diverged from the prototypic telomeric repeats (70% and 86% identical to [TTAGGG]n, respectively). A SATR1 (satellite) repeat cluster within the block common to 2qFus, 9pter, 9q13, and 9p11.2-B (asterisks in Fig. 3) also shows high variability in length, especially when compared with the overall high identity of these blocks. (emphasis added)The general topic of this paragraph is other telomeric repeats, not the one that's thought to represent the fusion event. But it does mention the telomeric DNA thought to be at the fusion site: "Like the array at the fusion site, these arrays are highly diverged from the prototypic telomeric repeats." That statement from this paragraph says that both these other telomeric arrays, and the array at the fusion site, are "highly diverged from the prototypic telomeric repeats." So in quoting this statement, my chapter does NOT misstate what the paper says about the telomeric DNA at the fusion site. It is not out-of-context.
(Also, recall that the paper said, in specifically discussing only the DNA at the alleged fusion site, that it has "degenerated significantly" from normal telomeric DNA. So saying it's "highly diverged" from normal telomeric DNA amounts to about the same thing.)
- Quote 3: "If the fusion occurred within the telomeric repeat arrays less than ~6 mya, why are the arrays at the fusion site so degenerate?": That question appears in the paper exactly as I quote it. The reason the authors ask it is quite rational: In 6 million years, this stretch of DNA should not have accumulated very many mutations. But if this was originally the site of the fusion of two telomeres, then it has experienced numerous mutational changes, including the loss of many telomeric repeats. As I also note in the book:
Telomeric DNA at the ends of our chromosomes normally consists of thousands of repeats of the 6-base-pair sequence TTAGGG. But the alleged fusion point in human chromosome 2 contains far less telomeric DNA than it should if two chromosomes were fused end-to-end: As evolutionary biologist Daniel Fairbanks admits, the location only has 158 repeats, and only "44 are perfect copies" of the sequence. (Science and Human Origins, pp. 95-96)So they naturally wonder why the presumed fusion site is so different from normal telomeric DNA if that's what it was a mere 6 million years ago.
Last year we had a whole discussion thread here on ENV that looked at this part of the paper in more detail, here.
Now, the authors of this Genome Research paper believe that a fusion event took place. Fine -- I'm very open to that. After the passage that goes with the third quote above, they go on to offer three possible explanations for why the telomeric DNA data diverges so much from what would be expected had a fusion event occurred merely 6 million years ago. None of these explanations are persuasive, but here they are: (1) they suggest the chromosomes weren't really joined at their ends; (2) they arbitrarily invoke a "high rate of change" in that stretch of DNA, or (3) they suggest there were errors in the sequencing process. Here's what the paper says exactly:
There are three possible explanations: (1) Given the many instances of degenerate telomeric arrays within the subtelomeric regions of human chromosomes (Riethman et al. 2001), the chromosomes joined at interstitial arrays near, but not actually at, their ends. In this case, material from the very ends of the fusion partners would have been discarded. (2) The arrays were originally true terminal arrays that degenerated rapidly after the fusion. This high rate of change is plausible, given the remarkably high allelic variation observed at the fusion site. The arrays in the BAC and the sequence obtained by Ijdo et al. (1991) differ by 12%, which is high even if some differences are ascribed to experimental error. (3) Some array degeneracy could be a consequence of sequencing errors.Any or all of these may be possible, but "possible" doesn't necessarily equate with "convincing." The fact that they resort to the first explanation shows how wrong Kenneth Miller is to say that "[t]he forensic case of the missing chromosome is settled beyond any doubt." The evidence isn't nearly as perfect as he claims it is.
The fact that they invoke explanations two and three shows how much the observed telomeric DNA diverges from what they expected to find if there was a fusion event. If telomeric DNA is predicted to degenerate quickly after a fusion event, then it is not very persuasive to cite telomeric DNA as evidence for such an event. Moreover, when you cite errors in your own experimental methods to explain data that contradicts your hypothesis, that appears quite arbitrary and doesn't inspire confidence in the hypothesis.
These explanations are so weak, so ad hoc, that in my opinion they are hardly worth mentioning. And given that, as I explain in the book, and David Klinghoffer recently recounted here, it's not unusual to find interstitial telomeric sequences that have nothing to do with fusion events, there are good reasons to question them. Thus, a fourth possible explanation that the authors do not consider is that this is not a fusion site, and there was no chromosomal fusion.
All the recent yelling and shouting about the evidence for (or against) chromosomal fusion has distracted from the fact that in Science and Human Origins, my main argument about chromosomal fusion isn't to question whether it took place.
In fact, I'm totally open to the real prospect that human chromosome 2 indeed experienced a fusion event sometime in the human lineage. What's more, I think there is indeed some other evidence that suggests fusion -- it's just that the evidence from telomeric DNA for an alleged fusion site isn't nearly as clear-cut as is claimed. In fact, in the appendix to Intelligent Design 101 I co-wrote with Logan Gage: "We accept that there is good evidence that human chromosome 2 is composed of two fused chromosomes." Again, my main argument about chromosomal fusion isn't that there's no fusion -- it's that if there is fusion, that doesn't demonstrate human/ape common ancestry. As I wrote in Science and Human Origins:
In The Language of God, Collins argues that this chromosomal fusion explains why humans have one less pair of chromosomes than apes, claiming "it is very difficult to understand this observation without postulating a common ancestor."43This comes from the very beginning of my section on chromosomal fusion, and it is my opening argument on this point. Later in the book I write:
To the contrary, it is very easy to understand this evidence without postulating a common ancestor.
Assuming that human chromosome 2 is fused as Collins claims it is, human chromosomal fusion merely shows that at some point within our lineage, two chromosomes became fused. Logically speaking, this evidence tells us nothing about whether our human lineage leads back to a common ancestor with apes. Nor does it tell us whether the earliest humans were somehow ape-like. (Science and Human Origins, p. 92)
But -- and this is the key point -- even if human chromosome 2 is the result of two other chromosomes which became fused, this is not evidence for human/ape common ancestry. At most, it shows our human lineage experienced a chromosomal fusion event, but it does not tell us whether our lineage leads back to a common ancestor with apes. (Science and Human Origins, p. 97)Likewise, I co-wrote with Logan Gage in Intelligent Design 101: "Evidence for chromosomal fusion in humans simply indicates that, at some point within our human lineage, two chromosomes became fused. This tells us nothing about whether we share a common ancestor with apes."
So my main, or "key" point on this topic is that Carl Zimmer and others can present all the evidence for human chromosomal fusion they wish, and it still doesn't tell us whether we share a common ancestor with apes. Unfortunately, this crucial point has been lost in Zimmer's clamor over a citation he could have found if he'd just read the book.
If Zimmer or anyone else would like to reply to my arguments in the chapter, they are most welcome to do so. But I suggest that in the future, they first consult the book.