Of Mice, Men, and Coelacanths

_{December 11, 2006

Intelligent Design}

Share: Facebook; Twitter; LinkedIn; Flipboard; Print; Email

Alternate title: The Sound of the Tree of Life Exploding

Explosion
Comparing the sequence to other species also turned up a big surprise. When the researchers compared the human ultraconserved element to all the DNA sequences in the public database GenBank, the closest match was to DNA from the coelacanth…

Okay, so maybe it didn’t explode but some branches are bent and something is definitely fishy here. Rubin, Haussler, and Bejerano sure do turn up some interesting things. Our closest relative on the tree of life according to ultra-conserved DNA is a fish that’s been around unchanged for at least 360 million years.

Mobile DNA part of evolution’s toolbox
Thursday, May 4, 2006
Written by Howard Hughes Medical Institute

Ã¢â‚¬Å“The big question is whether this is a special case or whether it’s the tip of the iceberg,Ã¢â‚¬Â says Haussler. A report on the research is published in the May 4, 2006, issue of the journal Nature.

Haussler and his colleagues were led to the discovery through their work on the ultraconserved elements of the genome. One ultraconserved element in particular caught their eye. Ã¢â‚¬Å“We were very interested in this sequence, because it had a number of copies elsewhere in the genome,Ã¢â‚¬Â says postdoc Gill Bejerano, who is the first author of the study. Close copies of the sequence were ubiquitous in amphibians, birds, and mammals, indicating that it served an important function. Ã¢â‚¬Å“We found it in every species for which we have genomes, from frogs to humans,Ã¢â‚¬Â says Bejerano.

Comparing the sequence to other species also turned up a big surprise. When the researchers compared the human ultraconserved element to all the DNA sequences in the public database GenBank, the closest match was to DNA from the coelacanthÃ¢â‚¬â€an ancient fish thought to have gone extinct millions of years ago until a live specimen was caught in 1938 off the east coast of South Africa. The coelacanth is a descendant of the ancient marine organism that gave rise to the terrestrial vertebrates more than 360 million years ago. Humans are therefore separated from the coelacanth by hundreds of millions of years of evolution, yet the two organisms still share critical DNA sequences.

Read the whole article at the link above.

Comments

It was unexpected because there were multiple very similar copies in coelacanth, not just one -- if there were just one, you would just say it is the ortholog (corresponding instance in another genome) of the mammalian ultraconserved element. The message of the paper is that a (repetitive) transposable element in coelacanth gave rise to this ultraconserved element in mammals through exaptation. The following paper gives a much better example of the problem that I think you intended to seize upon in this blog entry. It's a little less sexy because it involves flies and not mammals, but it's recent work that shows the current state of the art (which is somewhat limited) in explaining why many gene trees disagree with the species tree. http://genetics.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pgen.0020173GeoMor_{December 12, 2006
December
12
Dec
12
12
2006
09:08 PM
9
09
08
PM
PDT}

Geomor Then why was the finding characterized as "unexpected"?DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
08:52 PM
8
08
52
PM
PDT}

DaveScot, the emphasized sentence says that PCB2 is more similar to the coelacanth sequence than to ISL1 (and other similar sequences in the human genome), which was obvious in the figure you previously pointed to. "Paralogs" refers to other copies of the sequence in the human genome. Please consider carefully the implications (or lack thereof) of similarity in multiple copies within the human genome versus copies found across several species, given that PCB2 is the only copy that is ultraconserved.GeoMor_{December 12, 2006
December
12
Dec
12
12
2006
08:39 PM
8
08
39
PM
PDT}

It's also interesting to note that coelacanths have a genome size and chromosome number comparable to humans. http://www.genomesize.com/result_species.php?id=4420 human c-value=3.50 chromosomes(2n)=46 http://www.genomesize.com/result_species.php?id=3598 coelacanth c-value=3.61 chromosomes (2n)=48 http://www.genomesize.com/result_species.php?id=906 chicken c-value=1.25 chromosomes (2n)=78 http://www.genomesize.com/result_species.php?id=294 clawed frog C-value=3.65 Chromosome Number (2n)=36 http://www.genomesize.com/result_species.php?id=5438 opossum C-value=5.22 Chromosome Number (2n)=18 More interesting than the coelacanth are the lungfishes such as this one: http://www.genomesize.com/result_species.php?id=3604 lungfish C-value=112.78 chromosome Number (2n)=38 Over 30 times the size of the human genome. I wonder what mysteries lurk therein. DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
08:18 PM
8
08
18
PM
PDT}

Here is where they (HHMI) got it from. uc.338 homologs. My emphasis. http://www.osti.gov/energycitations/servlets/purl/882757-lklotn/882757.PDF An Enhancer Near ISL1 and an Ultraconserved Exon of PCBP2 are Derived From a Retroposon Gill Bejerano, et al
One of the most evolutionarily constrained regions in mammalian genomes is ultraconserved element uc.338, a mammal-specific 223bp region perfectly conserved between human, mouse and rat, overlapping a short protein coding exon of PCBP26. This small region was observed to have multiple paralogs within the human genome, overlapping protein coding exons of otherwise unrelated genes, as well as conserved intronic and intergenic regions (Fig. S1). Unexpectedly, this region also has multiple homologs in coelacanth that are closer in sequence to the human ultraconserved element than many of its human paralogs (Fig. 1c).
It should be noted that the coelacanth genome has only a small fraction of it sequenced so far so there's much more yet to discover in it which probably prompted Haussler's "tip of the iceberg" comment.DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
07:44 PM
7
07
44
PM
PDT}

bFast
For chimp/human I count 80 matches. I count 54 that are colored, and an additional 26 which are not.
They are only not colored if they don't match, unless it is a deletion. So either you are way off or I am looking at the wrong page. For the record, looking at the segment labeled ISL1 on page 33, I see that the human/chimp/chicken are tied for closest to coelacanth. With the other vertebrates in the following order, Dog/Rat, Mouse/Opposum, and Frog. Not exactly the predicted tree of life but we are going off of a partial sample. I see from looking at the diagram at the top of page 33 that ISL1 is not 100% the same in human and chimp and the two are significantly different than the chicken.Jehu_{December 12, 2006
December
12
Dec
12
12
2006
07:40 PM
7
07
40
PM
PDT}

GeoMor
This can be verified by reading the actual paper that this news article is about (Bejerano et al. 2006); unfortunately I donÃ¢â‚¬â„¢t think itÃ¢â‚¬â„¢s open access, but at least the key misunderstanding here can be resolved from the abstract:
One of these, a more than 200-base-pair ultraconserved region, 100% identical in mammals, and 80% identical to the coelacanth SINE,Ã¢â‚¬Â¦ http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=16625209&dopt=Citation
That quote from the abstract clearly refers to PCB2, not ISL1.Jehu_{December 12, 2006
December
12
Dec
12
12
2006
07:20 PM
7
07
20
PM
PDT}

bFast Mibad. You're right of course. In my haste to find what the Howard Hughs Medical Institute writeup was talking about in the exploding quote I failed to count the uncolored sequences. I'm not convinced the quote is a mistake yet. Bejerano links to HHMI article I quoted in his own UCSC webpage, by the way. http://www.soe.ucsc.edu/%7Ejill/ (see Living Fossil Transposon) which kind of bolsters the notion that HHMI made no mistake. DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
07:09 PM
7
07
09
PM
PDT}

I have notice another intriguing thing in the article (the link in post #22, page #13) It says:
all functions we understand in our genome are encoded using redundant codes.
(emphasis in original) On ISCS's brainstorms, Peter Borger extensively discusses that natural selection is not capable of maintaining redundant code. Further we have had a major debate on this forum -- something about magnito systems in light aircraft -- where we reach the same conclusion. I challenge the anti-ID community to even make a decent simulation that shows a redundant system that doesn't loose its redundancy under influence of random mutation.bFast_{December 12, 2006
December
12
Dec
12
12
2006
04:20 PM
4
04
20
PM
PDT}

DaveScot #27 chimp/human 54 matches For chimp/human I count 80 matches. I count 54 that are colored, and an additional 26 which are not.bFast_{December 12, 2006
December
12
Dec
12
12
2006
04:11 PM
4
04
11
PM
PDT}

Geomor The coelacanth sequence is the only one in black because it's the template against which all the others are compared. That's because it is the oldest phenotypically unchanged lineage in the group by the testimony of the fossil record. It's odd enough that the SINE protein product (PCBP2) in frogs, birds, and mammals are identical and equidistant from the coelacanth. Ultra-conserved protein is one problem to explain. The second problem is that the SINE regulatory region (ISL1) doesn't follow the pattern of the protein and surprisingly the human/chimp ISL1 resembles the coalencanth predecessor moreso than it does to critters with a closer common ancestor.DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
01:21 PM
1
01
21
PM
PDT}

Interesting coelacanth trivia from wiki
Coelacanths give birth to live young called pups. Their reproductive behaviors are not well known, but it is believed that they are not sexually mature until after 20 years of age. Gestation time is 13 months, females give birth to between 5 and 25 babies, which are capable of surviving on their own immediately after birth.
The list correlates better with generation (sexual maturation) time in all the species rather than divergence time. I wasn't aware before just now that coelacanths took so long to mature and had so few young. They're the slowest to reach sexual maturity in the whole list and live by some estimates at least up to 80 years (they were only known as fossils until 1938).DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
01:00 PM
1
01
00
PM
PDT}

I know little about this subject but I find it very interesting. What do you all make of this article? - "Evolution of 'irreducible complexity' explained" This part stood out to me:
The experiments showed that the receptor had the capacity to respond to aldosterone long before the hormone actually evolved. Thornton's group then showed that the ancestral receptor also responded to a far more ancient hormone with a similar structure; this made it "preadapated" to be recruited into a new functional partnership when aldosterone later evolved.
Sounds like a telic process to me.Lurker_{December 12, 2006
December
12
Dec
12
12
2006
12:52 PM
12
12
52
PM
PDT}

By my count of ISL1 comparisons: human/coelacanth 54 matches chimp/coelacanth 54 matches dog/coelacanth 51matches mouse/coelacanth 50 matches rat/coelacanth 50 matches matches opossum/coelacanth 51 matches chicken/coelacanth 54 matches frog/coelacanth 45 matches chimp/human 54 matches dog/human 49 matches mouse/human 46 matches rat/human/rat 47 matches opossum/human 48 matches chicken/human 50 matches frog/human 40 matches The human ISL1 is closer to a coelacanth than it is to any of the other species listed except for chimps and they're identical there. This makes little sense in light of evolution. Human/coelacanth should have the fewest matching bits not the most. Humans and chimps are closer to a coelacanth with 54 matches than they are to dogs, mice, rats, possums, chickens, or frogs which is 50 matches at best and the best case is a chicken of all things instead of another mammal. Otherwise it looks reasonable except perhaps that frogs shouldn't be such an outlier from coelacanths compared to mammals and birds since they've presumably been separated from a common ancestor the same length of time as any of the mammal/bird lineages but frogs probably just mutate a lot faster than the others which explains that. DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
12:35 PM
12
12
35
PM
PDT}

Thanks bFast for actually counting. It's pretty clear just by glancing at that figure that the coelacanth sequence is the most distant. It is purposely drawn to highlight the differences between the very closely related mammalian sequences and the coelacanth sequence. Jehu, I don't know exactly how the horizontal transfer would work. Now that I'm going back and reading comments again, I see that I may have misinterpreted what you were saying and thought that you were thinking "fluke" referred to a fish swimming around in the ocean, in which case horizontal gene transfer with a land mammal would be pretty spectacular. Sorry if I misunderstood that. At least relatively speaking, horizontal transfer between a blood parasite and its host is easy to see =) One thing that might make it a little easier to envision is that it could be transfer from the host to the parasite, rather than the other way around. One also has to wonder if, when they took a sample of the blood fluke for sequencing, they took it from some human blood sample and there could have been human genomic contamination. Anyway, I don't know anything about this guy, so that was all pure speculation.GeoMor_{December 12, 2006
December
12
Dec
12
12
2006
07:03 AM
7
07
03
AM
PDT}

Sorry Dave, I don't see what you are talking about. When I count the matches between the human and the coelacanth, I get 54 matches. Between the dog and the coelacanth, I get 52. Between human and dog, I get 68. (It's kinda rough countin' but if we use the same "deletion" logic they use, that's what I get.) Therefore the ISL1 is more similar between the human and the dog then between the human and the coelacanth.bFast_{December 12, 2006
December
12
Dec
12
12
2006
06:21 AM
6
06
21
AM
PDT}

Some pre-Socratic Greek philosophers suggested humans were evolved from fish. So they must be feeling pretty smart right now.geoffrobinson_{December 12, 2006
December
12
Dec
12
12
2006
04:22 AM
4
04
22
AM
PDT}

geomor and reed I don't know what is being referred to in the Nature article as being 100% identical among mammals but it isn't the ISL1 regulatory region. That is identical between humans and chimps then the next closest match is to the coelacanth. Language might be imprecise but the sequence data isn't. Refer to page 33: http://bejerano.stanford.edu/talks/BejeranoGoogleOct06.pdf and compare the sequences for yourself.DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
04:17 AM
4
04
17
AM
PDT}

It's not misleading. http://bejerano.stanford.edu/talks/BejeranoGoogleOct06.pdf Go to page 33 where the ISL1 regulator is graphically compared between human, chimp, dog, mouse, rat, oppossum, chicken, frog, and coelacanth. The human and chimp versions are identical and most closely match the coelacanth. For a laugh, go to page 7 to see Bejerano make the gratuitous, obligatory "secret handshake" to assure readers he's not "one of them". This secret handshake is prerequisite to publication in respected journals. :razz:DaveScot_{December 12, 2006
December
12
Dec
12
12
2006
03:57 AM
3
03
57
AM
PDT}

Well in that case the following statement is a bit misleading.
When the researchers compared the human ultraconserved element to all the DNA sequences in the public database GenBank, the closest match was to DNA from the coelacanthÃ¢â‚¬â€an ancient fish thought to have gone extinct millions of years ago until a live specimen was caught in 1938 off the east coast of South Africa.
As for horizontal transfer of genetic information; no it is not clear to me how the DNA of a blood fluke gets into the line of its host. It doesn't exactly inject its DNA into the cells of its host does it?Jehu_{December 12, 2006
December
12
Dec
12
12
2006
12:55 AM
12
12
55
AM
PDT}

Thank to GeoMor for clarifying the significance of the article. I think that DaveScot understood what the article was about when he posted this, but I'm still not clear on what this means: "Our closest relative on the tree of life according to ultra-conserved DNA is a fish thatÃ¢â‚¬â„¢s been around unchanged for at least 360 million years." It sounds to me like the article was saying that the sequence in question has deteriorated similarly in coelecanths and humans.Reed Orak_{December 11, 2006
December
12
Dec
11
11
2006
11:42 PM
11
11
42
PM
PDT}

This is the third time I'm trying to post this. I'm not sure if it's being intentionally held back, or what, but perhaps someone will recognize that it is a rather important clarification to make. The ultraconserved element in question is found almost exactly in all the mammalian genomes we have, and also very close in an amphibian (frog), while the fish sequence is less similar -- basically consistent with the "tree of life". This can be verified by reading the actual paper that this news article is about (Bejerano et al. 2006); unfortunately I don't think it's open access, but at least the key misunderstanding here can be resolved from the abstract: One of these, a more than 200-base-pair ultraconserved region, 100% identical in mammals, and 80% identical to the coelacanth SINE,... http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=16625209&dopt=Citation We don't yet have much genome sequence from other species in between (besides the frog) to see what went on, but we would strongly predict that the sequence will be found in some form in such genomes. Something very interesting indeed had to happen for the transposable element to come under selection in a particular location, and that's why this is in Nature. Jehu, in regards to horizontal transfer. Please note that DaveScot has referred to two separate studies here; the term "fluke" in the second (Xie et al.) does not refer to the fish of the first (Bejerano et al.), but rather to the mammalian blood parasite Schistosoma japonicum, colloquially known as "blood fluke". It should be clear why it is at least plausible that horizontal gene transfer between a blood parasite and its host's genome could occur. In summary, this blog entry and most of the comments seem to be based on a misunderstanding of what the study was actually saying, although I can see how it happened from the way the HHMI news article was written. There is no inconsistency with NDE or tree-of-life in either the Bejerano or Xie studies.GeoMor_{December 11, 2006
December
12
Dec
11
11
2006
10:58 PM
10
10
58
PM
PDT}

DonaldM, this finding is exactly what evolutionary theory predicted! There you got it.Srdjan_{December 11, 2006
December
12
Dec
11
11
2006
08:59 PM
8
08
59
PM
PDT}

I wonder how long it will be before the Darwinists will claim that this finding is exactly what evolutionary theory would have predicted!!DonaldM_{December 11, 2006
December
12
Dec
11
11
2006
06:48 PM
6
06
48
PM
PDT}

Is it just me, or has there been a _lot_ of these kind of papers published in the last year? Is this an indication of a new trend in biology?johnnyb_{December 11, 2006
December
12
Dec
11
11
2006
06:14 PM
6
06
14
PM
PDT}

"Comparing the sequence to other species also turned up a big surprise." I've noted it before: Darwinists are regularly surprised by real life findings. (I THOUGHT that fish looked an awful lot like my aunt Jean...)dacook_{December 11, 2006
December
12
Dec
11
11
2006
06:06 PM
6
06
06
PM
PDT}

I think we need to know more about epigenetics and other possible routes of inheritance before we go drawing any conclusions about a tree of life. I think the gene-centrism required by neo-Darwinism has lead the science astray on this.Inquisitive Brain_{December 11, 2006
December
12
Dec
11
11
2006
05:16 PM
5
05
16
PM
PDT}

jpark320 // Dec 11th 2006 at 4:56 pm "Sorry for the grammar, but gotta catch a flight soon!" Don't forget: 3.4 ounces or less containers of liquids/gels/paste, in quart-sized plastic bags, removed from luggage prior to screening. ;)russ_{December 11, 2006
December
12
Dec
11
11
2006
03:00 PM
3
03
00
PM
PDT}

Jehu, it looks to me like if you follow the phylogenic tree painted by one gene you get one branching, if you follow the tree painted by another three you get another branching. This is a much worse problem than mere "no mapping clarifies which connection is correct", rather it is, "this evidence declares this connection, this other evidence declares a different connection." This should be a problem. 1 + 1 = 43bFast_{December 11, 2006
December
12
Dec
11
11
2006
02:42 PM
2
02
42
PM
PDT}

bFast
Patrick, when I look at a bush, it forks just like a tree. It seems, however, that you and your article are not talking about a Ã¢â‚¬Å“very busyÃ¢â‚¬Â tree, but a tree-like structure that doesnÃ¢â‚¬â„¢t simply fork as you climb up from the roots to the leaves. Rather it seems to show a tangled mess that resists a fork-based mapping.
I think of it as branches hanging in thin air, not connected to anything but with certain "stastical probabilities" that make theoretical connections to the other branches. Kind of like the fossil record.Jehu_{December 11, 2006
December
12
Dec
11
11
2006
02:30 PM
2
02
30
PM
PDT}

Prev 1 2 3 Next

You must be logged in to post a comment.

Leave a Reply