We have been gibbering for many, many pages, at least nominally on the topic ‘Common Design vs Common Descent’. I’d like here to discuss an interesting fact I discovered during the course of this. I mentioned it twice, but readers seemed underwhelmed by what, to me, looks like a genuine scientific discovery (I haven’t searched exhaustively for priority). More importantly for present purposes, it provides a useful test-bed for several concepts that regularly do the rounds.
Sal Cordova had latterly been pursuing an attempt to demonstrate using freely available phylogenetic tools that tetrapods are not descended from aquatic, gilled vertebrates – what we conventionally call ‘fish’. In the process, Sal chose to use gene sequences of what turned out to be cytochrome oxidase (COX) – a component of the electron transport chain in mitochondria and in prokaryote plasma membranes. This molecule is ultimately what that breathing thing is all about – the molecular oxygen we inhale, in diatomic clumps, is held together by electrons shared between the atoms; COX donates electrons passed along from cytochrome c which enable the oxygen atoms to be separated; these react with protons (a different reaction for each of the two atoms) to form water. In the process, additional protons move across the membrane to generate an energetic imbalance that can be tapped, much like a pumped-storage hydro-electric scheme, to make ATP.
Sal thought he was looking at cytochrome c – this molecule is actually one step ‘upstream’ from COX, supplying it with the electrons which the latter passes on to oxygen. In eukaryotes, it so happens that the cytochrome c gene is in nuclear DNA, while COX is in mitochondrial DNA. This creates a significantly different evolutionary environment for the two genes, due to the proximity to reactive molecular species in the latter and the different dynamics resulting from recombination in the former. It would be unwise to draw too many evolutionary conclusions without an awareness of this distinction.
During the course of this discussion, Rumraket took some sequences of the actual cytochrome c, the molecule Sal thought he was looking at, and ran a Uniprot process to generate a phylogeny for a set of sequences. In an idle moment, I thought it might be interesting to explore the taxonomic neighbourhood a bit. It seems to me fundamentally inconsistent to dismiss phylogenetic inference on the grand scale while accepting it more locally. Of course it’s not clear whether any Creationists accept it locally either! I’ve tried repeatedly to get someone to say at what taxonomic level they think Common Descent stops being a valid inference; no luck so far.
So here’s what I did. I was interested to see what the cytochrome c’s of close relatives of the salmon looked like, rather than the grand-scheme analysis being pursued by the ‘I-ain’t-no-fish’ brigade. So I took Rumraket’s sequence for the Atlantic Salmon Salmo salar. The BLAST came up with several paralogs – gene duplicates – in the salmon, and orthologs – genes in separate species – with >90% sequence identity. I say this a lot, but the only process I am aware of that leads to sequence identities in anything like that ballpark is nucleic acid polymerisation. This mostly happens during cell replication (DNA-DNA) and gene transcription (DNA-RNA), somewhat less often by reverse transcription from RNA to DNA. So, absent another process that generates such extensive identity, common descent (at sequence level) has to be a front runner for the reason for any such highly alignable sequences.
So, here are the top few hits for Salmon cytochrome c sequence B5XFR7 (link here but be quick; I don’t think these are archived for long):
Rainbow Trout (Onchorhynchus mykiss) C1BFD3 99%
Salmon paralog B5DFW1 99%
Salmon paralog B9EMZ7 98.1%
Rainbow Trout paralog C1BGL1 98.1%
Rainbow Trout paralog C1BG90 95.2%
Salmon paralog B9EMJ0 94.2%
Rainbow Trout paralog C1BFB8 94.2%
Large Yellow Croaker (Larimichtys crocea) 94.2%
Sea louse (Caligus rogercresseyi) C1BPA2 93.3%
Rainbow smelt (Osmerus mordax) C1BKE6 93.3%
… ?
Whoa, back up a bit … the sea louse? That’s a crustacean. Now, I look up this organism and, it turns out, this is a parasite of both Rainbow Trout and Atlantic Salmon, particularly in Chile (yes, they farm Atlantic salmon in the Pacific). So, pending more work, a viable explanation for the high sequence identity with Salmon and Rainbow Trout is gene transfer from one or the other (hard to tell which). If not HGT, it was a remarkable coincidence that the words ‘salmon’ and ‘trout’ should leap off the page when I looked up the organism.
A naive user of free phylogenetic software might try and argue that this anomaly completely destroys the use of such software. After all, a phylogeny based only on cytochrome c might perch this crustacean slap dab in plaice in the middle of the Salmonids (it could also be used to show that salmon are more closely related to trout than they are to their own genes, in clumsy hands …). This is obviously a bit fishy, because we know (ask yourself: how do we know?) that it doesn’t belong there.
So, here are a few questions for Creationists to consider.
1) Do you accept that high sequence identity is reasonably inferred to indicate an origin in a single ancestral DNA sequence?
2) If so, do you agree that the paralogs – multiple copies in the same species – likely share common ancestry with an original single sequence?
3) How about within the family Salmonidae, to which both Atlantic Salmon and Rainbow trout belong (the trout is actually in the same genus as the various Pacific salmon)? Do you think sequence commonality highly likely indicates a genetic relationship here too?
4) Do you agree that the likeliest cause of high sequence identity in the sea louse version is due to gene transfer from salmon or trout, given its intimate association with these species?
5) What makes us think that the sequence in sea lice is anomalous, rather than indicating an actual whole-genome pattern, without even looking at the genome?
6) For broader genomic comparisons, collecting together multiple gene trees, where in the standard taxonomy would you place the division between the region that is reasonably inferred to result from common descent, and that which is not?
Another issue relates to function, and the need or otherwise for functional variation among species or among paralogs. As discussed, the task of cytochrome c and COX is chemically quite low-level – to pass electrons on down a chain. On a design paradigm, the fact that this function needs to vary at all is rather curious, unless we allow that variation can occur through largely neutral mutation. Even stranger is the fact that variation within a single species among its various paralogs is of the order of that between some orthologs in separate genera of the Salmonidae. If one answered ‘yes’ to 2 but ‘no’ to 3, one is saying that the same amount of sequence identity is diagnostic of sequence relationship + mutation in one case, but not the other. One might appeal to evolutionary unnattainability of a transition, but it doesn’t seem to me that the gap between salmon and trout is unbridgeable, starting from a common ancestor. After all, as noted, Pacific ‘salmon’ are more closely related to trout than they are to Atlantic Salmon (I’m curious no other Onchorhynchus species came up, but this may be due to lack of sequences). So, it seems plausible to me that all these gene variants are commonly descended, through fully connected paths of nucleic acid polymerisation.
This pattern of otherwise inexplicable variation of non-morphological genes is repeated throughout the genome and throughout nature. ‘Common-Designists’ would have it that there’s a Psilocybe way of dehydrogenating lactate, and a yeast way, and a bottlenosed dolphin way. And there’s a dandelion way of passing electrons on, and a squirrel way, and an E coli way, and a sea louse way … but hang on; the sea louse way is pretty close to the salmon way, in this gene copy at least. If this is an HGT event, it indicates that there is much less species-specificity in function than some would argue. If HGT, this is a natural experiment, placing a variant in a distant relative and seeing how it fares. Assuming it’s not a pseudogene, the sea louse is apparently able to use the cytochrome c of a very distant relative in addition to its existing repertoire, with no obvious adverse consequences. More work would establish which of its isoforms tend to be preferred, but nonetheless the bare fact rather goes against the expectation that these proteins are highly clade-specific for design reasons. And indeed it does seem much more likely that low-level function can be passed around without excessive penalty than more taxon-specific genes – the variants are mutational fluff, not critically functional. This is one reason why Amino-acyl synthetase phylogenies are actually quite poor – extensive HGT is possible, because really, how many different ways do you need to stick an amino acid on the ACC- end of a tRNA?
.. or it’s contamination, when the sea louse was being sequenced, with DNA from its host organism?
Joe Felsenstein,
Yes, I did consider that, but assumed that people preparing these samples were pretty careful. I don’t know, of course, and guess it would not be too hard to find out – one would expect 100% identity of the sea louse gene with a modern salmon or trout gene (not the one I started with). I’ll run that when I have time.
That might be a bad assumption. There’s quite a lot of contamination in genetic databases. Host DNA, stomach contents DNA, pathogen DNA, cloning vector DNA, lab tech DNA, lab tech’s lunch DNA: it’s all been found at one point or another. On the other hand, confirmed cases of horiztontal transfer between eukaryotes are extremely rare. That sea louse sequence would really have to be confirmed by fresh lab work.
OK, I ran the query (annoyingly, I had to type out the sequence, as the alignment page appears not to allow Ctrl+C).
The sea louse gene is 99% identical to the Salmo paralog B9EMJ0. Which is consistent with the distances of that paralog and the sea louse gene from my starting sequence. That argues against contamination, and in favour of an HGT event from that paralog in Salmo subsequent to Salmo/Oncorhynchus divergence.
Of course, whether it’s HGT or an artefact is secondary to the point that one can just look at a gene, any gene, and kinda know if it doesn’t belong there, pretty much irrespective of its function. That knowledge is encapsulated in the taxonomic tree, which the molecular trees recover a remarkable fraction of the time, and which informs our identification of anomalies.
Try inspecting the source code (ctrl+u in Chrome)
From OP:
“After all, a phylogeny based only on cytochrome c might perch this crustacean slap dab in plaice in the middle of the Salmonids…
Subliminal pun?
Allan Miller,
The task of cytochrome c was more then passing an electron down a chain as I recall?
Acartia,
No, it’s an extended pun. Dab and perch are fish, too.
@ Allan
How might you account for the Salmon sequence getting to the sea louse germ-line? Gene, gut, gamete?
ETA Great pun, in any case. 😁
No because life hasn’t been on Earth long enough.
Since I don’t accept premise 1, question 2 is N/A.
Don’t know, but if takes longer than a few million years to create such a relationship, I don’t think common ancestry is the explanation.
Re-sequence, confirm it’s not contamination. Same for genes humans only share with chickens or genes humans only share with zebrafish. If not HGT, then God did likely created it that way. I heard of recombinant DNA and that’s how they made the some of the first genetically engineered corn (the BT corn). HGT in Eukaryotes would seem tougher than in prokaryotes.
It falls outside the nested hierarchy that is usually seen. The same could be said of genes we share only with chickens and zebrafish. I heard of one that we supposedly share only with spiders, but haven’t been able to nail down that citation.
Those where the sequence divergence would take longer than 10,000 years to achieve would be created differences rather than mutation acting on common descent of the gene.
Saying there is no adverse effect could just as well be testament to the robustness and redundancy of the design, not that the sequences or genes have no function. This was confirmed by experiments such as those by Brenda Andrews reported at the ENCODE 2015 planning conference (not the same as the ENCODE 2015 users conference). See:
https://www.genome.gov/27560819/from-genome-function-to-biomedical-insight–encode-and-beyond/
One non-morphological protein is Topoisomerase. I was in the lab of one researcher who has studied Topoisomerases his whole professional life. Pasted on his wall was the mass spec evidence of phosphorylation on multiple serine residues.
Differing sequences may have differing functional phosphorylation patterns. Same for acetylation, methylation, or any other “-ation” including glycol conjugation. There is a lot we don’t know.
You’re premise that it is not functional in a species specific way could be deeply flawed, and it is only presumed based on a few mutations not having obvious immediate effect, but Brenda Andrews experiments shows that naïve view is flawed. I pointed out the problems of such naive viewpoints 12 years ago, and publications since have confirmed my argument:
Genes and pseudo genes and paralogs can have different roles. The pseudo gene transcripts though not translated have regulatory function in the transcriptome sometimes functioning as sponges in miRNA interference pathways. Who knows what else they do. Fuz Rana discusses one gene that is functional in humans but is pseudo in rats, but the pseudo in rats if part of a CE RNA regulatory network….God apparently has a sense of humor….
An electrical engineer looking at the miRNA regulatory networks would immediately recognize feed back control loops. The pseudo gene RNA transcripts play a part, not to mention the DNA segments can be regulatory binding sites for all sorts of molecular machines.
Synonymous sites that are sequence different on paralogous genes could also serve the function of being enhancer sites to regulate other genes in Cis or even in Trans on the chromosomes. Recall my discussion of enhancers on exons? Extra paralogs also serve as ways to amplify expression and provide redundancy. Even in prokaryotes under environmental stress, one strategy of expression amplification is gene duplication….
Now that we are studying the 4D nucleome, some of the positioning of gene duplications might also make sense. The only people who don’t seem excited to see these developments are people who prefer to view biological complexity as clumsy and poorly put together. I suppose one could view a Rube Goldberg machine as the product of ingenuity or mindless stupidity. Depends on ones perspective.
It doesn’t matter what it does (or if it has one, zero, or multiple functions). What matters is the branching order of the tree it yields on a phylogenetic analysis. There’d be no reason for it to correlate with trees drawn from other data sets, (like other cytochromes, or tRNAs or rRNAs or a host of other core metabolic genes, or comparative morphology etc. etc.) if common descent isn’t true.
Why you might ask? Because cytochrome c doesn’t cause those other traits. Cytochrome c doesn’t make a spine appear in a fungus, or a bacterium (or eel, or spider, or flower) grow four limbs, or what have you. It doesn’t cause ribosomal RNAs or proteins to have a certain sequence (nor is cytochrome c caused to have the sequence it does by those factors).
So there you go, that is why you blathering about cytochrome c potentially having a role in apoptosis is completely irrelevant.
This is pretty much assuming your conclusion against the evidence before you.
Nope. I was an old fossil record believer until I studied the evidence starting with the Faint Young Sun Pardox. The Faint Young Sun requires miraculous fine tuning to make the Cambrian explosion happen. It’s possible there was no fine tuning and no Cambrian era.
So the irony is, like so many things, evolutionary theory depends on miracles to make it happen.
See:
https://en.wikipedia.org/wiki/Faint_young_Sun_paradox
There are more anomalies which I’ve discussed before. You all aren’t interested in considering them because you’re so sure you are right.
Isn’t it obvious that the answer to the first question cannot be ‘yes’? It can be at best ‘maybe’.
Question: Do these two identical soccer balls have a common ancestor?
Answer: ???
“Sequence identity” is insufficient to tell anything about the origin of things, if we ignore what kind of things we are talking about. Overall, the natural environment and beings in it share the same elements, i.e. they have sequence identity, but do you accept it points to the same origin, i.e. creator? I guess not. And they are not the same thing either, so sequence identity by itself means nothing.
Rumraket,
A single mutation in the mouse version can cause its forehead not to develop.
So God created a planet and a star. The planet froze over because the star He created did not put out enough energy so God miraculously unfroze the planet. Sounds inefficient
So it was fine tuned to require a miracle. Interesting so the evidence that led you to young fossils may or not be true, but if it is true it required a miracle and if it is not true that also requires a miracle.
Yes you are.
Yes yes. And the entire scientific community used to believe in young Earth creationism until they … did science on the Earth and the universe.
I couldn’t find this conclusion anywhere in the wiki article. There was something about tidal, climactic and ratioactive heating though. But I’m sure that’s all just false because God.
It can also cause the electron transport chain to fail and kill the cell. That still doesn’t say why it yields a phylogeny that is highly congruent with all the others. Only common descent does.
As has been explained to you many many times before, the argument isn’t based on mere sequence identity.
The sequences are copied every cell division, and mutations accumulate in them. This directly predicts slow divergence and branching paterns.
colewd:
Have you had your own genome sequenced, by any chance?
Rumraket,
An assertion. Why would common design be excluded from this pattern?
How–and why–would common design cause it, except to mimic common descent?
Glen Davidson
This doesn’t even make sense. Common design doesn’t explain why the trees are congruent with each other.
Rumraket,
Please explain why this could not be a product of the design strategy.
Acartia,
A whole netfull of ’em, and deliberate. Plaice was the flag!
colewd,
Apoptosis is a pretty low-level function too. You surely don’t think the final form of the organism is coded in the cytochrome c isoform? There are several in the salmon alone. Butt my argument applies equally to cytochome c oxidase, and any number of other molecules. You have to find a dual function for them all, and it has to be species-specific and sequence dependent. I think you’d have your work cut out, and waving your arms in the general direction of apoptosis doesn’t cut it.
stcordova,
Long enough for what? You mean that no 2 DNA sequences with high sequence identity can ever be caused by common descent from a single DNA molecule? In a PCR assay, in parent and child, in forensic analysis, in two members of subspecies? I know you don’t mean that, but you might like to give the question a little more thought.
Allan Miller,
Nice attempt at a burden shift. You made a statement both of us knew was false. Your basic premise is false and I think your argument needs a restart.
The truth is none of us know at this point all the functions of these proteins. This is a statement from ignorance.
Interesting claim. I would love to see you try to support it.
stcordova,
This is special pleading, and your resort to analogy a bit desperate.
I’m not saying for definite that it is HGT, but if it is, this does not support a compelling need to make them all different – and certainly not to make them all different in a way that maps pretty well on the morphological tree. Bear in mind you have to do this for every nook and cranny of the genome. Even those SINE inserts you won’t talk about – a SINE held in common at a particular location in pigs, deer and cetaceans still accumulates mutations, as far as anyone can tell, which pattern follows the morphological trees too, and they are not exquisitely variant for any apparent functional reason, even though you might be able to write lengthy essays on A-to-I editing, or hear the word ‘transposon’ and trigger a brain dump.
It’s as if, for a Creationist, mutations never happen, and nothing ever varies ‘by itself’ as a result.
colewd,
Get stuffed. At the outset I talked of cytochrome c and cytochrome c oxidase. Two different moelcules, only one is involved in apoptosis AFAIK, and apoptosis is just as ‘low-level’ a function as electron transport. Also completely irrelevant outside animals, the only organisms Creationists seem to think exist.
Special pleading. And the genomes of organisms consist of far more than proteins. If you think that every single base – even deletion; the absence of bases – has to be like that with no possibility of variation because of a functional reason, you are at odds with the entire field.
Fundamentally, a cell either dies or it doesn’t. This isn’t something that appears to need exquisite control at the level of the cytochrome c isoform (something that might conflict with electron transport requirements anyway), and you have done knob all to support the claim that it does.
colewd,
Evolution occurs through mutations that are not fatal. Unless you think that all possible mutations are fatal, in which case we have the curious case of the functional isoforms that differ by just one amino acid.
Which strategy is that also how was it implemented might be relevant to whether it is possible?
Erik,
The only mechanism known to produce such extensive sequence identity in nature is nucleic acid polymerisation. I know you hope for another, but I’ve seen no good reason to drop descent as a default assumption – especially when this process, coupled with its occasional imperfection, leads to an expectation of cladistic patterns, which are met by real data. There is no apparent boundary where common descent ceases to apply, and common design takes over. I’d have thought there would be one – I can’t see a reason for common design to give the same pattern as common descent, poor analogies nothwithstanding.
The analogy put the emphasis on sequence identity. If this is all we are talking about, then we are failing to distinguish common descent from common design and your evolution is just as good as creation. Please let’s get rational and scientific.
I heard that species reproduce themselves. Is this superseded now? What on earth justifies this ludicrous focus on DNA sequence identity, as if that were the thing that told everything about species we need to know?
Erik,
Haha, you’re a card. An observed process is just as good as an unobserved process to account for something. Jolly good, you top scientist, you.
DNA is the medium of inheritance. You know, genetics. That’s how it happens, by template based DNA polymerisation, which happens also to be the only mechanism known to produce extensive sequence commonality. So, if I observe sequence commonality (and begging Your Grace’s pardon, it is important in biology), I’d like to know how it got there. Descent from a single original DNA molecule is a perfectly respectable hypothesis, since it actually happens. Common Design always seems a bit cargo-cult to me, particularly when we are talking of DNA sequences – maybe that’s why Creationists seem unwilling to consider ’em.
It doesn’t argue very strongly. It argues against contamination from any of the salmon sequences in the database, but it could be either an unknown salmon sequence or a sequence from a close relative of the salmon. I assume this louse is not limited to parasitizing a single species of fish.
How much sequence divergence should you see in 10,000 years? Are you sure you can fit all those species on the ark?
If there was no Cambrian era, why are the strata sorted so nicely, so that Cambrian fossils occur only in Cambrian strata, above the Late Precambrian and below the rest of the Phanerozoic? And why are all the other stratigraphic layers in a nice order too?
You mean the process of producing footballs is not observed? The analogy was somehow obscure to you?
Creationists (of certain stripes) also believe in creatio ex nihilo. That’s the company you are in.
But I’m not in that company. Material things don’t self-replicate. They are replicated from pre-existing material. This is how the entire physical world works. Living beings require a special explanation, likely non-physical.
John Harshman,
For sure, it could easily be wrong. I did hedge in the OP, allowing the possibility that it’s not HGT or, if it is, it may be a pseudogene. Nonetheless, circumstantially I would expect to find 100% match in existing data if it’s contamination, since the sea louse is an economic pest in economically significant species. Data would be skewed towards sequences from economic species, and farmed fish are the likely source of the sea louse materials.
But yes, by no means conclusive and I’ve reached the limits of tools within reach of my armchair; I’d like to see broader sampling of both the the sea lice and the salmonids.
Allan Miller,
You are making assertions here based on ignorance to cover up that you made a statement you knew was wrong. This is not like you so I am really wondering what is going on. Apoptosis is one of the most complex functions that we have in the cell. I suggest you bone up on it. The fact that a single mutation in an apoptosis protein can create a fetal mutation is very telling.
No, I am saying the assuming proteins have single function is not a credible position because discovery is proving this wrong as in the case of cytochrome c.
You need further study here. Apoptosis is a complex process which is part of the cell cycle. When and how is apoptosis triggered? Do you think internally killing a cell is trivial. Why?
Erik,
What, you’re still talking about the analogy? The real thing somehow beyond you? The process of producing identical sequences by other than DNA polymerisation has not been oberved.
I mean template based DNA polymerisation actually happens, not the descent of all life from an ‘ur’ sequence. You Creationists scoot straight to ‘universal’ whenever you see a discussion on common descent. I was talking much more locally: paralogs in salmon, or orthologs in salmon, trout and (possibly) sea lice.
Footballs are not real? And they are not identical enough?
Yes, because adaptation is trivial and not under dispute. Unless we discuss universal common descent, there is nothing at all to argue about, because you are not talking about evolution in the sense that would go contrary to creation.
colewd,
I’ll tell you what’s going on: you are misunderstanding or misinterpreting my motives. I do not consider the statement I made to be wrong, so I take exception to the accusation of revisionism. You’re right, I would not make a habit of denying something that was a short scroll away.
And a single mutation in cytochrome c can be fatal to the organism even if it doesn’t have apoptosis. I might haughtily suggest you bone up on that too. You misunderstand what I mean by ‘low-level’. I don’t mean unimportant; I mean it does not sit at the top of a control chain that is part of the variation in morphology between species. Apoptosis is induced when cytochrome c changes its bound state; this is quite coarse chemistry which does not need to vary exquisitely in every organism – the squirrel way of becoming unbound, and the dandelion way …
It doesn’t matter how many functions they have. The question is why the variation in this or any other molecule must follow the morphological tree.
You are obsessed with this one molecule. I can simply deflect it by talking of something else. Do you propose a universal rule applying to every base in every organism based upon your understanding of cytochrome c and what can only be a secondary function in a particular clade of organisms?
Let’s talk of an as yet unspecified transposon in an as yet unspecified clade of plants. Why does its variation follow the morphological tree, and that of every other gene? Is ‘we don’t know everything about it yet’ a better reason than common descent?
Erik,
Do you find obtuseness a useful substitute for sensible discussion? Footballs are not DNA molecules. They don’t undergo template based polymerisation. I’ve noticed this before with you; an obsession with the analogy, when the real thing is there to be discussed whenever you are mature enough to have a go.
Since my OP was about a gene in salmon and rainbow trout, does this fall within your definition of ”universal’ or not?
Do they not have sequence identity? Or identity all the way? The first question assumes that when the makeup of two things is the same, it reasonably indicates a common origin. Bad claim, equally bad analogy. That’s about it.
Funny how in your system we can make up definitions as we go along, as if there were no standard in place already. In my view, this is very much a problem.
Allan Miller,
I think this is inconsistent with a small variation in it producing a mouse with no forehead. This means that what is apoptosis at one stage is embryo development in another. Strong evidence that it could be part of the variation in morphology between species.
I accept you were acting in good faith.
Given the molecule is involved in embryo development there is no reason it shouldn’t. The changes in this case are very likely to be mutationally sensitive so this supports a designed sequence vs random changes.
You keep bringing this molecule up 🙂 I don’t think every genetic variation we see is due to design. Common descent plays a role. Your challenge is to show that no genetic variation that exists is due to common design. Thats keiths position and I wish him the best of luck.
Erik,
They aren’t DNA molecules, so I’m not really interested in them for present purposes. DNA molecules have sequence identity for a very specific reason – the DNA double strand is unwound and a complementary sequence synthesised upon each strand. The result is two molecules from one, each commonly descended from the first and (barring mutation) identical to it and to each other. This is the only process known to produce sequence identity (I know, I’m like a broken record), so I wouldn’t be in any rush to dismiss it.
That’s an oddly evasive way of answering a very simple question. I’m simply trying to find where you think ‘universal’ common descent starts. I know where I and most biologists would place their marker.
Let’s try another approach; do you concede any possibility that a gene in salmon which shares 99% sequence identity with one in rainbow trout originated in a single DNA molecule in a common ancestor?
Allan
This is facinating and timely.
Check out
http://sandwalk.blogspot.ca/2017/11/lateral-gene-transfer-in-eukaryotes.html?m=1
colewd,
That is not evidence that the actual variants in species contribute to the morphological differences between them. It’s trivial that some mutations cause dramatic effects on development; this does not mean they are involved in development but that they can initiate a causal chain which interferes with it.
You haven’t actually shown that it’s involved in embryo development. Developmental apoptosis and cell damage apoptosis are likely to involve separate pathways, for one thing.
But anyway, this is a rabbit hole. Unless every organism on earth does apoptosis, and every gene in every organism is involved in the final form of the organism, it’s a monumental red herring.
I certainly don’t have to prove the negative.
Your challenge, a much less preposterous one, is to show why the gene trees of genes not invoved in morphology would follow the morphological tree. You surely would not assert that there are no such genes (bearing in mind that a gene in this context is any stretch of DNA of interest).