Taking a new tack with the common descent/common design theme, I’d like to sneak up on it. As Sal Cordova likes to do, I offer a hypothesis for critique. Here’s an old phyogenetic analysis of mine. Does it show common descent? If not, what’s wrong with it?
This one is about crocodiles. Are crocodiles all the same kind? How do you know? If they aren’t how many kinds are there within the group and how do you know? If they are, are they a whole kind or just part of a larger kind? And if the latter, what is the larger kind?
J-Mac,
Any variation. If you have 5000 buckets’ worth of ‘varied things’ and only four buckets to put them in, you can’t keep all the variation that you had in the 5000 things. Where you gonna put it?
Ok John, from Table 1 in your paper I put in Accession number AY277489 for “Tomistoma Shlegelii c-myc (c-Myc) gene, exon 3 and partial cds”. I guess the title confirmed my claim this was a fragment that didn’t have the 1st exon. Is it correct exon 1 will usually have the methionine codon “ATG”?
But, the back door way I attempted by putting in the coded-amino acid sequence from the AY277489 entry:
EEQEEDEEEIDVVTLVEVNGVESGTESSTDSSGESSTDSTDGHS KPHHSPLVLKRCHVPIHQHNYAAPPSTKVEYPSAKRLKLDNGRILKQISNNRKCSSPR TSDSEENDKRRTHNVLERQRRNELKLSFFALRDEIPEVANNEKAPKVVILKKATEYVI
SIQADEHRLIAEKEQLRRRRDQLKHKLEQLRNSCA
resulted in the same BLASTP anomaly I posted earlier.
I post it here so anyone else can attempt to duplicate my results. The BLASTP did not return Tomistoma Shlegelii at all as I pointed out earlier!
The first results were:
It might be helpful if the protein sequence were in GenBank too. Do you have to make separate submissions for the nucleotide and the protein sequence? Did the protein sequence not get submitted, did it get deleted? What’s up???? For people looking into such papers as yours on gharials in the future, it would be good if whatever is causing the BLASTP anomaly get remedied.
I misidentified Mecistops cataphractus as the false gharial because it was the only 100% coverage/100% identity hit. So I surely assumed this was just a renaming of the species.
Thanks in advance.
Not actually true. You “highlighted” a few protein and DNA sequences, hardly anything to be impressed with. You didn’t say much about the paper itself. I have no objections to compliments, only to wasted bandwidth.
Do you agree with AiG? Do you further agree with me that all crocodylians belong to a single kind?
The 3′ UTR is a sequence that comes after the stop codon. To find the end, you need to sequence some mRNA. I didn’t do that myself, just accepted previous literature. Now that’s out of the way, would you care to hazard an opinion on where the “kind” boundaries are?
Usually. But if I recall, in MYC the first exon is entirely 5′ UTR.
I suggest you forget about protein sequence and BLASTP; there’s no point to it unless you get outside crocodylians.
We didn’t sequence any proteins. The translated sequences are in the entry already, and GenBank does that automatically. Again, there is no reason I can think of to look at the protein sequences.
It’s a renaming of Crocydylus cataphractus. If you transfer a species to a new genus it retains the specific epithet. Now you’ve spent way too much time on your problems with BLASTP, which there was no reason to use anyway.
So the above c-myc protein fragment is 100% covered, and 100% identical for two creatures:
The GenBank phylogenetic classification for Mecistops cataphractus
The GenBank phylogenetic classification for Tomistoma schlegelii
So even though they have 100% identical c-myc protein fragments, they are assigned different phylogenetic branches and nodes starting with a split at Longirostres. No problem so far. But…
for Gavialis gangeticus (Gharial) that was NOT identical, it gets assigned to the Gavialidae node above:
That is to say, the ones with the 100% identical c-myc fragments get assigned different branches from the Longirostres node, but the one that is NOT identical gets assigned the same branch (Longirostres-to-Gavialidae) from the Longirostres node and toward the node (Gavialidae). Why? Is that because of the introns in the gene that aren’t picked up in the protein phylogeny???
I’m not criticizing the fundamental conclusion of the paper. As I pointed out the even many YECs would agree with the paper. But this doesn’t seem 100% congruent, maybe at best a consensus.
Thanks in advance.
Thanks for your reply, but apparently GenBank didn’t generate a separate accession number for the protein fragment of Tomistoma schlegelii when it apparently did for Mecistops cataphractus. Seriously, there are people that might find it handy who are looking outside the crocodiles to have the protein fragement available for BLASTP. But it’s your loss if you don’t want the accession number created and thus people at least have a chance to see your work by a different channel such as BLASTP searches.
I don’t believe that you commented on Kantian Naturalist’s “In Defense of Modest Nominalism” thread a few months ago, but this issue was the main topic being debated.
I can’t see how it would be possible to delineate different “kinds” in any sort of useful way. The decision of where to draw the lines seems entirely arbitrary in virtually all examples that I’ve seen.
John Harshman,
Scratch that, I do see that you commented a few times early in the thread.
It’s because ordering by raw distance is not a good guide to phylogeny. Gavialis has a single autapomorphic change from the common ancestor of Mecistops, Tomistoma, and Gavialis, but the other two haven’t changed. As I have mentioned already several times, the protein sequence is useless (or nearly so) at this level. But the DNA sequences of the intron, exon coding sequence, and 3′ UTR all have enough variation to be useful.
As Erik already pointed out, the hypothesis of common descent wasn’t being tested.
GenBank creates a protein sequence and protein accession number automatically, as you have recently discovered. Why that sequence doesn’t turn up among the first hits when you do BLASTP is unclear to me, but its something you might want to take up with NCBI. I had nothing to do with it.
Erik is wrong. Do you share his odd view of epistemology? Any analysis of a phylogenetic data set is a test of common descent, unless you can come up with a different hypothesis to explain the data as well. The agreement among data sets and the internal consistency within each (here measured by bootstrap values) are tests. Again, unless you have come up with a different hypothesis that explains the results as well.
Thanks. I just did a BLASTN for AY277489 for Tomistoma schlegellii and saw Gavialis giganticus ranked higher based on smaller intron differences. It goes to show those non-coding regions are functional for phylogeneticists trying to build trees.
Hard to see how phylogenetic analysis is not a ‘test of common descent’. Hard to see what Mung, who accepts common descent, bases that acceptance on.
Two caveats: again, raw distance ordering is not a good guide to phylogeny, and it isn’t just the intron sequence that’s useful (“functional” is probably an unfortunate word choice) here.
John,
For making distinctions between phylogenetic branches for closely related species, is there a formula for estimating how long a stretch of DNA and what sort of DNA would be used to make an estimate?
Of course, it’s good to have larger samplings of the genome if one can afford it, but how did you arrive at 1100 being an adequate sequence length? Couldn’t there have been stretches of DNA of that length that wouldn’t have been sufficiently informative simply because you just picked the wrong stretch to sample?
Eukaryote Prokaryote for starters.
Plants and Animals.
Fish and Birds.
Fish and Mammals.
For crocodiles, no opinion, but YECs want as much common descent as possible so as to fit more things on Noah’s ark (unless the crocodiles could stay afloat and live in the water for a year outside of the ark).
Well, that makes it easy. A little bitty Ark to hold 7 people (or however many), one pair of a prokaryote (E. coli in their gut will do), one plant (a couple of pots of house plants maybe), a pair of goldfish, a pair of parakeets, and for mammals, some rats, or else just a pair of dogs.
And all the rest is done by vast rates of mutation afterwards, with lots and lots of natural selection. You know, natural selection, that thing that Sal is always trying to prove can’t do anything to bring about adaptation. Except when the Post-Ark Exception is in force.
Short answer: no. You can certainly make a guess and then see if it works.
1100 was what we had good primers for. And in fact the original purpose was just to have a good outgroup for some birds. It just turned out that the outgroup made a fine publication all by itself. Anyway, the only way to find out if a stretch of DNA will give you a decent tree is to try it on something. And of course in 2003 there weren’t as many complete genomes to try sampling from; developing a new gene for sequencing was much harder than you suppose.
Those can’t all be kind boundaries. There can’t, by definition, be kinds within kinds, so you can’t have an animal kind and a bird kind too. And anyway you definitely don’t think that all mammals are a kind, and probably not all birds.
Why should what you want be at all relevant to science? And if you don’t have an opinion how can you hope to identify kinds? Is this paper evidence for common descent of all crocodylians or is it not?
Joe Felsenstein,
Heh. Yes, my immediate reaction too was the extent to which one has been let off the hook with regards to the dimensions of the Ark. Even less room for ‘diversity’ though.
and …
John Harshman,
Kinds within kinds, haha.
Yeah, I can’t prove I had these thoughts too, but … I did.
I believe the number was 8:
Noah and his wife: 2
Three sons: 3 (Shem, Ham, Japheth)
Three daughter-in-laws: 3 not named
2 + 3 + 3 = 8
Yes, but “evidence for” is not “proof of”. A common genetic code is evidence in favor of universal common descent, but not absolute proof.
However, Complex novel features is evidence against universal common descent. If there aren’t complex novel features that preclude a common ancestor crocodylians, then the YLC/YECs who believe the crocs didn’t swim but were placed on the ark will be happy with your idea. See, they don’t think you and God disagree all the time….
But a YLC/YEC will not accept that a blade of grass and an elephant share a common ancestor because of the not-completely-universal universal genetic code. There wasn’t enough time to evolve them from a common ancestor. If life is young and the fossil record is young, I’d think even you wouldn’t think a few hundred thousand years would be enough for all plants and animals to evolve from a common ancestor.
The Faint Young Sun Paradox puts eliminates the oceans as a viable residence for the aquatic creatures for hundreds of millions of years in the past. The C14 traces suggest the fossil record is young, not to mention the very smooth contact domains between strata.
You asked how we would estimate the Baramin. Anything that can’t evolve from anything else in less than a 6,000 would be the guideline if one accepts the literal timeline from Luke 3, but if one made a timeline independent of the Bible, but based on other data like the Faint Young Sun Paradox, one would still get some Baramin because of waiting time problems, etc.
Regarding the C-myc, it’s perfectly believable a random mutation can create varieties of C-myc alleles. Baramin are not delineated this way, they are delineated mostly by un-evolvabilitiy from another creature. If the creatures are created, the gene trees across species are the result of common design, not common descent. Therefore the analysis of C-myc is insufficient to identify a Baramin. The YECs however are simply providing a guess….
But, I would suppose even you don’t think a plant and a human could evolve from a unicellular eukaryote or prokaryote in 6,000 years. Therefore, if the Biblical timeline is correct, then baramin exist even if we can’t identify them. If one doesn’t accept the Biblical timeline as a premise, then one could build other models using the Faint Young Sun Paradox and C14 or whatever arguments appropriate to build a model. Of course, there is one model that is closest to the truth, and I don’t think the mainstream model is the correct one.
Because of Noah’s Ark, the YLC/YECs have a hate/love relationship with common descent, mostly hate, but with an occasional drop of love.
http://nwcreation.net/evolution_creation.html
stcordova,
Would you say, based on this that, the variation in finch beaks is due to meiotic recombination?
John Harshman,
Didn’t Harshman ask for kinds within kinds in the OP?
“This one is about crocodiles. Are crocodiles all the same kind? How do you know? If they aren’t how many kinds are there within the group and how do you know? If they are, are they a whole kind or just part of a larger kind? And if the latter, what is the larger kind?”
So, why is he questioning it now? He no likes the answers that’s why…
“What’s wrong with this paper?”
It seems to fit the needs of everyone; YEC, OEC, Ark Creationists and Darwinists… in other words a paper that can be interpreted any way you want….
That’s what is wrong with this paper…
I don’t know, and I feel bad since this is a good question! There will be the big Internationational Conference on Creationism 2018 in July/August. If I remember, I’ll ask.
The view in the YLC/YEC community has evolved:
1. fast mutation
2. homologous recombination
3. created heterozygosity
Maybe all of the above.
I personally think we know sooooo little. Yeah we can genetically engineer things, but that doesn’t mean we know the level of detail we need to know to answer some questions.
For example, when John Sanford and other genetic engineers in the 1980’s were re-engineering plants and agricultural products, they didn’t even deal with the “-omic” issues I’m now highlighting. There were probably a few things we engineered and we were really lucky to get them to work. But that doesn’t mean we’ve figured out even a fraction of how things work.
The researchers involved in finch studies definitely have some ideas on the genetic basis of the variation. I don’t discount random mutation for making changes. Sadly we see the effects of this in heritable diseases, some of which oddly may still contribute to reproductive success even if it makes people sicker…
Awh, c’mon in the spirit of Christmas, finally a thread where no one really disagrees with the OP. My Christmas gift to John Harshman. I actually had something nice to say about his work for a change.
Btw, Merry Christmas J-mac.
Looks like Joe’s sombrero covered a part of his bible when he was reading about Noah’s ark…so he couldn’t add up right…
Or his math is not what it used to be…
If your story is right Joe, why don’t you go to the lab and prove it? Instead of speculating about numbers all your life? Have you done any mutagenesis experiments to see if your bluffing has any merit? Until then, you can romance your foolish audience all you want along with Hirschman who has no clue what he published in 2003 but for some reason he thinks it can still work with the help of YEC
Looks like Joe’s sombrero covered a part of he bible when he was reading about Noah’s ark…so he couldn’t add up right…Or his math is not what it used to be…
If your story is right Joe, why don’t you go to the lab and prove it? Instead of speculating about numbers all your life? Have you done any mutagenesis experiments to see if your bluffing has any merit? Until then, you can romance your foolish audience all you want along with Hirschman who has no clue what he published in 2003 but for some reason he thinks it can still work with the help of YEC.
BTW; Regarding mutations rate this is such a bluff, because Joe and The Gang assume the mutation rate has been steady… he doesn’t know what the mutation rate was 200 years ago… or more…I can experimentally prove that this mutation rate is garbage in one generation… none-heritable changes within one generation can be more profound then this mutation nonsense… he has to stick to it though because his speculation genetics paper has to continue to be in the top 1000000 reference papers on speculation genetics or something ….
stcordova,
Are you sure you want to put your name next to this very controversial issue?
YEC may not be happy…The ark ones…
Thanks Sal! You are not only a smart guy but also a really nice guy too..
Sal…there is one thing that I would never compromise on… and I may have told you that before…
My father and his father and my grandfather were the very few opposed to the communist regime that would never compromise… never…
My grandfather wasn’t even a Christian (or Catholic)… There is something about their determination that they had that makes me unmovable… Can you accept it?
The test is not a test if its itself based on a exclusivity option.
Common DESIGN easily can explains trees and better. THEREFORE it nullifys the trees as a test of common descent.
its not a scientific test in other words.
It’s based on 2 nuclide substitutions…which happens back and forth within one generation…
Bob! I’m so glad to …can you resolve the issue of crocks? Did they or didn’t they sleep on the Noah’s boat?
I’m not sure about crocs on the ark. the bible says the creatures with the breath of life went on the ark and so the idea is ‘breathers of air” needed to go on the ark to survive.
So the KIND that crocs come from might of been a landlover. indeed i think there are fossils of croc types who lived on land.
So the diversity of these types started from the land. tHis would make sense also about survival as its always creatures in the water that survived timelines.
What kinds are is hinted at in genesis.
The snake in one kind(to justify a general delegging).
our mimic, primates, must be one kind. yet birds were on the ark in many kinds. so no bird kind.
WONDREfil..cen Y ERxplain?.
And me too, and anyone with two braincells to rub against each other as well is my guess. Nested kinds doesn’t make any sense.
It is? What about the polyconstrained phosphoproteome you have been going on about? In humans, MYC is a proto-oncogene protein and has multiple phosphorylation sites:
Or perhaps phosporylated proteins can freely evolve in clades that do not contain a certain bipedal mammal whose most distinguishing feature is a hugely inflated ego?
John,
it looks like Sal is going to pretend that his indifference towards any clade that doesn’t include humans is to be celebrated as a big concession to you. Perhaps I can spark the discussion with some questions of my own:
1) In the paper you spend a lot of effort identifying a “second signal” in the morphological data (presumably the true phylogenetic signal). How do you distinguish that from the confounding signal? You aren’t cherry picking are you?
2) One of the explanations put forward for the basal position of Gavialis in the morphological tree is a wholesale character reversal (it’s called taxic atavism?). Does that happen a lot? What would cause such a thing?
J-Mac,
OTOH, have you built an Ark and tried to stock it with everything that would need it lately? A proof of concept, like. Or is this ‘experimental’ thing just for us, cartoon science which can only be done wearing a lab coat?
Robert Byers,
Only by saying, to any and all data, ‘that’s DESIGN’. For no reason other than you can utter the words.
You have to admit that having some sponge or fungus-like organism as the only animal on the ark does make things considerably easier. You might not even have to bring two of them, it can evolve multiple sexes after the good god is done drowning all living things on Earth.
Since we have only two trees to choose from, we can easily identify two signals: characters that fit the morphological tree better than the molecular tree and characters that fit the molecular tree better than the morphological tree. (Of course there are characters that fit either tree equally well, but they’re not interesting for this question.) Anyway, we don’t pick the characters; we just observe their fit to the two trees. In the morphological data set, there are of course more characters of the first sort than the second sort; that’s why there’s a morphological tree in the first place. But there are a substantial number of characters of the second sort, and that’s the second signal.
As far as I know it isn’t a common phenomenon, which is what makes this situation interesting. I have no good hypothesis for the cause.
With your great knowledge of science, you should know that there’s no such thing as proof. “Very good evidence” is the best we can possibly get. Can we agree that given the evidence, it’s very probable that all crocydylians are descended from a common ancestor?
However, Complex novel features is evidence against universal common descent.
No, they aren’t. We’ve been over this. If we somehow know they couldn’t have evolved naturally, that doesn’t eliminate guided evolution and so is not evidence against common descent.
Why is the ark relevant to a scientific question?
Well, it wouldn’t be enough if evolution proceeded by known processes. But again, guided evolution could do it. And you should know that the common genetic code isn’t the evidence for common descent of plants and animals. Instead, it’s the nested hierarchy in which eukaryotes are organized. You’re still arguing against a natural origin of novelty, not common descent.
Now who’s cherry-picking? Note that now your argument isn’t with biology but with physics and geology. We could discuss that, but this is probably not the thread for it. When you’ve brought these up before, people have explained why your little anomalies are not big problems.
Not really an operational method, is it? Unless you can quantify what can and can’t evolve in less than 6,000 years, you have nothing. Now, according to everything we know about evolutionary rates, very little molecular evolution can happen in 6,000 years, so almost every species by that criterion must be a separate kind from every other. Why, there must even be multiple kinds within many species. Of course you can invoke miracles to speed up evolution, but now we’re into guided evolution where anything is possible, and there goes your criterion.
.
There’s another non-operational criterion. How do you determine “unevolvability”? How do you decide whether gene trees across species, i.e. nested hierarchy, is the result of common descent or common design? How does common design explain a nested hierarchy?
Can we agree that accepting the biblical timeline takes you outside science? Your reasons for rejecting mainstream science boil down to your biblical literalism and a couple of cherries you have picked from the mass of data. That isn’t science at all. And of course you are assuming no guided evolution, which could certainly cause plants and animals to appear from a common ancestor in 6000 years. After all, creationists require absurd evolutionary rates to stuff the ark. What’s a little bit of extra magic when there’s a divine magician?
If there’s no common descent, how do you explain that nested hierarchy? However, let’s save common descent of all eukaryotes until later. Are we agreed on common descent of crocodylians yet? I will point out that there are several previous papers and many subsequent papers that arrive at the same tree from different data, and a little google search will find them. How much consilience is necessary before we can agree?
I have not had time to carefully read the paper until now. It is certainly a good and careful job. The late 1990s saw many systematists and molecular evolutionists getting nuclear gene data to complement the large amount of previous work on mitochondrial gernes. This paper is a solid advance on that front, and does a particularly thorough job of comparing the evolution of molecular and morphological characters.
The internal consistency of the signal from molecular sites is impressive — different sites provide independent signal supporting the molecular tree. Thus there is considerable signal of common descent.
I note that the difference between the morphological and molecular trees seems to come from the position of the gharials. Based on the runs done, I suspect that the problem is caused by covariation in the evolution of morphological characters. The analyses of morphological characters done here uses discretely-coded characters that are treated as if they evolved independently. I suspect that the postcranial characters had substantial evolutionary covariances, owing either to genetic correlations or selective covariances (covarying selective pressures). The methods available for morphological characters do not take this into account, so they are likely to be overimpressed by apparently reinforcing signal.
If the gharial is removed from the data set, which of course we do not want to do as its position is one of the major concerns of the study, the morphological and molecular trees are entirely consistent with each other. This suggests to me that selective covariance (selective correlation) on the line to the gharial Gavialus is the issue, with selection changing multiple characters simultaneously. Genetic correlation seems less likely given the diversity of characters involved.
Bingo.
Unless by common descent people mean blind, unguided, purposeless descent with modification.
ETA:
Did you means to say guided common descent? Because you are using evolution and common descent inter-changeably.
There’s no law that says that kinds cannot be nested. They would just not be nested due to common descent.
Can phylogenetic analysis tell us that two species are not related? If so, please provide an example.
My understanding of kinds are specially created groups of organisms. Obviously that means they must be independent and cannot be nested within each other. You can classify them in nested hierarchies if there are shared features that allow that but the higher groupings are not kinds in themselves.