Cool new paper by Robert Asher

i think that its actually does make sense. for instance: a fish protein should group with other fishes since fishes have many things in common: they have similar morphology, similar habitat etc. so it make sense to think that they will share more similarity among a specific protein than with say a mammal or a reptile. and if its true for a single protein it should be true for other fish proteins. i dont see why it doesnt make sense for you.

Yes. In order to see that it would be necessary to read what he wrote, which you clearly did not do.

3 Likes

@scd I think I understand what you are saying… You should read this post here, which includes a similar assertion and where Joshua follows up with this:

It is not that it doesn’t make sense, or that it isn’t intuitive, it is that it isn’t accurate.

2 Likes

But they’re NOT grouped based on similarity. I don’t know how many times I have to repeat that.

The TREES are similar, THAT is what you are being asked to explain. Not the fact that the proteins are similar. Since we know what the proteins do, it actually becomes mysterious that they’re not completely identical, since they do the exact same thing in each species, as I was careful to explain. But you can even put that aside.

That the trees are similar, which were derived from different independent sequences, some of which are even nonfunctional. Why should a nonfunctional sequence be constrained to yield a tree similar to one derived from a functional gene? Or to yet another nonfunctional sequence? Why do different species even have similar nonfunctional sequences in them, and (again), why do they reproduce the same trees?

I believe I actually explained why that would not be the case by pointing out how no such constraint actually operates on these sequences. Why should the tree derived from one protein affect another protein’s tree, and why should it affect the tree derived from a nonfunctional sequence?

I’m running out of ways of explaining this and I don’t know what to do short of you having to literally go take courses in molecular evolution. I suppose some times being tutored and having to work out problems and tests for yourself can’t be substituted for by just reading stuff in your own spare time. I will say, though, that the kind of mindset you take with you to a certain question makes a big difference for comprehension. When you’re reading something technical looking for something to disagree with, or perhaps being unwilling to let go of your preconceptions and “try on” another pespective, then argument and debate is hopeless. Wanting to understand is essential.

I emplore you to at least TRY to understand why we would say the things we do. You don’t have to agree with it, but if you can figure out why we say it(hint: it’s not fear of God’s judgement), as in try to consider “what would have to be true, for their arguments to be valid?”.

4 Likes

@scd Are you beginning to see the difference between what is intuitive and what is borne out from the data?

ok. maybe that image will help us to communicate better because a language barrier. here is the phylogeny of the cytochrome c protein:

%D7%AA%D7%9E%D7%95%D7%A0%D7%94

(image from britannica site)

you can see that the cytochrome c tree also fit with the creatures similarity in general. so for instance human and a monkey share more similarity on that protein than say human and a tuna. now, this similarity/difference (even the headline called “Phylogeny base on nucleotide difference”) comparison indeed give us a tree. and this tree indeed base on level of similarity\disimilarity. now, if i got it wrong again please tell me where im wrong and we will continue from there. i can also discuss about the non-functional sequence argument but lets focus each point at time.

First of all, in many ways not really.

Second, superficial appearances can be deceiving. The morphological similarities aren’t scored here in any rigorous way, so we are left trying to eyeball it from some silhouettes, which is completely meaningless and subjective. You need to show me a phylogeny based on morphology, scored by similarity.

I’m looking at the Tuna, and I’m trying to determine how “similar” it looks to the moth. It’s not obvious to me how similar they look at all(how do I score this? Where is the character matrix we should use?). They are separated by a score of 38.2 nucleotide differences in the cytochrome C gene. (I don’t really understand the algorithm used here, some of the minimum number of substitutions are indicated to be negative, which makes little sense to me).
I compare the moth to humans, and I get
Human-Moth 33.4
Tuna-Moth 38.2
Is human more similar to moth? By what characters?

I try Pigeon to moth and get 30.6. Are Pigeons even more similar to moths than humans and tuna? It’s not obvious to me that they are.

Rabbit (30.3) is more similar to moth than dog is (32) in nucleotide sequence from cytochrome C. But it’s not obvious to me that a rabbit looks more like a moth than a dog does. Hey, maybe they do to you, who knows? Without an objective method of scoring their characters, it’s impossible to tell.

Rattlesnake is apparently slightly more similar to humans than tuna by cytochrome C sequence. I don’t know why, they have no limbs, at least tuna do.

Is the turtle obviously more different from pigs than chickens are? Pigs and turtles run around on all fours, chickens fly. Cytochrome c says yes.

It gets even less obvious with the fungi. How similar do baker’s yeast look to pigs versus chickens to you? Is baker’s yeas obviously more similar than Candida is? Cytochrome c says yes. Etc. etc.

Now the problem here isn’t so much that you can find examples of morphological similarity also corresponding to genetic similarity, certainly you can. The claim was never that this can’t be done or isn’t ever the case.

The question you were posed is why we obtain this result(why are the trees similar, which has not been demonstrated in this case), when we have zero evidence for any functional constraints between the data sets. What is it about the morphology of humans that constrains the cytochrome c-sequence to be phylogenetically grouped closer to the cytochrome c-sequence of monkey, than the cytochrome c-sequence of rabbit? You seem to be saying that it’s because humans are morphologically more similar to monkeys than to rabbits. Even supposing that is actually the case, that still doesn’t explain why the trees are similar.

Why should there be any degree of correspondence here? Cytochrome c is a protein in the electron transport chain. It’s involved in core energy metabolism(involved in turning food into work), it has nothing to do with the order of appearance, shape, or size of anyone’s limbs, bones, organs, numbers or position of eyes. Even supposing cytochrome c was involved in some developmental patterning somewhere, why would that constrain it’s nucleotide sequence to yield a phylogenetic tree that mirrors the morphological tree? Even if they were both based on similarity scores that would still not explain why one’s tree mirrors the other when they are independent data sets. That’s the whole point here.

If what you want to accomplish is to show that consilience of independent phylogenies isn’t evidence for common descent, what you need to be doing is demonstrating their non-independence in such a way that their putative non-independence would end up constraining the trees inferred from them to be similar for whatever functional reason.

4 Likes

“so for instance human and a monkey share more similarity on that protein than say human and a tuna”

Actually, you can’t read that off the tree. I’m sure that claim is true for most of the species shown, but the tree doesn’t show it. What it shows are inferred numbers of changes along particular branches, and the sums of those changes will not match the pairwise distances between two taxa. Nor does “based on nucleotide differences” mean what you think it does. I suspect that this tree was built by least squares fit of a simple matching distance matrix, but one can’t be sure without a real reference.

3 Likes

I know! Repeating it infinitely won’t be enough. :smile:

1 Like

Perhaps, but maybe not. In this case I suppose the results are nearly identical either way.

It’s not a parsimony tree, given the fractional branch lengths. Could be neighbor joining. Conceivably maximum likelihood. Wait, there’s a negative branch. Scratch likelihood. It’s got to be some distance method.

2 Likes

It’s from here: Evolution - Species, Genetics, Trees | Britannica

This is what it says in relation to the figure:

Distance methods

A “distance” is the number of differences between two taxa. The differences are measured with respect to certain traits (i.e., morphological data) or to certain macromolecules (primarily the sequence of amino acids in proteins or the sequence of nucleotides in DNA or RNA). The two trees illustrated in the figure were obtained by taking into account the distance, or number of amino acid differences, between three organisms with respect to a particular protein. The amino acid sequence of a protein contains more information than is reflected in the number of amino acid differences. This is because in some cases the replacement of one amino acid by another requires no more than one nucleotide substitution in the DNA that codes for the protein, whereas in other cases it requires at least two nucleotide changes. The table shows the minimum number of nucleotide differences in the genes of 20 separate species that are necessary to account for the amino acid differences in their cytochrome c. An evolutionary tree based on the data in the table, showing the minimum numbers of nucleotide changes in each branch, is illustrated in the complementary figure.


%D7%AA%D7%9E%D7%95%D7%A0%D7%94

The relationships between species as shown in the figure correspond fairly well to the relationships determined from other sources, such as the fossil record. According to the figure, chickens are less closely related to ducks and pigeons than to penguins, and humans and monkeys diverged from the other mammals before the marsupial kangaroo separated from the nonprimate placentals. Although these examples are known to be erroneousrelationships, the power of the method is apparent in that a single protein yields a fairly accurate reconstruction of the evolutionary history of 20 organisms that started to diverge more than one billion years ago.

Morphological data also can be used for constructing distance trees. The first step is to obtain a distance matrix based on a set of morphological comparisons between species or other taxa. For example, in some insects one can measure body length, wing length, wing width, number and length of wing veins, or another trait. The most common procedure to transform a distance matrix into a phylogeny is called cluster analysis. The distance matrix is scanned for the smallest distance element, and the two taxa involved (say, A and B) are joined at an internal node, or branching point. The matrix is scanned again for the next smallest distance, and the two new taxa (say, C and D) are clustered. The procedure is continued until all taxa have been joined. When a distance involves a taxon that is already part of a previous cluster (say, E and A), the average distance is obtained between the new taxon and the preexisting cluster (say, the average distance between E to A and E to B). This simple procedure, which can also be used with molecular data, assumes that the rate of evolution is uniform along all branches.

Other distance methods (including the one used to construct the tree in the figure of the 20-organism phylogeny) relax the condition of uniform rate and allow for unequal rates of evolution along the branches. One of the most extensively used methods of this kind is called neighbour-joining. The method starts, as before, by identifying the smallest distance in the matrix and linking the two taxa involved. The next step is to remove these two taxa and calculate a new matrix in which their distances to other taxa are replaced by the distance between the node linking the two taxa and all other taxa. The smallest distance in this new matrix is used for making the next connection, which will be between two other taxa or between the previous node and a new taxon. The procedure is repeated until all taxa have been connected with one another by intervening nodes.

1 Like

The relevant bit is “Science, vol 165, Jan. 20, 1967, p. 281”. Look that one up.

From what I can gather it’s the Fitch–Margoliash method used to construct the tree(which is well explained in the paper), and the one shown is the best among a mere 40 different trees tested by the algorithm, which I suppose was standard for 1967 computers?

So a least-squares method, as I had supposed. Yes, searches have become faster and more extensive since 1967.

ok. i got your point. you are basically asking why there should be a connection between morphology and enzymatic activity (supposedly there is no connection between the two). so all we need to find is that there is a connection between the two. am i right?

The scientists are using math, @Robert_Byers1. Their math has been soundly executed and explained, albeit in a form that a majority of folks have not studied.

You are asserting that you can grasp these concepts without grappling with the math. This assertion is completely, 100% wrong.

Instead of shouting with all caps, you would do better to go back, read the papers carefully, and ask questions about the math you might not understand.

If you just keep shouting and avoid doing the hard work, I will be forced to assume that your posts about science are not worth reading.

Best,
Chris

4 Likes

Showing a connection would be a first step in a bigger job before you, but yes that would be one of the things you would have to do. After you have found this putative connection you think there is, then you need to demonstrate that this connection should constrain tree-topologies to agree.

In other words, not just that there is SOME physiological consequence on anatomy derived from (say) the activity of some enzyme. We know there are such connections. For example, if you have certain metabolic diseases caused by mutations in certain enzymes, this can result in stunted bone growh or muscle development (and many other possibilities). So it is not merely that you need to show that one has some effect on the other, but that the effect it has functionally demands that phylogenetic trees that can be derived from the species that have these genes should be forced to match each other.

2 Likes

ok. here is apaper that show a connection between supposedly morphology-neutral genes and morphology:

“Evidence for an ancient adaptive episode of convergent molecular evolution”

first note that this happen in similar creatures (snakes and agamid lizards). so we probably will not find it among a lizard and any kind of mammal or a fish or a bird. this fact alove prove that similar morphology also effect similar non- morphological genes. it also show that there is a connection between morphology and these genes:

“The molecular convergence between snakes and agamid lizards may thus have driven by similar adaptive pressures on metabolic function affecting both lineages”

since metabolic function may be effected by the creature morphology.

Nope. In fact it’s evidence against that claim, because snakes are more similar morphologically to their actual relatives, iguanid lizards, than they are to their convergent buddies, agamids. Congratulations: your citation argues against your point. It’s a difficult point, but creationists in my experience manage it more often than you might expect.

Sorry, no. They’re affected (with an “a” not an “e”; spelling pet peeve) by the creature physiology.

4 Likes