We are cluttering up a thread that’s not about convergence with discussions about convergence. Here, we can discuss this specific issue.
Convergence, or homoplasy, occurs when two individuals share a feature that is not inherited from their (assumed) common ancestor. Such a feature may be a phenotype – for example, sharp teeth in many predators, streamlining in fast swimmers, wings in diverse groups – or a genotype. The former is thought to typically arise by natural selection. The same fundamental phenotypic reward – holding on to prey, catching or escaping – leads to similar outward manifestations of the phenotype, although typically the developmental pathway in the different species is different. There is ‘virtually no’ parallel convergence in genotypes. Convergence in gene sequences generally arises as a result of stochastic variation. If one has two diverging lineages, the limited number of changes available to each DNA base pair (3) results in a significant chance that a base pair held in common by two sequences is not there due to homology (common ancestry) but homoplasy.
It’s important to separate out these two senses. Much cross-talk arises because people are talking of one as if it were the other, sometimes with the lame justification that ‘it’s all molecules’, or the convergence directly constrains a biochemical system. These are misunderstandings; it’s the distinction between phenotype and sequence which is important, not whether the phenotype is ‘molecular’ or not.
Now, the reason for this discussion is that some people are arguing that homoplasy should cause one to reject common descent. So here is their opportunity to defend that view.
[eta – whatever one thinks of Biologos, they provide here an accessible introduction to homoplasy and its relation to homology]
Seems to me homoplasy is a strong argument for the non-random agent of selection being the niche environment. Diverse organisms (their gene pools) will stumble upon similar adaptations. The fact they can be found by distantly related species using different genetic sequences, proteins and so on confirms the richness of functionality in sequence space.
Mung brought up antifreeze proteins in fish. This provides a couple of useful points.
Firstly, he did so in order to express incredulity at my assertion that molecular sequence comparisons exhibited homoplasy mainly for stochastic reasons. My assertion was actually related to the genotype, not the phenotype – the ‘molecule’ I had in mind was DNA. Proteins sit in something of a middle ground between genotype and phenotype. Their sequence is derived from genotype, but the actions of the translated protein are firmly in the arena of phenotype. The sequence folds, for physical reasons, or has direct physicochemical effects, and is thus more than merely a string of that ineffable quantity, ‘information’. Phenotype does constrain genotype – selected sequences form a subset of possible sequences – but that correspondence is not 1:1, because of the redundancy in the genetic code (scare quotes omitted). If there is a phenotypic constraint on a particular acid at a degenerate site (more than one codon for the same acid), there is still latitude in codon usage even if the acid itself is tightly constrained. Within that reduced, but still variable, portion of the codon table, there are degrees of freedom, and stochastic variation, which can be informative on phylogeny.
The second point is that we can only see it’s a homoplasy at all because it fails to conform to the hierarchic structure given by other sequences. Far from invalidating common descent, it relies upon common descent to be at all noteworthy! But furthermore, the underlying genes themselves aren’t convergent – that’s evidence that it’s a (phenotypic) homoplasy, and not a homology.
Tripeptide, by the way. That’s 3 acids.
From the same issue of PNAS:
Proc Natl Acad Sci U S A. 1997 Apr 15; 94(8): 3485–3487.
PMCID: PMC34156
Origin of antifreeze protein genes: A cool tale in molecular evolution
John M. Logsdon, Jr. and W. Ford Doolittle
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC34156/
🙂
ETA: That’s just what you need for a good word salad!
Mung,
And now for a substantive critique … 🙂
Whether amino acid sequences are genotype or phenotype appears to depend on who you ask. Andreas Wagner, for example, includes amino acid sequences. But I appreciate the clarification as to what you mean when you say genotype.
A substantive critique from me, LoL? You have me confused with someone else. 😉
I am fairly confidant that I can produce numerous examples of convergence at the level of proteins. But if that’s not what we are talking about they won’t be relevant. DNA sequences I’d have to do more study. But either way, I don’t think they swamp out or otherwise ‘disprove’ common descent.
So what would I be critiquing? (Yeah, I know, I’m no fun.) But if I played devil’s advocate people would accuse me of being a *gasp* Creationist.
Similarity of protein is supposed to indicate how closely related two organisms are. Yet we have similarity at the level of protein but not at the level of DNA. So similarity does not indicate the two organisms are closely related, and the DNA sequence would indicate that they are not (given that similarity of DNA sequence also is supposed to indicate how closely related two organisms are).
So here we have a case where the “twin nested hierarchy” fails. The genotype and phenotype don’t give you the same tree. The whole big deal about the “twin nested hierarchy” is that the two are supposed to give the same results.
This is why I comment as I do about the “twin nested hierarchy” and also ask WHICH hierarchy. Because you want the one where the two match up. You don’t want the one where the two do not match up.
You can have proteins coded in DNA by sequences that vary at points where the code is redundant but that does indicate homology plus drift. You can have proteins that have same or similar properties (fish antifreeze example above) but have evolved convergently, the DNA sequences for those will not be homologous.
Similarity in DNA sequences indicates homology and relatedness. Similar function in proteins that are unrelated in DNA codon sequence are examples of homoplasy.
But they do. Linnaeus’s tree was built purely on studying phenotypes but when DNA sequencing became available, it largely confirmed the original hierarchy
Where they don’t, go with DNA sequence homology. Trust me!
Thanks for the link PeterP (and welcome to TSZ)
From that paper:
DNA sequence similarity is supposed to be a more certain guide because … ?
Mung
Well you can’t get a paternity suit to stick just because the child looks like her father.
Mung,
I’m a bit doubtful that he actually does say that. As I say, when we get to proteins, we have 2 things, primary sequence as an ‘informational’ construct, and physics – the consequences of implementing that sequence as atoms with cascades of effect. So, of course, even DNA has a phenotype. It’s still important to be clear which one is talking about. I guess I wasn’t.
Mung,
Because that’s what’s replicated. Proteins aren’t replicated. Given the conservative nature of template-based DNA replication, one can get a far clearer view of relationship by accessing primary sequence than one can any ‘munged’ version of this.
Of course, primary sequence is munged by mutation. That’s where things get interesting. Homoplasy (genotypic) is but one of the things that degrade the information content of relationships over time. (Of course, it’s just an inevitable consequence of mutation).
We have actual parents and their children in a known relationship which allow us to identify reliable markers in DNA for establishing paternity.
Is that how we arrived at DNA sequence similarity is supposed to be a more certain guide for comparing humans to non-humans? Have you learned nothing from Erik’s posts?
Mung,
That’s so 1970’s! DNA sequencing gives us a much richer dataset.
Not in every last little bit. Remember that the important sequence in the antifreeze proteins is 3 acids – 9 DNA positions – albeit repeated. These are embedded in genomes billions of base pairs long. A 9-bit sequence is hardly enough to invalidate (or, for that matter, validate) the entire relationship.
They do not match up with exact perfection at every node for every base pair, for a variety of reasons. One would hardly expect them to. But the fact that they both tend to converge on the same broad consensus trees is powerful support for each approach – the molecular and the phenotypic. Each is a test of the other, albeit not wholly independent.
Mung,
I certainly haven’t.
Is this is becoming a pattern?
For example, page 47 of The Origins of Evolutionary Innovations.
Genotype: A sequence of ribonucleotides or amino acids.
Phenotype: Secondary and tertiary structure, biochemical activity.
Mung,
Fair enough, he’s wrong then.
I think that’s a bit too harsh. He’s trying to create genotype – phenotype maps that are at least somewhat tractable.
I can sort of see why Wagner might say it. He means the informatic content of the protein as opposed to its physical manifestation. But, I think ‘genotype’ is the wrong word; an important part of the concept – indeed the reason for its introduction, by Johanssen (Rumraket would understand the source!) – is to indicate the heritable component.
I know who Johanssen is, lol.
I like quantum convergence best…where many totally unrelated animals “evolved” quantum entanglement systems to be able to use quantum effects to “see” Earth’s magnetic field to navigate in way more superior to the ones more intelligent scientists (one would hope) designed…
Evolution of an eye is already a mystery… quantum-entanglement-based compasses in birds’ eyes, insects, fish, many mammals, of yet unknown biochemical reaction, taking place in their eyes or elsewhere …not to mention the neurological processing in their brains…is another one of those mysteries driving the great evolutionary paradigm…whatever that is today…
This example must be just one of those countless mysteries of evolution no one can logically explain…
Well, then one should be able to get it to unstick by claiming the similarity in DNA is due to convergence?
They just happened to grow up in a similar niche, so its not surprising the DNA turned out the same.
It seems to me that you generally wouldn’t construct a phylogeny by using both the amino acid and DNA sequence of the same gene, and then pretend they corroborate each other as if they were independent data sets. It is pretty clear there is a direct causal relationship between the two. The DNA sequence directly determines amino acid sequence. So if you build a distance-based phylogeny using DNA sequence, and then build another distance based phylogeny using protein sequence, you’d be expected to get trees corroborating each other. But that would obviously be fallacious, as their sequences are correlated by a process independent from genealogy, as the DNA data set directly determines the amino acid data set. Genetic code redundancy will bring some caveats to that, but roughly speaking you’d expect them to produce the same phylogeny.
No, because we are talking about a single generation of separation. There couldn’t have been any time for selection to yield convergence in either sequence or phenotype.
Convergence is the result of many hundreds of thousands of generations of selection.
So no, you couldn’t invoke that excuse to get off a paternity suit, nor does it constitute a problem with the inference of convergence.
Except when it isn’t.
Improbable Destinies
This is false. See the link in my previous post.
What am I supposed to see here?
Why do you say that? Relatedness in the sense of DNA sequence identity is most close in clones and identical twins, progressively less so as relatedness decreases through family, local group, sub-species, species and so on. The relatedness is there from the moment of creation of the zygote. Convergent evolution takes generations.
It links to a thread topic. Can you be a little clearer?
Homoplasy: The Recurrence of Similarity in Evolution
Headed out to see if I had this in storage. Traffic was horrible. Perhaps tomorrow.
Rumraket,
Why do you say you are only talking about a single generation?
One persons DNA sequence could be exactly the same as another persons, just because, in their lineages they both ended up getting the same lucky accidents. Entirely unrelated. Same niche, same lucky accidents.
If it can happen for an eye…
Evolution can take place in less than a generation!
I keep directing people to this cool book I’m reading. They are afraid of it. Especially keiths. He’s running from it like mad. He squirmed, stalled, and then skedaddled.
Makes me laugh.
Mung:
That’s your wet dream, isn’t it?
Why does that excerpt make you laugh?
Convergence in biology is a great problem for evolutionism. the odds of such likeness in unrelated creatures, from selection on mutations, is just unbelievable. iN fact I think its a future emphasis of ID/YEC. As more investigation goes on more convergence is found .
Never mind these dna stuff. jUst plain head shapes in creatures proves there is a problem.
Its very likely a creator with a common design would place within a common design of biology, remember post fall, bodies able to switch body plans as needed.
for example the funny concept of saying egg laying proves a common descent.
Yet Platypus , birds, crocs, snakes(some), dinos(just more birds or others) .
Yet a creator would have in the design the simple design of EGG LAYING as needed for circumstances. Don’T draw wrong conclusions, much less proof, for common descent.
Its another failure of evolutionary biology scientific investigation.
they didn’t allow for another option for likeness in body plans.
They convinced themselves too quickly. Starting with those old dudes before evolution came when making classification systems.
Error on error. As usual.
Amen!
Robert Byers,
Are you a Benny Hill follower Robert?
Allan Miller,
To infer orthodox common descent, converging on the broad consensus tree is not enough. You need to convince yourself that the biochemical differences are due to changes from reproduction alone along with any changes to the germ-line prior to sexual reproduction.
How does orthodox common descent differ from unorthodox common descent? What do you think you mean by “biochemical differences? If you refer to DNA sequence differences or to differences that arise from DNA sequence differences, isn’t the close resemblance between those differences and the sorts of things that are observed to happen from mutations good enough evidence?
Two insertions.
This innovative book is certain to provoke discussion of homoplasy and provides compelling evidence for particular theories of evolutionary change.
– Homoplasy: The Recurrence of Similarity in Evolution
Theories of evolutionary change. Heh.
colewd,
Wha? I just don’t understand this, so I’ve no idea what I’m supposed to be convincing myself of.
I see a lot of hand-waving and vague nods in the general direction of something to do with homoplasy from Mung. Again, not sure of any actual point.
They don’t have any points to make, as usual. The most you can say about convergence is that it’s statistical unlikelihood is directly proportional to the length of sequence at the nucleotide level. But I can’t think of any case of large levels of nucleotide convergence, so unless someone can point out such a case I don’t see any problem.
I see what you mean (about the single-generation thing) but no, it hasn’t actually happened for any eyes. There might be convergence at the amino acid level of proteins, but I’m not aware of any similar case for DNA sequence.
You seem to be laboring under the misapprehension that molecular biologists have invoked convergence to explain the independent origin of highly similar DNA sequences in separate very distantly related organisms (and apparently, you think his has been invoked to explain eyes, where do you get this?). They have not, so that excuse is simply not available to you in the case of a paternity test.
Mung,
The original paper’s in Danish, was the point.
Hmm, who is this Johanssen guy you speak of?
Rumraket,
Him! Coined the terms gene, genotype & phenotype.
Robert Byers,
Yeah, let’s sweep the entirety of molecular phylogeny under the carpet, since Creationism has no answer to it. Now, on to the superficial …
Rumraket,
Wrong wrong, bad comprehension. What I said was if the same lucky accidents could happen multiple times to give rise to eyes, then the same lucky accidents could happen to give two unrelated individuals the same DNA.
Its as simple as that.