More Recent Comments

Thursday, December 08, 2011

Homology

 
There's an interesting discussion about homology going on in the comments section of Fishing for Creationists. The creationists are claiming that homology disproves evolution.


33 comments :

Bilbo said...

Larry,

If I understand what you claim to be the official definition of homology, the various limbs in the diagram are only homologous if they all develop from the same genes, correct?

efrique said...

Bilbo,

can you point to where Larry said "the same genes"? I've read the post he's referring to and he quite clearly says something substantively different. Are you trying to trap him into something you can quotemine?

Jud said...

Rather than concentrating so exclusively on parsing definitions to the exclusion of the sense of the thing, look at how outwardly different-appearing structures (arm and hand, leg and paw, flipper, wing) in fact are quite similar at two deeper levels, skeletal (look at the colored groups of bones) and yes, genetic.

The actions of identical or extremely similar groups of genes are modulated differently in development of the different species, so that for example the actions of the genes governing the growth of the bat's "finger" bones are allowed to proceed until those bones are longer than the equivalent (homologous!) bones in human fingers.

Anonymous said...

Bilbo,

I'm not Larry but I'll answer anyway:
No, not quite. The limbs are homologous if the common ancestor of those vertebrates had limbs and there is continuity with that ancestor. The genes controlling it may have changed however. Its not hard to see how this can happen- a random insertion of a transcription factor into a genetic network creating a redundancy followed by the loss of the original TF. I dont think there have been many cases of this during vertebrate limb development though
Consider this: the turtle shell is clearly not a homologous structure- only turtles have them, but the genes involved early in shell formation are the same ones used for the limb. This is an example of cooption. In its simplest form all it requires is a single nucleotide change in a crpytic transcription factor binding site to create a bonafide TF bs to deploy and entire genetic network.
Bottom line is the notion of homology has gotten very complicated with the explosion of knowledge in recent years ( and in my view should be retired) and IDers use this confusiion to their advantage

RodW

mrr said...

I agree that homology is a complex notion. Notably, it is used (and sometimes defined) in different ways in different fields.

As for many other complex notions, which are necessary to describe a complex world, this means it's hard to work honestly with the concepts, and easy to abuse them if you don't know better.

If I may make a little plug-in for our own work, we made an effort to clarify all homology-related terms some time ago. It is discussed on the Byte size biology blog. Notably, in relation to this discussion, the following definitions might be useful:

Homology: Similarity that results from common evolutionary origin
comment: This broad definition encompasses all the working definitions proposed so far in the literature.

sub-categories:
Biological homology: Homology that is defined by sharing of a set of developmental constraints, caused by locally acting self-regulatory mechanisms of differentiation, between individualized parts of the phenotype.
comment: Applicable only to morphology. A certain degree of ambiguity is accepted between biological homology and parallelism.

Historical homology: Homology that is defined by common descent.
informal comment: this is the most widely used definition in mainstream evolutionary biology, and includes the types of molecular homology, such as orthology.

Structural homology: Homology that is defined by similarity with regard to selected structural parameters.

You can browse all these definitions here.

Joachim Dagg said...

@efrique:
Based on the English Wikipedia entry on homology, Larry wrote: "Thus, when two structures are homologous they are, by definition, descended from common ancestral genes and share the same embryological history. If they do not share the same ancestral genes then they are not homologous for the purposes of classification. Insect wings, for example, are not homologous to the wings of birds."

@Larry, Bilbo
The German Wikipedia entry details the morphological criteria for homology as used before the advent of the molecular age.

1. criterion of position.
for example, insect legs are always found at the thorax, no matter how differently formed these legs are.

2. criterion of specific structure.
best example is the identity in fine structure of vertebrate eyes in comparison with the difference in fine structure with octopus eyes despite outer similarity.

3. criterion of continuity.
If two organs cannot be homologised directly, but a series of intermediates exists that can be homologised serially, then the end points of the series are also homologous.

4. It's a comparative approach.
It's useless to speak about morphological homology without also considering analogy and convergence. Two traits are homologous only in comparison to a thrid that is not.

As you can see, there was nothing DNAish in the definition of homology before the discovery of DNA.

Molecular research revealed a still deeper level of homology, homeobox and hox genes, and Jud already said something on the question of how these levels interact. It is an interesting research topic, not a problem upon which to dump a proven theory.

Paul Nelson said...

True story, from my experience. In the 1980s, Leigh Van Valen used to assign his students his widely-cited paper, "Homology and Causes," Anatomical Record 1982 (freely available as a pdf at www.leighvanvalen.com). The paper opens with the sentence, "Homology is the central concept of anatomy, yet it is an elusive concept" -- a statement sure to draw the curiosity of any student. Just below, on the first page, Van Valen defines homology as "correspondence...caused by continuity of information."

But his notion of information was subtle and surprising. In class discussion, someone (not me -- might have been Mike Foote) pointed out that "information," as Van Valen defines it in the paper, could include pre-Darwinian concepts, such as might be found in the writings of Richard Owen or Georges Cuvier.

Van Valen said, with a mischievous twinkle in his eye, "Yes, that's right."

Check out the paper; it's well worth a look.

Larry Moran said...

Bilbo asks,

If I understand what you claim to be the official definition of homology, the various limbs in the diagram are only homologous if they all develop from the same genes, correct?

I did not say that there was a single "official" definition of homology. What I said was that when you're talking about phylogeny or taxonomy there's only one correct definition.

So, the answer to your question is "yes." If those limbs are homologous we would expect to see a common ancestor with homologous genes for blood vessels, muscles, cartilage, nerves, bones, etc.

The difference are due to the timing of gene expression and that's controlled by regulatory proteins during embryogenesis. Thus, we would expect to see species-specific differences and that will almost certainly include different genes for regulation.

Larry Moran said...

@Paul Nelson,

Please stop your selective quote mining and try to look at the big picture.

Let me remind you what Phillip Johnson said ...

The features that create the classification, such as hair or fur in mammals, are called homologies. They're supposed to be inherited from a common ancestor. But, in fact, in a great many cases the "homologies" are traceable to different parts in the embryo and to different genes. So, in short, the animals get them by an entirely different route and this is strongly inconsistent with the common ancestor hypothesis to explain them.

It's also a well-known fact among embryologists but it never comes out to the general public because, well, it's so unpalatable a fact and so difficult to explain under Darwinian theory.


Do you honestly believe that's an accurate representation of the current state of knowledge in evolutionary biology?

Paul Nelson said...

Larry, the "current state of knowledge in evolutionary biology" about homology is total chaos. But what Johnson says about the divergence of embryonic precursors (for such basic features as germ cells, for instance), while abbreviated, is accurate.

You give no indication in what you write of having grappled seriously with this problem, which is the daily concern of evo-devo workers who venture beyond their own model systems. But saying as you did, in a previous comment in the Fishing for Creationists thread, that the only anomaly is the divergent formation of lens crystallins, is so wildly wrong that I doubt you know the evidence in any depth.

Can say what "evolution" rules out -- excludes definitely -- with respect to homology?

Anonymous said...

Hey Paul, Van Valen hinted at something he called information, which you think refers to your garbled sense of information? Big deal! Is there a point you want to make from the paper? Or challenge here? No. Or if you want to run the gauntlet and get walloped again, caught as usual quote mining...try it. That's the fun thing about IDiots. They always come back for more.

Truti

Jud said...

Paul Nelson writes:

Larry, the "current state of knowledge in evolutionary biology" about homology is total chaos.

So scientists are now uncertain as to whether fish gave rise to dolphins and birds to bats?

The more I read of what you write, the more I am reminded of this old marketing teaser for a Firesign Theater record:

"Dogs flew spaceships! Aztecs invented the vacation! Everything you know is wrong!"

Anonymous said...

Paul,
Larry, the "current state of knowledge in evolutionary biology" about homology is total chaos.

That is nonsense for several reasons a few of which are,
-there is nothing to support the conclusion in any of the papers you quote mined
-your use of the word "chaos" is careless and adds absolutely no value to this scientific discussion
-your carefully constructed pack of lies - oops! cards - has once again been laid waste

You were caught lying yesterday twice, once referring to Hall and the again to Eberhard.

That Johnson is a science and integrity challenged charlatan has been established, as you are dishonest quoteminer and a repeat offender. You are now coming across as a pompous self-obsessed gasbag who trots out one of those tricks that no one one but you and your flunkies admire. Having been shown up to be yet another empty showboy of an IDiot your self-righteous anger is showing up. The same thing happened to your fellow IDiot Johnson in the days of Usenet, who scampered away like a coward when scholars began to needle him.

You give no indication in what you write of having grappled seriously with this problem, which is the daily concern of evo-devo workers who venture beyond their own model systems. But saying as you did, in a previous comment in the Fishing for Creationists thread, that the only anomaly is the divergent formation of lens crystallins, is so wildly wrong that I doubt you know the evidence in any depth.

In your anger at getting tripped up repeatedly over the last two days, you, who have have not even read through the papers you fatuously quote, is accusing Larry of being "so wildly wrong". Everything that Larry has written and even that you have quoted, makes clear what homology is as a conclusion and evolution is as an explanation. Nothing you have written supports your wild claims. Sadly for you this is a blog run by a scientist not some dissembling washed out prickly IDiot. And every time you play out your cheap parlor tricks here, you will be shown up for the charlatan you are.

Bilbo said...

Thanks for your answer, Larry. Paul, instead of waiting for Larry to agree to a debate, why don't you just write something up at ENV. I'm sure Larry (and no doubt others) would respond. And then you could offer a rebuttal, and Larry would offer one, ad infinitum.

Anonymous said...

As usual Paul the IDiot will scamper back to IDiot Institute's safe confines. There he will in the company of his fellow mountebanks declare how he quoted this, quoted that, dropped a few dozen names, and dazzled everyone at Sandwalk. Then he will wash his face, straighten its frozen form out of shame defeat and anger, and put on his best mask of David Berlinski inspired asinine assurance, and pen one link filled vacuous lie filled rant.

He's been doing it for years. Nothing new.

Truti

heleen said...

@Larry Moran
If those limbs are homologous we would expect to see a common ancestor with homologous genes for blood vessels, muscles, cartilage, nerves, bones, etc
Quite, we would expect to see. Now Larry Moran is correct: homology is not defined by common ancestry, but is part of the evidence for common ancestry.
NB: I agree with the German Wikipedia homology definition, the older and clearer one, rather than with the English Wikipedia that succeeds in making evolution something assumed before we know anything, rather than something we have evidence for.

Larry Moran said...

@heleen

Is the Panda's thumb homologous to your thumb and therefore evidence of evolution?

If your answer is "no," as it should be, then explain how you arrived at that answer.

Paul Nelson said...

As Larry has ignored my offer for an open debate on this topic at Evolution News & Views, I'm going to take Bilbo's excellent suggestion and start the discussion myself, with the comments open. Larry can then join in the comment thread if he wishes.

The post can't go up until Monday because the web staff at the Discovery offices is away for the weekend.

Heleen, if you're interested, Elliott Sober has written extensively about the problem of assuming common descent in definitions of homology. Please contact me at nelsonpa@alumni.uchicago.edu for pdfs.

P.S. to Jud: Yes, it may sound like the mad ramblings of a crazy person, but much of what biology "knows" about homology is wrong. That's the nature of discovery: one never knows what certainties will be overturned. More about this on Monday at ENV.

heleen said...

@Larry Moran
Is the Panda's thumb homologous to your thumb and therefore evidence of evolution?

If your answer is "no," as it should be, then explain how you arrived at that answer.


Larry Moran now has arrived at the correct position that homology is evidence of evolution, evidence for common descent.
First, in the 'PJ' line, Larry Moran defended that we should define homology as derived from common descent. That definition reduces evolution to an assumption without caring about the evidence for evolution.

And the Panda's thumb is the radial sesamoid, as follows from its position in relation to all the other handbones. Don't ask silly questions.

heleen said...

I wrote comments in the line too. http://sandwalk.blogspot.com/2011/12/fishing-for-creationists.html
In that line Larry Moran cited a textbook definition of homology that actually muddles understanding.

Glen Davidson said...

One thing about debating with someone as prone to misrepresentation as Paul Nelson, if you debate with him don't let him focus only on the edges and problem areas. This is what these pseudoscientists do, they don't acknowledge the implications of the non-difficult cases, they just keep raising questions where there are some issues.

Of course Paul is actually opposed to honest science altogether, as it comes to results that he denies.

It's probably best not to enter into any debate with such a discredited buffoon. If one does, however, one should insist that he'll give the "ID explanation" for vertebrate homologies (Behe's too, since his evolution takes place by miracles, hence the constant derivation isn't explained), and tell us in a meaningful sense how finding the results predicted by evolutionary theory shouldn't be understood as evidence for evolution. If he fails at either one, why should one go on?

IDiot Paul, you have never explained to me why the "macroevolutionary" patterns of vertebrate evolution and those of bacteria differ, apparently according to the known mechanisms of horizontal gene transfer in the latter and not in the former. There are two reasons you haven't, one being that you can't, and the other is that you argue totally in bad faith, without a speck of intellectual honesty.

Glen Davidson

Larry Moran said...

heleen said,

Larry Moran now has arrived at the correct position that homology is evidence of evolution, evidence for common descent.

Heleen, your responses are very annoying. You avoided answering the question.

Here's the correct answer. The Panda's thumb and Heleen's thumb are not homologous because they are not derived from a common ancestor with those types of thumbs.

On the other hand, Heleen's thumb and the thumbs of Howler monkeys are homologous.

The evidence of common genes and a common embryological/developmental pathway allows us to conclude that two structures are homologous. In other words, they developed from structures present in a common ancestor.

The evidence is gene similarity and embryological development (and perhaps other things). The consclusion is homology.

If you're a creationist, you could say that both structures were created independently and their similarity simply reflects a creator with a limited bag of tricks. According to creationists, no two structures can ever be homologous unless they redefince homology to mean "looks similar."

In that case, the Panda's thumb and Heleen's thumb are homologous to a creationist because they have no criteria for distinguishing between true homologoy and pseudohomology.

If you believe that evolutionary biologists are confused about the Panda's thumb, then you will agree with the creationists.

Larry Moran said...

Paul Nelson says,

Heleen, if you're interested, Elliott Sober has written extensively about the problem of assuming common descent in definitions of homology.

Here's a quotation from Elliot Sober's book Evidence and Evolution.

My focus has been on how a similarity (or a dissimilarity) that characterizes a pair of species provides evidence that discriminates between the common-ancestry and separate-ancestry hypotheses. Isn't this to ignore the fundamental biological point that it is homologies that provide evidence for common ancestry? There is a large literature on how the concept of homology should be understood, but the question in hand has in fact a simple answer. Homologies are usually taken to be be similarities that are present because of inheritance from a common ancestor; the wings of sparrows and robins are homologies in this sense. A homeoplasy, in contrast, is a similarity that is not due to inheritance from a common ancestor but instead arose because independent origination events occurred in separate lineages; the wings of birds and bats are an example. So defined, the concept of homology already has built into it the claim of common ancestry. If our goal is to test the common-ancestry hypothesis against the separate ancestry hypothesis by looking at data, then it would beg the question to say that our data consist of "homologies" in this sense. What counts as an observation in this problem must be knowable without one's already needing to have an opinion as to which of the competing hypotheses is true. This is why similarities are the right place to begin.

(Allow me to anticipate a quibble ... the wings of bats and birds are not homologous as wings but they are homologous as limbs.)

Anonymous said...

(Allow me to anticipate a quibble ... the wings of bats and birds are not homologous as wings but they are homologous as limbs.)

As I understand it, the wings of bats are homologous to human hands, while the wings of birds are homologous to human arms.

TomS

heleen said...

Larry Moran, your responses are very annoying. You avoided answering the only interesting question. The question is whether homology is defined as common descend or evidence of common descend. If homology is defined as common descend one cannot conclude to common descend from homology.
That is the simple point ‘if A is assumed, any pretence of concluding to A is void’.

Note that Larry Moran (Saturday, December 10, 2011 12:58:00 PM, this line) says The conclusion is homology.

Larry Moran started (Thursday, December 08, 2011 12:02:00 PM in the http://sandwalk.blogspot.com/2011/12/fishing-for-creationists.html line)
Here's the definition from Evolution (2009) by Douglas Futuyma.

Under the phylogenetic concept of homology, which is fundamental to all of comparative biology and systematics, homologous features are those that have been inherited, with more or less modification, from a common ancestor in which the feature first evolved. That is, homologous structures are synapomorphies


However, on (Thursday, December 08, 2011 3:46:00 PM, same line) Larry Moran said:
"Homology" is a conclusion based on evidence. The evidence is based on significant structural, or sequence, similarity, shared developmental pathways, and shared genes.
Evolution is the explanation for why structures (or genes) are homologous

Here I fully agree. However, if one has define ‘homology’ as derived from common descend, one cannot make this conclusion.

Then, in this line (Saturday, December 10, 2011 12:58:00 PM), Larry Moran says: The evidence is gene similarity and embryological development (and perhaps other things). The conclusion is homology. Quite, the conclusion is homology, and if you want it from gene similarity and embryological development rather than morphological position that’s OK to me.
And: The evidence of common genes and a common embryological/developmental pathway allows us to conclude that two structures are homologous. In other words, they developed from structures present in a common ancestor.
That is, homology as evidence from the structure leads to the explanation that this derives from a common ancestor. Here, homology is evidence and common descent the conclusion. Even for Larry Moran. That is, Larry is using two different ideas through each other.

I hoped Larry Moran would see his two statements are incompatible. The last one is valid.

About the thumbs: I know all that. Larry Moran should address himself to the reasoning he employs, not drift sideways. I’m not a creationist, and have not given any reason to suppose so.

Larry Moran said...

@heleen,

I'll stick with the definition in the best textbook on evolutionary biology.

Everything I said is consistent with that definition.

The only people who might disagree are those who insist on splitting hairs to make an obscure semantic point.

Anonymous said...

Heleen,

Larry is being kind in comparing your quibble to hair-splitting. The problem is more serious, you are being obstinate in your ignorance.

Larry has made it clear that homology is a conclusion and evolution is an explanation.

Quoting you,
The evidence of common genes and a common embryological/developmental pathway allows us to conclude that two structures are homologous. In other words, they developed from structures present in a common ancestor.

And then you add, mistakenly,
That is, homology as evidence from the structure leads to the explanation that this derives from a common ancestor. Here, homology is evidence and common descent the conclusion.

Homology as evidence from structure, it is NOT. Homology is not a conclusion from mere structural similarity as you are repeating mistakenly.

The IDiot's gambit on homology as an icon, is to accuse scientists of circular reasoning. You by talking of structural similarity = homology are just becoming fodder for the IDiot's mill.

Nullifidian said...

That's the creationist logic: if things differ, it proves separate creation, and if things are similar, it proves the existence of a divine plan. You can't possibly expect creationists to define their positions so that they are capable of refutation, otherwise they'd have nothing going for them at all.

Nullifidian said...

If homology is defined as common descend one cannot conclude to common descend from homology.
That is the simple point ‘if A is assumed, any pretence of concluding to A is void’.


Why not?

This is where creationists go off the rails. They assume that everything in science must be demonstrated according to the canons of formal logic. However, as any philosopher will tell you, formal logic cannot add to the scope of one's knowledge. It can only bring out what is implicit in the premises. E.g.:

All men are mortal.
Socrates is a man.
Therefore Socrates is mortal.

This only works because Socrates fits in the category of "men".

When attempting to expand one's base of knowledge, then, one must adopt other methods. In the hypothetico-deductive method, there is nothing wrong with positing that homology is the result of common ancestry and then using the existence of homology as evidence because homology is a prediction that flows from the hypothesis of common ancestry. If one said that homology was the result of common ancestry and then no homologous structures or proteins were ever found, that would put a major question mark over the hypothesis of common descent. But there are homologous structures, and these structures were identified as homologous even before common descent was ever proposed as a serious scientific hypothesis.

Jud said...

Paul Nelson writes:

Yes, it may sound like the mad ramblings of a crazy person, but much of what biology "knows" about homology is wrong. That's the nature of discovery: one never knows what certainties will be overturned.

The "nature of discovery" is intensive, extremely careful work by very, very smart people, often over a very, very long time. It is ridiculously easy to actually be a crazy person, or rather close to one in terms of being prone to mad ramblings, and to thereby produce 'inspired' crackpot results that differ from the established wisdom because they are wrong. It is ridiculously hard to study and work so very, very hard and meticulously with such extreme intelligence that one clearly understands not only the good scholarly work that has gone on for generations, but how to extend good, careful, correct work beyond present bounds to new discoveries. That is why true discoveries are so very, very rare, in contrast to the web pages of ID supporters where we read nearly every week about some new "discovery" or other supposedly overthrowing the established wisdom. All you've done is put crackpottery on an assembly line basis, turning out half-baked ideas by the dozens.

heleen said...

@Larry Moran
An appeal to authority? Rather than think for yourself?

Start from the beginning:
What is the evidence for evolution?
Homology , (other arguments).
Everything in paleontology and molecular phylogeny depends upon homology.

If one uses the definition cited by Larry Moran, where homology is defined as a consequence of common descent, evolution is included in the definition of homology. Therefore, under that definition, using homology as evidence for common descent is circular and not a valid argument. Therefore, most of the evidence for evolution vanishes. Evolution is made into an assumption rather than a scientific theory.
The cited definition (rather a popular one, I know) is misleading in a way that is easily remedied by sticking to the previous usage of homology. Larry is using two different ideas on homology through each other, one based on the definition that includes descent and one based on the valid older definition based on structural relations.

I’m not quibbling: if one deals with creationists one cannot afford having no defense against the accusation that one assumes what one wants to prove. Concise reasoning is of importance.

Enrique said...

heleen, YOU are on circular reasoning, not others, due to your asumptionof calling "homoly" to mere morphological or functional similarities.

Take the wing of a a bat and a bird. ¿Are they homologous? They are similar. Is their resemblance proof of common descent?

No, mere similarity is not proof of evolution. You investigate further, seen inner details of structure, its developement, the genes involved and son on. If after that study you can rule out the posibility that such wings are similar by chance, if you have proofs that the explication of teh similarity is coomon descent, tehn you cal it momlogy.

A sentence like "Everything in paleontology and molecular phylogeny depends upon homology" is then not circular reasonig but a paraphrasis of "Nothing in biology makes sense but in the light of evolution". It means that Paleontology cannot be a mere describing discipline, an study that simply realizes that some animals have parts similar to others. And use those parts for classification.

A lot of laughable classifications by Linnaeus, Cuvier and others are due to use superficial traits for classifications. That sentence warns that Paleobtology need to to better than mere look for superficially similar things.

When you engage into if a trait is secondary, or adquired by convergence and collect evidence about it, then you are doing paleontological science, otherwise you are collecting stamps. Thats the meaning of that type of sentences. No circular reasonig at all. They are assertions that common descent evidence has been collected and discussed.

They are simply language-saving constructs. How do you now some traits are homologous? Why do you accept these hypothesis and reject those? You can explain that evidence in several pages.
A short way to express that is to say "A and B are hologous". That sentence is NOT the proof by itself, nothing substitutes for teh pages of evidence. Similarity alone is not evidence. "A and B are hologous" is a conclusion not a proof.

"Homology is evidence for evolution" is not circular reasoning, is a short way of saying that you need to study a morphological similarity to rule out any other explanations. You need to investigate the detailed morphology, embryology, developental, molecular, physiological evidence you present are accepted as proof. That's the proof, all that collected evidence, the word "homology" is shorthand for that (claiming homology without proof is vane).

Paul Nelson said...

My post (on homology) at ENV has to wait in line behind previously-scheduled material. The post will have the title "Take the Homology Quiz," will link to this and other Sandwalk threads, and will be open for comments.