I see the phrase used from time to time in debates on evolution vs creation. I don’t want this thread to be such a debate, but I don’t understand the buzzwords when it is explained.
I merely would like the concept explained to a nonbiologist using nontechnical terms, along with a noncontroversial explanation of why the concept of protein homology might necessarily apply to one model but necessarily not apply to the other.
Homology basically means similarity. When you look at protein homologies, you are comparing two different protein samples by looking at how similar the amino acid sequences are. Think of proteins as really long words. Words are made up of a combination of different letters, proteins are made up of different amino acids. Each amino acid can be represented by a letter. That being said, lets look at three made-up proteins:
Protein 1: GHTYUIOSKA
Protein 2: GHUFGIOSKA
Protein 3: HGGASOIKUF
Proteins 1 and 2 have a high homology. They are very similar in the sequence of amino acids. Both sequences have a G in the first position, and H in the second, etc. Proteins 1 and 3 have a low homology. The amino acid sequences have little (in this case, nothing) in common.
If two proteins have a high enough homology, you can hypothesize that they are related in function. Homologies can also be used to propose or further support evolutionary relationships. Human proteins, in general, share a high homology with chimpanzee proteins. They also share a high (but smaller) homology with dog proteins. The idea is that the more closely related two species are, evolutionarily speaking, the higher homologies their proteins should have.
If you could give specific examples of when homologies are applied in one case but not another, I’ll try to answer why that was done without turning this into a GD thread. 
Thanks, Alphagene.
Would it also be reasonable to expect higher homologies because of taxonomic similarities that have nothing to do with ancestry? For example, the word “niggardly” has nothing to do with any racial slur, despite its high homology with a word that does.
All I have is this, from Where’s the evidence against evolution?.
[ul]
[li]For example, some creationists explain protein homology on the grounds that “similar organisms need similar proteins.” If creationism were true, we might find that similar organisms did indeed have similar proteins. More likely, we would find nearly 100% sequence conservation, since God would use the same hemoglobin to serve the same purpose across all species.[/li]
But from what you say, Alphagene, similar organisms do have similar proteins.
[li]How do you explain protein homology[/li]
With no attempt to explain it himself in his own preferred model. That same question is repeated numerous times, until finally this:
[li]Please, Joel, enough! I asked you a question: how do you explain protein homology from a creationist standpoint? No answering questions with questions, no irrelevant quotes, just find me a valid creationist answer or admit that creationism can’t provide one![/li]
A very unfair question to put to a nonbiologist of whatever belief system, especially since the author never explained how it fits his own model, other than:
[li]Protein homology is a different issue: evolutionists have an answer that fits all the facts and makes powerful predictions, while creationists don’t. Protein homology leads to new discoveries, when evolutionists explain it. Does “God did it” help to explain the functioning of proteins?[/li]
Little more than a gratuitous assertion, and a meaningless rhetorical question.
Finally, I said this:
[li]I don’t know protein homology from protein heterology, but I do know that replacing religion politics with science politics is not an improvement. I wish both sides would, for the benefit of those of us less knowledgable in these matters, … elucidate more.[/li][/ul]
And so I came over here, looking for answers.
“Would it also be reasonable to expect higher homologies because of taxonomic similarities that have nothing to do with ancestry? For example, the word “niggardly” has nothing to do with any racial slur, despite its high homology with a word that does.”
Not necessarily. Two proteins can serve a similar function and have different sequences. In this case they are called analogous proteins. This would be similar to your example of niggardly in that it looks like nigger (same function in a protein) but has a different meaning (different sequence of the protein) and a different origin. The idea behind protein homology is that the different proteins are coded for by genes that had the same origin, but over time the genes mutated and thus evolved differently in the different organisms. They still have similar sequences and often similar functions, but are different in detail even though they had the same origiin.
Thing 1
Thanks. Is there any way to attribute protein homology to ancestry without begging the question? In other words, without saying, “If they are of the same origin, they should be homologous. Since they are homologous, they are of the same origin”?
Evidently not. Thanks all.
Being a biologist by training, I would think that two similarly related species would have similar protein homologies. That said, however, the protein homologies are probably what defines that similarity. Otherwise you have some scientist sitting in a room with a dead bird and a dead bat saying “Gee, they both fly, have similar diets, and are about the same size, so why are they so genetically different?”
I also think you can have two species similarly related that can have vastly different proteins if the protein gives one species a tremendous selective advantage the other doesn’t have such as disease/ weather/ temperature resistence or greater yield of milk, fruit, etc.
Finally, going back to the bird and bat example. You can have some features evolve more than once in evolution such as flight. Likewise, you often will see amino acid sequences of proteins that are vastly different (i.e. have very low homology) yet have a similar ‘profile’. That is, they form a similar 3-D shape and perform a similar function. I’m at a loss to provide a pretein example for this, but I know they exist based on what I’ve read.
Thanks, Yarster! But I might put bats taxonomically closer to mice than birds. I’m not sure why, but they just look like flying mice to me, sort of like “flying” squirrels.
At any rate, I would think the protein homology phenomenon would fit any arbitrary hypothesis that posited any sort of DNA “theme”, whether that would be ancestry, design, or some combination.
Then let’s see you explain protein homology from a YEC standpoint. Frankly, it seems to me that if you make a prediction (if species are descended from each other, you will see certain patterns in their DNA) and that prediction is correct (the patterns turn out to really be there) that’s the scientific method. Do you consider the scientific method to be a tautology?
Incidentally, Lib, I think you’ve grossly misrepresented me. Firstly, I didn’t demand that Joel explain protein homology all by himself: I told him that even if he asked creationist experts, no matter what their credentials were, he would be unable to find a creationist explanation of protein homology. Moreover, you have rather unfairly quoted me as making comments which I made when I thought Joel was evading the question. When I found out that he had, in fact, made a perfectly appropriate reply, I apologized. I really don’t see why you’re dredging up statements which I apologized for, unless, of course, you’re trying to paint me in an unfair light.
Secondly, I did indeed explain protein homology in layman’s terms, despite your false accusations to the contrary. Moreover, I did explain how evolution fits the observed facts- again contrary to your false assertions. Thus far, your commentary in this thread suggests that you haven’t read my explanations, since you are asking questions which I already answered in some detail.
-Ben
No, sorry, Ben. I did not intend to disparage you in any way. In fact, I didn’t even recall that you were the inquisitor. I merely responded to Alphagene’s request for context. But you’re right that I missed your explanations, both when I read the thread originally, and when I searched it for Alphagene, using the phoneme “homol”.
To answer your questions, yes, I think the scientific method is a tautology (as are all epistemologies, including scriptural ones). And no, I can’t explain protein homology for creationism, since I know very little about it. But as I said before, it makes sense to me that a designer might use a theme rather than an ecclectic pallet.
**
Thanks, Lib- apology heartily accepted!
I think that the problem here is that when you ask us if we can use homology non-tautologically, most people would interpret that to mean:
This is very different from:
Protein homology is only tautological in the way that creationists try to use it: they state that similar animals have similar proteins, and that genetic similarity reflects anatomical similarity, but if there’s a conflict, then the genetic similarity means that everyone since Linnaeus has made a mistake on the anatomical similarity.
As used by evolutionists, there’s no tautology. On the basis of familial relationships observed in the fossil record, one can make predictions about what sort of homologies one will see in genes, and the genes bear out those predictions even when they conflict with the predictions one would make on the basis of overall anatomical similarity. Such a use of homology is only tautological to the extent that you (ie Lib) believe that logic and common sense are based in circular reasoning, but we certainly don’t have a “you date the fossils from the rocks and the rocks from the fossils” situation here.
-Ben
Just to add some things to the mix, on the issue of why a creator might use similar themes, here’s a point–not conclusive, but certainly some evidence.
Certain bases in DNA are more succeptable to mutation that others–they’re just more reactive. If a base gets changed, and amino acid may get changed (I know this is oversimplifying–there’s a degenerate code, etc). If an amino acid changes, there’s a good chance the proteing changes. So if we see mutations that are likely to happen, that’s also evidence. A creator doesn’t have to deal with that; you can just put whatever would be usefull in there.
So proteins can offer an insight into this–some changes are more likely than other…
I should also point out once again the problem with the whole “you don’t know how God would create” argument:
FOR THE UMPTEENTH TIME (drumroll)…
You can’t just wave your hands and say “God did it”; doing so is no different from last Thursdayism. If you want to posit God as a scientific hypothesis, you must be specific about what that means.
The reason I critique certain assumptions about how God would create is not because I believe anything about God- instead, these assumptions have been put forth by creationists, on the rare occasions when they delineate the concept of special creation in any detail. I did not originate the concept that God would create similar proteins for similar animals: essentially all creationists who offer an opinion on the subject say that protein homology can be explained using that assumption. Furthermore, my criticism based on segregational load comes from the idea that God created a perfect creation, and that disadvantageous mutations occurred after the Fall. Segregational load cannot be explained under such a scheme; I claim, therefore, that segregational load is evidence against that sort of creation “science” (and I additionally claim it as evidence for evolution, in that it is the sort of process one would expect from evolution, and additionally it sheds a lot of light on the processes of evolution.)
-Ben
Errr, did you read my post? I’m not sure if you’re pointing out (correctly) that I don’t know how God would create or if you think (incorrectly) that I’m arguing for creationism; just curious?
My point was that things happen in ways that are more consistent with random mutation (e.g. more easier mutations vs less probable one) than with design. Of course I’m basing it on my understanding of rational design, which is why I’m not just waving my hand.
Lib, my understanding of the word “homology” is a little different than what was described above. The word was originally used by biologists in speaking of morphological structures, and refers to their similarity deriving from a common ancestral structure. For example, the bones in your hand have counterparts in the wing of a bat, and since the reason for this is that we have both inherited that bone pattern from a common ancestor, these structures are said to be homologous. Compare that to the relationship between a bat’s wing and an insect’s wing. These structures serve the same purpose (flight) and are “analogous” structures, but they are not homologous. On another note, the results of convergent evolution can produce structures of great similarity, like the eyes of vertebrates and cephalopods for instance, that are NOT homologous (hence homologous and similar are not really synonyms). Sometimes things that don’t seem very much alike, for instance your hand and a whale’s fin, turn out to be homologous when looked at closely - dissect a whale’s fin and you’ll find a whale version of every bone in your hand.
In the case of proteins, the sequence of amino acids is inherited through genetic mechanisms I’m sure you already know. But since a lot of mutations can take place without altering protein structure or function (that is, with one amino acid is substituted for another), you will find subtle differences in the detailed sequence of amino acids in the same protein in two different species. If the branching of the two species was relatively recent, you see high homology, expressed as a very similar amino acid sequence; if it was further back, you see less homology. What’s more, you can compare the sequences of proteins for many different species, and based on homologies put them into different biological classes by how similar they are. LO! These classes correspond, in most cases, to those deduced from morphology. (The relatively few cases where there are disagreements have been interesting in their own right, and have led to further investigations and discussions. But that’s a long discussion.)
A technical detail: Amino acids that are highly important for structure or function of the protein cannot mutate to something else so easily. So homologies look at differences in mutations that are relatively silent at the protein level. If you posit a designer creating life immediately, there is no chemically compelling design reason for these silent amino acids to be anything but random (I’m simplifying slightly, they could be semi-random). But the fact that they are conserved across species is an indication that those species derive the genetic information specifying these molecules from a common ancestor: They are homologous proteins.
At any rate, Myrr’s point is a good one: evolution allows us to predict the general behavior of the sequence homologies we see; creation allows us only to discover them empirically - it lacks the explanatory power of evolution.
Hi. A few words before this descends irreversibly into the GD realm.
I like this last Thursdayism. It falls well enough outside of any testable hypothesis that it works as a legitimate alternative to evolution. The problem with biblical creation is again, it implies certain things about the world around us. Science has done good at predicting things about the world around us, and time and time again, creation has lost out. Last Thursdayism makes no predictions, therefore cannot be disproven.
Next, protein homology. There is an added dimension here, and I believe it is the one that implies time’s arrow : Biochemistry and gene function.
Take some protein sequences (dachshund in Drosophila melanogaster vs. man):
YNSPPPISTSDPSANECKLVEYRGQKVAAFIISNETMLCLPQAFELFLKHLVGGLHTVYT
Y++P P+ + P NECK+V+ RG KVA+F + ++CLPQAF+LFLKHLVGGLHTVYT YSTPSPVENT-PQNNECKMVDLRGAKVASFTVEGCELICLPQAFDLFLKHLVGGLHTVYT
KLKRLDIVPLVCNVEQVRILRGLGAIQPGVNRCKLLCCKDFDILYRDCTTA
KLKRL+I P+VCNVEQVRILRGLGAIQPGVNRCKL+ KDF+ LY DCT A
KLKRLEITPVVCNVEQVRILRGLGAIQPGVNRCKLISRKDFETLYNDCTNA
SSTETLLRNIQSLLKVAADNARQQERQISYEKAELKMDVLREREVKDSLERQLVDERKLR
SS ETLL NIQ LLKVA DNAR QE+Q+ EK ELKMD LRERE++++LE+QL E+K R SSIETLLTNIQGLLKVAIDNARAQEKQVQLEKTELKMDFLRERELRETLEKQLAMEQKNR
Top lines=Drosophila, middle lines=similarity, bottom lines=human. Letter for identity,+ indicates similar amino acids.
We know that they do the same type of job. Better yet, we can take the human copy, stick it in Drosophila, and it does the same thing still. We also can introduce mutations in the gene that mutate specific codons encoding amino acids, and we can see which ones specifically knock out gene function. These codons must be the most important ones. Through phylogeny these are also the ones that are most conserved. At this point, we are not past the assertion that God uses similar proteins for similar functions.
Now compare to a species closer to human (mouse) and we see a sequence more similar to humans. Now compare a species closer to melanogaster (D. simulans) and we see closer to D. melanogaster. And yet, all of these rescue mutants in Drosophila – the protein has identical function. To me, this doesn’t fit with the creationist “similar proteins for similar functions” – these are similar proteins for identical functions.
This is not an isolated incident in biology, either. Many genes look like this. The proteins are all variable, but serve the same function. The padding in the protein has changed more than the “meat” - active site, important structural motifs. It doesn’t fit the “similar proteins for similar functions.” I cannot second guess God, but I see little reason why it is advantageous to shuffle the padding around while keeping the core the same, if it is really all for naught – you end up with 2 proteins that act exactly the same.
Without function and biochemistry, protein homology is nearly useless. That is how we know, in biology, the equivalent of whether a word means stingy or is a racial slur.
The dachshund lines came out a bit funny. Lemme try again.
YNSPPPISTSDPSANECKLVEYRGQKVAAFIISNETMLCLPQAFELFLKHLVGGLHTVYT
Y++P P+ + P NECK+V+ RG KVA+F + ++CLPQAF+LFLKHLVGGLHTVYT
YSTPSPVENT-PQNNECKMVDLRGAKVASFTVEGCELICLPQAFDLFLKHLVGGLHTVYT
KLKRLDIVPLVCNVEQVRILRGLGAIQPGVNRCKLLCCKDFDILYRDCTTA
KLKRL+I P+VCNVEQVRILRGLGAIQPGVNRCKL+ KDF+ LY DCT A
KLKRLEITPVVCNVEQVRILRGLGAIQPGVNRCKLISRKDFETLYNDCTNA
SSTETLLRNIQSLLKVAADNARQQERQISYEKAELKMDVLREREVKDSLERQLVDERKLR
SS ETLL NIQ LLKVA DNAR QE+Q+ EK ELKMD LRERE++++LE+QL E+K R
SSIETLLTNIQGLLKVAIDNARAQEKQVQLEKTELKMDFLRERELRETLEKQLAMEQKNR
Meant of course, to show the homology evident over many 100s of millions of years of evolution, if the gene is important. Notice the stretches here of little homology. Also, these are the most conserved stretches of a 1000 amino acid protein, the rest are nearly random when compared. And yet, exactly the same function in Drosophila.
Thanks again for the great patience and for simplifying things a bit. Maybe what biology needs is an Einstein, i.e., someone who can relate both to scientists and the public at large — someone who can talk to us in terms of elevators that fall and twins who meet paradoxically.
One more question, please. Does the protein sequence (the string of letters) change within a species over time?
Hard to say, really. We have no access to protein sequences from animals which are not currently alive. Since we have only been sequencing proteins for under 75 years, this is far too short to notice any sustained change of protein sequence.
Note, this means all molecular phylogeny is implied from currently living organisms.
It is clear that there is something called genetic polymorphism. This means that the same genes look slightly different between individuals in a same species. Polymorphism technically refers to all types of change (the good, the bad, and the neutral), but is used commonly to refer to neutral change – change not associated with a phenotype, or silent mutation which doesn’t even change an amino acid.
I’ll use the technical definition. Polymorphism is found all over the show, and is probably what causes individual variation within a species. The hypothesis is that a small population with an polymorphic individual or two moves into an isolated region. Conditions change, and the polymorphism becomes selected for. Repeat a bunch of times, throw in genetic changes based on genomic reorganization, and you can possibly lead to a situation where the new population cannot interbreed with the old guys.
Luckily, the world has seen fit to give us a couple of writers who can do just that.
Stephen Jay Gould and the late Isaac Asimov.
Two great thinkers who can explain the most esoteric science in language that High School kids can understand whil not talking down to anybody.