I think I see what you are saying. I’m not familiar with how similar my genetic code is to a fungus or whatever. If there are large traceable DNA code blocks between all species, then clearly they must all come from one ancestor.
I’m not a geneticist, I’m a chemist. Chemically, I see no reason why only one type of genetic code system survives from multiple events. In fact, it’s precisely the type of behavior I would expect from a self-assembling system.
I think I see what you are saying. I’m not familiar with how similar my genetic code is to a fungus or whatever. If there are large traceable DNA code blocks between all species, then clearly they must all come from one ancestor.
I’m not a geneticist, I’m a chemist. Chemically, I see no reason why only one type of genetic code system survives from multiple events. In fact, it’s precisely the type of behavior I would expect from a self-assembling system.
They are identical.
I’m not clear on what you mean by a “code block”. If you mean genes, then those have changed in clearly traceable ways from species to species. It’s a trivial problem now to calculate phylogenies of gene sequences. If you mean the actual genetic code itself, there’s only one, and it’s stayed completely the same with the exception of a few trivial counterexamples.
I’m not parsing this sentence at all, but you seem to be agreeing with the main point - that the best evidence is that this all happened once - which is good?
Think of it this way: A book by Tolkien is very different from a book by, say, Douglas Adams. That would be the human and fungus in your example: The contents of our genes are both the same. But Tolkien and Douglas Adams both used the same dictionary: That’s the genetic code. As a result, a literature critic would certainly say that Adams and Tolkien were both writing in the same language. If, instead, you compared either one to Cervantes, you’d find it was a different language, but in the world of biology, we don’t see any Cervantes.
blah blah blah
You know what I meant. You demand very strong evidence to be convinced of a new position, while refusing to really examine how strong the evidence for the current one is. It is the essence of conservatism.
My post was a cute argument from definitions. “Incontrovertible” means “impossible to deny or disprove.” You say new evidence must be incontrovertible. But if it is accepted as such, then automatically the old evidence was rather more probabilistic. QED, you demand that new evidence be held to a higher standard than the old. I’m upset you didn’t get this little proof. 
I think we can all agree, based on Chronos’ argument, that all RNA/DNA-based protein transcription that we have hetherto examined came from a single innovator.
That doesn’t exclude that other aspects of our metabolism/cell structure drifted horizontally from other biogeneses around the time of codon innovation (or later). It doesn’t exclude that non-RNA/DNA based life is crawling deep inside the earth. And it of course it doesn’t exclude that different-codon life may yet be found. (The last proposition may be said to have the least likelihood, as it’s been subject to the most rock-turning. In fact, the vast majority of rock-turning.)
Let us be careful not to frame the question in such a way as to fit our answer.
But if you want to demonstrate that there are other types of life, you have to demonstrate that there are other types of life. And as everyone has said, this has never been done. It’s certainly possible that there are some weird bacteria-like organisms that arose idependently of all other life, but you have to actually find them.
In what way? If they were identical, then I would grow in decaying matter and taste good on pizza. The fact that all of our DNA uses the same four nucleotides is exactly what I’m stating would likely have resulted from multiple abiogenic events. Different nucleotides simply aren’t likely to produce a surviving genetic code even assuming they can be synthesized naturally in the first place. The codons for each amino acid are the same between species because that is what the codons do. I don’t see where probability works in at all there. DNA is transcribed in a chemical reaction. The end result wont change no matter where the DNA came from. If TGC produces cysteine how could it produce anything else in a different species?
You appear to be using the term “genetic code” in a way that is far different from the generally accepted scientific definition. When I say “genetic code” I’m referring to the correlation between codons and associated amino acids.
There seems to be a gap in your knowledge here. It’s not a simple chemical reaction where codon + whatever -> amino acid. There are tRNAs and aminoacyl-tRNA synthases involved. Basically, the ribosome helps match a tRNA with a codon on the mRNA. The tRNA has an anticodon that pairs up with the codon, and it also carries a specific amino acid, which is then transferred onto the growing protein. The genetic code essentially boils down to which amino acid is attached to which tRNA. That association is performed by a group of enzymes called aminoacyl-tRNA synthases. The enzyme recognizes a specific anticodon and a specific amino acid and hooks the tRNA to an amino acid.
The key point is that as far as we can tell, this association is completely arbitrary. Experiments have been done where these synthases have been mixed and matched to change the amino acid-anticodon pairing, and the system still works just fine. The best evidence we have suggests that this code is a “frozen accident”, and one of many possibilities that could have arisen.
You are evidently not understanding what is meant by the genetic “code.”
The genetic code is analogous to Morse code. Morse code uses three symbols (dot, dash, and space) to represent the letters of the Latin alphabet. These symbols are analagous to the four nucleotides, while the amino acids are analogous to the letters. Which combinations of dots and dashes code for which letters is arbitrary (although Morse gave more common letters easier codes). A different code would use different combinations of dots, dashes, and spaces to correlate with each letter. There are an extremely large number of such possible codes. The genetic code differs in that there are four symbols, rather than three; and that all codons have three symbols, rather than being variable in length as in Morse code.
The odds of two identical genetic codes having arisen by chance is analogous to someone in South America having independently originated a code exactly the same as Morse code, with the same combinations of dots and dashes coding for exactly the same Latin letters, without ever having heard of Morse code.
In a different code, TGC would not code for cysteine, but for lysine, or something else.
But wouldn’t that require a different set of enzymes to interperet it that way?
Yes, or rather a different variety of transfer RNA (if independently evolved life even used the same elaborate transfer system to decode the genetic system). I’m not sure what you are getting at.
Why are you so certain that different enzymes would be as stable as the ones we have now? You can’t just rearrange things and expect that they will all work just right. Small changes in a chemical structure lead to large changes in reactivity. I don’t think that you can just change the enzyme and expect it all to work just so. I think that after the first billion years a thermodynamic minimum would be reached.
Yes, but why would the exact same enzyme evolve to connect the exact same tRNA codon to the exact same amino acid?
As Colibri says, the codes for the amino acids are completely arbitrary. There’s no reason for UUU and UUC to code for Phenylalanine, while UCU, UCA, UCG, AGU, and AGC codes for Serine.
Take a look at this table: Genetic code - Wikipedia
There are some non-arbitrary properties of the table, since multiple codons that code for the same amino acid are typically one letter substitutions for each other. But otherwise, there are 64 possible combinations of triplet codons, and there is no chemical reason for any codon to have any relation to any amino acid.
And this code is identical in every single species, except for a very few that have one different codon: Genetic code - Wikipedia
All living things use the same alphabet to code for genes, even though different organisms have different genes. So suppose you find a book written in the Latin alphabet. Even if you have no idea what language the book was written in, even if you have no idea what the words mean, you still know that the author of the book didn’t independently invent the Latin Alphabet by coincidence. And we see the same thing in living organisms. There is one alphabet, despite all kinds of different documents/genes. And there’s no way that is a coincidence, or somehow the only way it could have happened.
With all due respect, you don’t seem to be grasping the complexity of the issue. The system of transcription of DNA and its translation into protein involve not one (or twenty) enzymes, but multiple steps, including messenger RNA, various transfer RNAs, ribosomes, and several different enzymes. The mechanism differs somewhat between prokaryotes (bacteria and archaea) and eucaryotes (other organisms), and there are many different variations on the RNAs, enzymes, and other components. Yet all of these variations result in the translation of the same set of codons to the same set of amino acids. Because the components of the steps vary between organisms, it’s basically impossible that the relationship between codons and amino acids is due to some “thermodynamic minimum.” .
Because that is the most stable system. It would be pretty extraordinary if it weren’t after a billion years. A different enzyme would have a different stability. It would have different reaction rates and different selectivity. I just don’t think you can arbitrarily change enzymes and expect all the other variables to remain the same. It never works that way in the chemistry world that I know, and that’s with a timescale of hours. With a timescale of a hundred thousand years, the slightest difference in reaction rates would lead to massive changes.
So after a billion years, I expect multiple abiogenic events to produce exactly the same enzymes. This isn’t some extraordinary coincidence. It’s precisely how a self-assembled system behaves. Each piece is dependent on the other pieces. It may even depend on earlier enzymes that don’t work as well, but after a billion years the conclusion is the same.
To be clear, the claim isn’t that all living things have an identical set of ancestors, just that there is a common ancestor. Whatever proto-critter first came up with the genetic code we use now would be such a common ancestor.
WarmNPrickly, the problem is that, first of all, we’ve experimentally produced the appropriate tRNAs for other genetic codes, and they seem to be just as stable as the current set, and second, evolution doesn’t do global maxima, just local ones. Or rather, it does a whole bunch of local maxima, which may or may not include the global one. And the vast number of possible genetic codes leaves a heck of a lot of room for local maxima. You might as well ask, if cockroaches are so successful, why hasn’t everything else evolved into a cockroach: It’s because to get from human to cockroach, you’d have to go through a whole bunch of intermediates that are inferior to either.
As I pointed out above, there isn’t a single set of enzymes (or other components such as the RNAs), but many variations. Given differences between species, there may be millions of different pathways that yield exactly the same result. So this argument fails rather decisively.
Well yes. In fact I’ve seen weird things like a semi functional triple helix made by just changing the phosphate bonding positions. Of course we aren’t going to see stability differences on a hundred thousand year scale in the laboratory.
No doubt. Hopefully I will figure where the problem is soon.
But that’s not how biological systems work. Did you even read the wikipedia article?
You aren’t appreciating just how many possible systems there could have been. Four billion years of evolution is pretty small compared to–well, let’s see. How do we calculate this?
There are 64 possible triplet codons. There are 20 amino acids, and one stop. Therefore, there are 21^64 possible combinations…arg, I can’t figure out how to convert that to regular scientific notation. But that number is a lot bigger than 10^64. If we round 4 billion years of life up to 10 billion years, that is only 10^10. 10^10 is very very very very small compared to the number of possible permutations of codes. We’re looking at 10^54.
We’re not just talking about most creatures share the same code, we’re talking 1 and exactly 1 code. When you look at actually existing organisms, there just isn’t this sort of convergence to one and only biochemical solution, because organisms live in very different environments. So you look at a protein like Hemoglobin, and we find that while every mammal has Hemoglobin, the exact amino acid sequences that make up that Hemoglobin vary tremendously. When you’re looking at convergence between a couple of possibilities, then it’s certainly true that there might be one way that’s better than the others. But if there are 21^64 different possibilities? That’s just ridiculous. You don’t understand how large 21^64 is. It’s so much bigger than a billion that the difference between 1 year and a billion years is just a rounding error.