How similar would alien life be to us at the molecular (DNA) level?

[Shamol]

So yesterday there was a rather heated discussion on discord on this topic. The question itself has lots of assumptions built in, but it's still interesting if for nothing else then just to understand the basic aspects of molecular biology.

Let's say life is found on some other planet. Let's say this life is similar to us in functional terms- it replicates by autocatalysis, it has an energy harvesting system so it can function against entropy, different replicating entity compete with each other which drives evolution, it progressively becomes more complex thereby accumulating a lot of organized complexity (these are plausible assumptions about any life in the universe, although I understand still arguable), and in other ways as well. Let's further grant that this life, at the molecular level, is based on DNA, or in the least some other equivalent molecular alphabet system that allows us to compare their DNA with ours (in the same way we compare human-chimp DNA). All of these assumptions are arguable and require individual threads on their own right, but let's grant them nonetheless.

So with these assumptions in place- how similar would our DNA be with alien DNA?

One argument for high similarity can be given in terms of functional similarity. In addition to replication, metabolism, evolution, increased complexity, having a developmental program and such, it's plausible to think life on earth and life elsewhere would have more functional similarities. Not only would we both require energy, but perhaps we would have similar sources of energy and similar energy harvesting mechanisms. Maybe their biological context would depend on molecules like water and lipid as well, which would mean functionality have to evolve with these features in the center.

Even if we grant all of this though- it's virtually impossible that our DNA would be even remotely similar to alien DNA. And the reason can be given in one statement- Biological codes are linguistic/symbolic in nature, and language is arbitrary.

Just to give one tiny example, here's the basics of how DNA codes for protein. Simplistically, three DNA bases (letters- ATGC) comprise a "codon", and each codon encode a single amino acid. There are four letters (=4^3=64 codons) and 20 major amino acids. So that means there's a grammar to this process- there's a set of rules to what codons make what amino acids. This set of rules is called the genetic code.

Here's the issue: this code is completely arbitrary. There's no physical/chemical rules that determines a codon specifying any particular amino acid. The "translation" process between the two languages occurs via a set of mediator molecules called tRNA, which bind to a codon on one side and an amino acid on the other:

The whole process and DNA-amino acid correlation is, again, arbitrary. Evolution used whatever code it hit upon. So even if alien life had proteins, and even if their proteins were made of amino acids, and even if they had twenty of them, and even if the translation process was similar, there's absolutely no reason that their DNA-amino acid translation code would be similar to ours. There's just no physical or chemical determinism to appeal to to make that case, it's all historical contingency and evolution making use of whatever it chanced upon.

Not just DNA, biology as a whole is awash these sorts of arbitrarily arranged codes (for the cell/molecular biology folks- signal transduction is another example. The first-second messenger molecules' correspondence is arbitrary and mediated by a receptor, without any sort of determinism). So yeah, there's no reason why alien DNA would be remotely similar to our DNA.

There is one caveat to this that I can think of which I might mention later.

 

[Dracowyn]

Just like I said in the chat, I doubt the DNA of aliens (if some alien trees of life do have DNA similar in buildup as ours) would have anything in common with ours. Simply because they'd have no relation to any species on Earth. The DNA of every species alive today is the result of billions of years of evolution, from the first simple cells that had DNA back in the Archaean to modern life nowadays there have been countless of mutation that allowed evolution to happen.

Even if alien life would have some relation to us, as in what if the panspermia theory is correct and life does spread through the universe from one planet/moon to the other and share a same origin, the billions of years of evolution on both planets would probably result in the DNA barely sharing anything.

For an instance, we do share 60% of our genes with bananas. But only 18% with baker's yeast and only 2% with E.coli while those are both earth species from the same tree of life like we are.

That said, morphologically, there could be similarities through convergent evolution. This happened numerous times on Earth, non related species living in a similar environment, having a similar lifestyle and eventually starting to look alike. Some examples are the extinct ichthyosaurs and modern day dolphins (ichthyosaurs being reptiles while dolphins are mammals). Placodonts and turtles (while not being closely related, but both groups still being reptiles, they looked so much alike it's hard to tell them apart) and bat wings and pterosaur wings, both evolved from skin between the fingers (in the case of bats between all 5 fingers, in the cause of pterosaurs an elongated 4th finger and the wing membrane connecting the tip of it with the body).

 

[Kylie-chan]

At a molecular level, not at all unless RNA somehow spread very far, and you've articulated the genetic/molecular bio reasons why not very well. However, your philosophical reasons are wrong: our concept of language is based on what we ourselves know and has nothing to do with the chemical properties of amino acids. We saw an analogue between language/code and between the functionality of amino acids and symbols (which are pretty broad) and assigned them that type of conceptualisation. Our concept of language has nothing to do with whether alien DNA might be analogous or not chemically, let alone abstractly.

To illustrate, and this is something I'll briefly add my two cents on: atomic elements (or even subatomic... again, the way we describe compounds is partly rooted in convenience and conventional linguistics) themselves can be viewed as morphemes. Meaning anything chemical can be a language, really. What is the genetic code except carbon chains? It's that very arbitrariness you hit on that debunks a philosophical argument for similarity but supports the chemical and biological arguments for similar properties. Evolution is arbitrary but not random, because some things are likelier to work than others.

At an atomic level? I think it's almost certain to be carbon-based, at least in the universe we've observed. In which case there might be some superficial or even meaningful similarities in composition and chemical properties. Silicon is really the likeliest substrate (for a very non-complex and self-similar organism) and it simply lacks the versatility and immense reactivity of carbon, and the environment would have to be lacking in some very common universal elements for the organism to thrive.

Weirdly I was having a discussion with someone recently about this also.

 

[Nihilego]

This entire discussion works on the assumption that alien life would use DNA, which is in my opinion very unlikely to be valid.

DNA alone is not enough. In fact, the sugars from which DNA is assembled (on Earth, at least) require processing by protein machinery before their polymerisation into a long DNA chain. Beyond this, its maintenence, regulation, and activity in transcribing proteins is all handled in itself by other proteins. The real question here is if "aliens" are protein-based, and if so, if these proteins are able to assemble and handle DNA at all. This seems enormously unlikely to me; DNA is an extremely stable polymer but there are probably hundreds of thousands of conceivable molecules which could form code-like polymers given the protein machinery to support them.

So the real question here is if alien proteins, presuming they use them (again - an assumption which may not be true, although which in my opinion is more plausable than the DNA assumption), are of similarity to those found on Earth. The answer to this, in my opinion, is no - probably not. If you understand that DNA, which uses only 4 bases, is complicated then when you consider that proteins contain 20 different amino acids, the order of which dictates their structure and function, you might begin to realise just how many ways these things can work. The assumption that they'd just so happen to utilise DNA is a very, very large one.

RNA is a much more interesting candidate than DNA, though, due to the fact that its monomers require less handling to assemble and moreover due to the fact that it can have catalytic activity - unlike DNA, RNA can catalyse its own reactions, making it a much better candidate as a building block for life. It's horrendously unstable outside of very tightly-controlled environments, though, so proteins are probably still a better shot.

 

[Kylie-chan]

This entire discussion works on the assumption that alien life would use DNA, which is in my opinion very unlikely to be valid.

DNA alone is not enough. In fact, the sugars from which DNA is assembled (on Earth, at least) require processing by protein machinery before their polymerisation into a long DNA chain. Beyond this, its maintenence, regulation, and activity in transcribing proteins is all handled in itself by other proteins. The real question here is if "aliens" are protein-based, and if so, if these proteins are able to assemble and handle DNA at all. This seems enormously unlikely to me; DNA is an extremely stable polymer but there are probably hundreds of thousands of conceivable molecules which could form code-like polymers given the protein machinery to support them.

So the real question here is if alien proteins, presuming they use them (again - an assumption which may not be true, although which in my opinion is more plausable than the DNA assumption), are of similarity to those found on Earth. The answer to this, in my opinion, is no - probably not. If you understand that DNA, which uses only 4 bases, is complicated then when you consider that proteins contain 20 different amino acids, the order of which dictates their structure and function, you might begin to realise just how many ways these things can work. The assumption that they'd just so happen to utilise DNA is a very, very large one.

RNA is a much more interesting candidate than DNA, though, due to the fact that its monomers require less handling to assemble and moreover due to the fact that it can have catalytic activity - unlike DNA, RNA can catalyse its own reactions, making it a much better candidate as a building block for life. It's horrendously unstable outside of very tightly-controlled environments, though, so proteins are probably still a better shot.

Very good post, I glossed over the fallacy that life at a molecular level is life at a DNA level, since I was more interested in dismantling some of the others and was entertaining that assumption. Chemically different genetic material likely exists in the universe anyway, maybe even with protein analogues.

I do not think it entirely implausible RNA may have proceeded divergently on different planets, and if other life with comparable molecular basis to ours exists I'd expect something similar. I believe the existence of nucleobases in space is one of the many likely and unlikely things NASA has considered, and pre-RNA world hypothesis still exists even. (Although it is origin of life we're discussing here) DNA/RNA analogues are used on Earth tho.

 

[Shamol]

However, your philosophical reasons are wrong: our concept of language is based on what we ourselves know and has nothing to do with the chemical properties of amino acids. We saw an analogue between language/code and between the functionality of amino acids and symbols (which are pretty broad) and assigned them that type of conceptualisation. Our concept of language has nothing to do with whether alien DNA might be analogous or not chemically, let alone abstractly.

To illustrate, and this is something I'll briefly add my two cents on: atomic elements (or even subatomic... again, the way we describe compounds is partly rooted in convenience and conventional linguistics) themselves can be viewed as morphemes. Meaning anything chemical can be a language, really. What is the genetic code except carbon chains? It's that very arbitrariness you hit on that debunks a philosophical argument for similarity but supports the chemical and biological arguments for similar properties. Evolution is arbitrary but not random, because some things are likelier to work than others.

I understand the genome can be referred to as a language only in a highly metaphorical sense, and I'm not entirely sure how far that analogy would hold. As you pointed out, other aspects of biology (and even chemistry) can fit a linguistic description perhaps just as easily. What I was trying to point out was the fact that the DNA-amino acid code correspondence is ultimately arbitrary- and the language metaphor fleshes out this one aspect rather adequately I think.

As for evolution being arbitrary but not random, there are some discussions to be had about the different layers of randomness. Consider the fact that some RNA viruses use what strikes us as an altogether unnecessary replication system- using a reverse transcriptase enzyme to convert the RNA into DNA and then back into RNA. This strikes us as unnecessary and random. And yet, it happened. Which means in whatever microenvironment this instance of evolution took place, it was non-random in the way you mentioned- evolution hit upon an innovation that made sense in that restricted context. Despite this general principle holding true on both earth and (say) Mars, there's no reason to think that the underlying environmental and other conditions would share enough variables to drive evolution in a similar direction (prima facie at least). In brief, while evolution is arbitrary and in a restricted sense non-random, that may or may not mean much in the proverbial larger scheme.

So the question that we ultimately have at our hands is- is there any overarching "logic" to the genetic code? Is the assignment of amino acids to individual codon triplets entirely arbitrary, or is there a method to this? I haven't studied this in any detail, but there is evidence that the genetic code is "error minimizing" in significant ways, and there are definite biases in how the code was constructed (link, link).

From the second paper:

Since discovering the pattern by which amino acids are assigned to codons within the standard genetic code, investigators have explored the idea that natural selection placed biochemically similar amino acids near to one another in coding space so as to minimize the impact of mutations and/or mistranslations. The analytical evidence to support this theory has grown in sophistication and strength over the years...

So in a very, very restricted scenario where you have aliens who share the same basic information storage and interpretation machineries as us (DNA/amino acids/proteins)- there's a chance there would be some convergent evolution due to the adaptive advantages our current genetic code has.

As Razor Leaf pointed out though (and as I mentioned in the disclaimer of the first post), these are really big if's. It was more of a fun thought experiment if nothing else.