Hacker News Reader

Comment by enkid | Hacker News Reader

DNA doesn't make a copy of itself, the cell makes a copy of the DNA. DNA doesn't really do anything, it requires RNA and proteins to actually do anything.

View on HN

well, quine program doesn't do anything by itself too, you need computer and electricity for it to work

There's a definitional problem here. Would a book with instructions for how to order additional copies be a quine? What's the difference? I don't know exactly.

I guess the right question to ask about DNA is, does it contain all the instructions required to produce a copy of the DNA itself? DNA translation involves a bunch of RNA, sure. But DNA copying is mostly done by DNA polymerases, which are typical peptide based proteins encoded in the DNA itself, right?

So in a real sense, the DNA contains the blueprint for the machinery to copy the DNA itself, but not the instructions for building that machine from the blueprint, the same way a software quine doesn't necessarily include the source code for a compiler.

Definitely a definitional problem, but that's my shot at resolving it a bit.

Then again, DNA polymerase is not perfect, and DNA is not immune to single nucleotide changes having a real effect, so it's only a quine in the "spherical cow" way of looking at it anyway. But it's certainly the most "quinish" thing in nature that I can think of.

I just don't think it's accurate to say DNA has instructions in the same way a computer does. It's more like a recipe for a chemical reaction. If we expand the definition of a quine away from being self reproducing code and to self reproducing chemical reaction, it seems like things like crystalization or fire could be considered a quine.

I actually like that.

But I may have been a bit unclear about my mapping of computer and biology terminology to this question.

I tend to think of a computer programs as being mathematical machines, and proteins as being molecular/chemical machines. Motor proteins "walking" along microtubules is the most vivid example of this.

I'm sort of torn between both worlds here, honestly. I like both perspectives. Viruses especially have a very quinish quality to them in that they're stripped down to just enough stuff to get a host cell to make more viruses.

I like the virus as quine analogy much better than generic DNA as a quine, personally. Seems much more meaningful.

DNA includes instructions for how to build all the machinery needed to make a copy of itself, including the surrounding cell, and RNA, and proteins, and the organism that is capable of gathering food/fuel to obtain the all the raw materials to build everything.

No it doesn't, it holds a recipe for creating proteins under specific chemical and biological conditions. If you put human DNA in a chicken egg, it wouldn't hatch a human.

And if you write a quine computer program on a piece of paper, it won't create a copy of itself further down the page.

Or if you put a fully-functional cell in the vacuum of space, or the heart of a star, it won't replicate either.

All replicators need a suitable environment to replicate in. The fact that inhospitable/non-viable environments exist does not negate the fact that they are actually replicators.

If you build a cell from only eukaryote DNA instructions, it won't have any mitochondria and be unable to copy itself.

Mitochondrial DNA is still DNA.

Symbiotic replicators that rely on the presence of each other to reproduce (like cycles of quines?) are also still replicators.

I think also the Golgi apparatus has no DNA. You can't build one if you don't have one. But apparently we're still not sure where it comes from.

The program doesn't make a copy of itself, the runtime makes a copy of the program. Program doesn't really do anything, it requires runtime and cpu to actually do anything.

Reply on news.ycombinator.com