This question is tied to the “Glass Trees” question I asked earlier which everyone seemed to like. Since y’all had such good opinions on that, I figured I’d bring this particular quandary to y’all in hopes that you could bring peace to my nearly month-long torment over what to do.

3 Days in my Worldbuilding bakery, and this is my little bundle of joy. <3 The setting: A planet 90% the mass of Earth, with an iron core to provide a magnetosphere. Planet has similar composition to Earth, though has a moon which causes tides roughly 2.5x greater. (Orbital mechanics aren’t important here, this is just to describe the setting.) Life forms are carbon-based, using sugar-phosphate based genetic biopolymers. Atmosphere is mostly comparable with that of early Earth, during the Ediacaran/Cambrian.

The Setup: So I wanted to have a genetic compound used for the biosphere that was different from Earth, but not so different that it would be incomprehensible. My main needs for this mirror to DNA were:

  • It should be comprised of the same or similar core elements as DNA; no swapping Carbon with Silicon here, we stayin’ organic on this planet.
  • It should be unable to interact with DNA or RNA. I want to try avoiding any possibility of “this alien virus could infect you and mutate you into a monster”, or any cross-species breeding going on. Terran life and these aliens should not be able to interact on a genetic level, ideally.
  • It should function in roughly equivalent ways - self-replicating, coding for proteins, allows for Darwinian evolution, all that bacon.

The Problem: …I am not good with chemistry. I’m skilled with biology & geology, I’m a passionate amateur for astronomy, and I know enough physics (quantum & otherwise) to follow a conversation. But I have always struggled understanding chemistry in all its forms; without a smart person to balance me out, I’m stumbling in the dark. That being said, I did try to do some research, and even though I only understand 50%-70% of it, I at least have enough to make inferences.

What I was using for a while, but turned out not to work. My Research: My first candidate was HomoDNA, an alternative form of DNA that replaces Deoxyribose with Dideoxyallopyranosyl, which is a homolog hexose. I operated for a while under the assumption that this HomoDNA would work similarly enough to DNA & RNA to be comparable. However, thanks to a paper that a friend of mine sent me, I now know that HomoDNA doesn’t form stable Watson-Crick pairings like DNA & RNA does. Plus, it doesn’t seem to form stable helical frames either. So Homo-DNA is out.

D-pyranosylribose (p-RNA), the pentapyranosyl isomer of RNA (whatever that means), has shown to form pairing comparable & superior to DNA & RNA. It also does not bond with standard DNA & RNA, which makes it more viable for my "aliens cannot infect Earthlings" criteria. I don’t know much about it besides the small reference in that fore-mentioned paper.

Among synthetic or Xeno-nucleic acids (XNA's), Fluoro-arabinose (FANA) and Anhydrohexitol (HNA) are also favored alternatives to DNA. They form stable base pairing, can also pair with DNA, and also work with synthetic catalytic enzymes, which is prevalent to the idea of the chemical origins of life. However, I don’t know how difficult these synthetic strains would be to form under natural circumstances; I recall that something mentioned regarding Hajimoji-DNA is that it cannot exist outside of a lab environment, so they must introduce something that allows it to form and prosper that doesn’t exist in nature.

And for the record, I did look into Hajimoji-DNA. I am likewise intrigued, though I feel as if having a structure that codes for 8 different nucleotides would lead to a higher risk of mutation & cancers, as well as having more potential to code various proteins. It feels like an impractical thing for nature, in that regard.

My Question: (finally) My question that I present to you all is, which of these DNA alternatives seems to be the most likely one to arise naturally on an alien planet? Which of them would fit the criteria I listed above best? Is there another substitute that you feel fits better? Any advice and information you could provide would be helpful! (Just keep in mind, I’m no chemist, so y’all might need to dumb a few things down for me.) :P

ADDEN.: RESPONSE TO GNA SUGGESTION From the same paper I’ve been referencing up above. I did some research into Glyco-Nucleic Acids (GNA), as folks in the comments suggested; I did confirm, the structure is the same, just the paper calls it “glycerol” instead of “glycol”. This also seems promising, but there is some confusion over how stable it is. The research paper “De Novo Nucleic Acids” cites GNA as having a lower melting temp., thus forming a more flexible backbone structure and destabilizing Watson-Crick pairing. However, the Wikipedia article linked says the exact opposite: that GNA requires a higher melting temperature, and thus forms more stable Watson-Crick pairing. I’m unsure if this is an error on someone’s end (as in whoever’s writing the paper/articles), or perhaps Glyco-Nucleic Acids and Glycero-Nucleic Acids ARE different? Hope y’all can help.

  • 1
    $\begingroup$ Possibly look at the alternative models for DNA that Paulings' and Franklin's teams were looking at? (Various single/triple spirals) I'm no biologist but I understand that there's no way that DNA or RNA can interact with the cell machinery that unzips/rezips DNA or decodes RNA. $\endgroup$
    – user86462
    Jan 16, 2023 at 6:59
  • 2
    $\begingroup$ What's wrong with glycol nucleic acid using P, Z, S, and B bases, and possibly with your alien lifeform using a different set of amino acids to make their proteins? $\endgroup$
    – user73910
    Jan 16, 2023 at 7:24
  • 1
    $\begingroup$ @AtlastheWorldbuilder Glycol nucleic acid does not use glucose, but propylene glycol. $\endgroup$
    – user73910
    Jan 16, 2023 at 12:15
  • 1
    $\begingroup$ This is what I am talking about, by the way: en.wikipedia.org/wiki/Glycol_nucleic_acid $\endgroup$
    – user73910
    Jan 16, 2023 at 12:22
  • 1
    $\begingroup$ Ooh, Glycol Nucleic acid might beat my answer - higher temperature Watson crick base pairing would mean at least one less obstacle to a high temperature organisim $\endgroup$
    – lupe
    Jan 16, 2023 at 13:03

3 Answers 3


TL;DR: I suggest either GNA or PNA

First, a couple comments on the alternatives:

Regarding FANA:

Very few fluorine compounds occur in nature and, as far as I know, almost all are produced by plants or microorganisms as defense mechanisms [1]. There are some fluoroalkanes that are produced by volcanoes, such as fluorobenzene [2]. Unfortunately, the amount of organofluorine compounds produced this way is low. In any case, if you do have an environment that produces some organofluorine compounds, you would need to figure out how they would be converted into fluoro-arabinose, and how the FANA can produce more of itself, preferrably using fluoride ions in the environment as the source of fluorine. Given that only one enzyme is known to catalyze the formation of a C-F bond, that might be a tall ask, especially if your planet is supposed to have a history similar to that of Earth. Furthermore, there's an argument to be made that proto-life that relies on fluorine on an earth-like planet is going to get outcompeted by proto-life that does not. Unfortunately, there haven't been any studies about the possibility of abiogenesis of a fluorine-containing nucleic acid, so all of this is speculation. Furthermore, sources conflict on whether fluorine or carbon is more abundant in Earth's crust. One thing to note, however, is that fluoride salts are not very soluble in water, which would mean that fluorine would not be as accessible to your proto-life as you would think judging by how abundant fluorine is in the planet's crust. So, ultimately, I'm going to say that, while it's plausible that there would be a planet where FANA is the information storage molecule, it won't be Earth-like.

Disclaimer: I could be completely wrong in my reasoning for this

Regarding p-RNA:

The other answer suggests p-RNA as the information storage molecule. Now, RNA as we know it is more prone to hydrolysis compared to DNA, which may be the reason why DNA evolved to be the information storage molecule [3]. Unfortunately, I haven't been able to find any studies regarding how prone p-RNA is to hydrolysis or whether it can evolve some mechanism to effectively protect itself from damage.

Now, onto my suggestions:

Glycol Nucleic Acid:

Firstly, glycol and glycerol nucleic acids are different molecules. Glycol nucleic acid uses propylene glycol as the backbone, glycerol nucleic acid uses glycerol. Glycol nucleic acid is more stable than DNA or RNA [4], and it follows base pairing rules. On the left is glycol nucleic acid, note the lack of an oxygen atom linking the backbone and the base. enter image description here

Given the simplicity of its backbone, it's entirely plausible for it to have originated under conditions similar to those of early Earth.

Other suggestion: Peptide Nucleic Acid

enter image description here

This nucleic acid uses N-(2-aminoethyl)-glycine as its backbone. It is hypothesized that PNA may have existed at some point in Earth's past [5]. There have also been some studies that suggest that PNA is, under certain conditions, capable of folding into structures besides the double helix [6], so it may be the case that PNA has some capability to act as a catalyst as RNA would. HOWEVER, just how capable PNA is in that specific regard has, to my knowledge, not been studied, so it might be that PNA is not as good as RNA in regards to being able to fold into a shape with catalytic activity.

Three problems with this idea:

First, source 4 says that the double helix formed by PNA is wider and more slowly-winding compared to DNA, I'll explain why that might be problematic later.

Secondly, N-(2-aminoethyl)-glycine, the backbone of PNA, is a neurotoxin. So, while alien viruses are probably not going to be infecting humans with any degree of success, it might be unwise for a human to eat alien life that's based on PNA.

Thirdly, PNA does bind to DNA and RNA. Nucleotides based on PNA have been used in antisense therapies. That is to say, if a fragment of PNA were to bind to a bit of messenger RNA or even DNA in the nucleus, that fragment will not be able to be read by the body's translation machinery. However, that would require that PNA somehow gets inside a human cell and that that bit of PNA is complementary to some bit of DNA or RNA in the body. Given that totally alien life is probably going to evolve to use a different set of amino acids for its proteins and that alien viruses are going to evolve to target alien life and replication machinery used by that life, that's not going to happen. Furthermore, human translation machinery will not read PNA, so you can rest assured that, even if an alien virus were to somehow get inside a human cell, it won't be able to replicate.

Alternative Bases:

Life on Earth uses adenine, guanine, cytosine, and thymine as the nucleobases in DNA (RNA uses uracil instead of thymine). However, scientists have also created 8 synthetic nucleobases that could serve as the basis of encoding genetic information [7]. In addition to using a different backbone, you can also use a different set of nucleobases.

Possible Issues:

In nature, humans and other eukaryotes use a complexes known as histones to make DNA more compact and as a means of regulating gene expression. I do not know if the proposed nucleic acids would be compactable or not. Also in regards to GNA, I have not been able to find studies as to how resistant it is to hydrolysis.

In regards to your three criteria:

  • The proposed nucleic acids still use the same elements as DNA. Check.

  • While PNA could bind to DNA, Earthling cells' transcription and translation machinery won't read PNA or the DNA/PNA structure, so you can rest assured that alien viruses won't be able to infect Earthlings. Sidenote: Whether or not the alternative nucleic acid can bind to DNA is not enough to determine whether it's possible for a virus to infect a DNA-based organism. The important criteria here are whether the virus can gain entry into a cell and whether the cell's translation machinery will be able to read the genetic code and make more viruses.

  • "Maybe" on whether PNA could catalyze chemical reactions to allow for the emergence of early self-replicating molecules or if there would need to be some RNA-analog as well. But, if conditions are right and life can evolve to be more complex, it certainly could code for proteins. As for Darwinian evolution, there only need to be four things for that to occur:

  • Things must be capable of reproduction. Check.

  • The offspring must have some variation from the parents. Provided that the replication of the genetic material introduces some errors, check.

  • These variations must be heritable. Check.

  • Certain variations must make the thing more capable of surviving and reproducing. Check.


[1] https://www.tcichemicals.com/US/en/support-download/chemistry-clip/2013-10-08

[2] Gribble, G. W. (n.d.). Naturally Occurring Organofluorines. Organofluorines, 121–136. doi:10.1007/10721878_5



[5]Banack, S. A., Metcalf, J. S., Jiang, L., Craighead, D., Ilag, L. L., & Cox, P. A. (2012). Cyanobacteria Produce N-(2-Aminoethyl)Glycine, a Backbone for Peptide Nucleic Acids Which May Have Been the First Genetic Molecules for Life on Earth. PLoS ONE, 7(11), e49043. doi:10.1371/journal.pone.0049043



  • $\begingroup$ This is fantastic! You have given me so much information to work with from this, you are amazing! :D Doing my own digging into PNA, it does seem to fit alot of the criteria for what I'd be looking for in life like this! I might need to do a bit more digging, but you've given me some solid leads and some good sources to work from! Thank you thank you THANK YOU!!! $\endgroup$ Jan 18, 2023 at 20:03
  • $\begingroup$ (Also, I think this little bit of biochem-digging might've uncovered a glaring error in the Wikipedia entry on GNA, in that case, since Wikipedia says the two are interchangeabe when they definitely aren't. However, I'm not sure what to do with that or how to correct that edit, so I'll leave that up to you in case you're the one who wins any special prizes for it. :P) $\endgroup$ Jan 18, 2023 at 20:04


So, caveating that I'm more of an evolutionary biologist than biochemist, and these days more of a programmer than either.

One of the issues here is not making the things - I'd strongly believe that all the above, given the right enzymes, are synthesizable by biological life.

However, any form of DNA existing in a similar configuration to we have now would be a strong hint that this life is related to our planet's life in some way. There's sort of nothing inevitable about DNA arising as the stable information storage molecule.

RNA on the other hand..

RNA is both an information store, and a functional molecule. Arguably the most important enzyme is an enzyme made of RNA, and it is basically a hangover from when everything in the cell was an enzyme made of RNA.

RNA also forms simply, from, we think, the primordial soup - the conditions on earth before life arose. The idea goes that RNA strands formed, some of them became self replicating, and they grabbed all of the resources and began battling it out. A few billions of years of evolution later, throw in a cell membrane, proteins, DNA, complex carbohydrates, and you have us!

So, the argument here is that those same conditions arose somewhere else. Rather than go with DNA, RNA simply got better. Now we have aliens with no DNA, and possibly some extremely different chemistry. So, prehaps, missing DNA, maybe much more functional RNA molecules in the cell. We might see lower protein involvement, with proteins used for membrane spanning and some more complex machinery

  • $\begingroup$ I really like this idea! I’ll have to consider how far I want to take it, though. Most of the unicellular creatures I’ve already come up with have structures similar to Nucleii and ribosomes, and I’m not sure if I’m equipped to rewrite a whole new cellular biology system right away to fit this. Would it even work with Nucleii? :? $\endgroup$ Jan 16, 2023 at 19:29
  • 1
    $\begingroup$ So, I'm not sure - which is sort of a fun implication - I'd take a small guess that you might end up with a less centralised structure. However, you still need coordination of some sort, and some checks on the bits of RNA reproducing wildly, so you might see some centralisation and a kind of nucleus like setup $\endgroup$
    – lupe
    Jan 16, 2023 at 21:50

DNA and RNA with additional letters dont need proteins at all. Protein folding acts as a catalyst and structural material. with additional letters rna can do this job. https://www.quantamagazine.org/new-letters-added-to-the-genetic-alphabet-20150710/

P.S. Peptide nucleic acids are super interesting. Id love to learn more about them for anyone interested.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .