Say, in the context of finding microbial life on Mars (i.e., organisms that evolved from the start with six nucleotides, not just taking current terrestrial organisms and swapping out the nucleic acids and ribosomes).
Say, in the context of finding microbial life on Mars (i.e., organisms that evolved from the start with six nucleotides, not just taking current terrestrial organisms and swapping out the nucleic acids and ribosomes).
Sounds like a greater risk of transcription errors. It’s possible for RNA to be transcribed into erroneous DNA. Normal pairs are AT and GC. But if something goes wrong, you can see adenosine paired with cytosine, for example. Best case, it produces junk. Worst case, it’s cancer.
Now instead of one to one match, it’s a possible two to one. If a nucleotide gets damaged, but is still a valid combination, you can’t tell it was damaged. That’s a problem.
Additionally, you no longer have matching strands of DNA. Right now, if you have one strand, it encodes the same information as the other. Not only can you read either strand, but you can take two strands and rebuild them into two whole new strands. But if each nucleotide corresponds to two others, when you separate the strands, you don’t have enough information to rebuild it. Nor can you necessarily build the protein you wanted, because you don’t have that full information any more. A sequence doesn’t have a corresponding sequence on the other strand.
No, each nucleotide would still pair with one other—there would just be an additional possible purine/pyrimidine pair (i.e., A:T + G:C + X:Y).
As for transcription errors, you’d only have 36 potential codons (62) instead of 64 (43), so it seems like the process could be more robust.
Oh, I misunderstood. Yeah, in that case there isn’t such an increase in risk of transcription errors. I think it’s still increased slightly, because each nucleotide could become one of five others instead of one of three others, so it might still be easier for a stray high-energy particle to cause a conversion. Depends on how they’re formed.
But, I’m not any kind of biologist.
Translation errors seem like biggest problem, with the current system you have some amount of redundancy where sometimes a single point mutation still gives the same codon but you would lose that with less base pairs per codon