Hacker News Reader

Comment by peter303 | Hacker News Reader

Perhaps there is slightly slightly different ground energy state of the two different chiralities and Nature selected the slightly lower one. I have not seen evidence supporting this hypothesis.

A similar mystery is why DNA chose a base64 nucleic coding system for 21 amino acids and a few punctuations. Some of the codons are redundant and some unused. (There appears to be a fossil base16 coding system inside it.) And synthetic biologists have modified it into a working base216 system with additional nucleic acids and amino acids. Perhaps again there is some ground energy argument as to why this universal 3-billion year old coding system is as complex as it is and no more complex.

View on HN

> Some of the codons are redundant and some unused.

I recall reading about a paper some 10 or 20 years ago where they simulated different codon mappings and tested them for resistance to genetic mutations. IIRC they found that what we got is fairly close to optimal, in terms of which codons are redundant and to what degree etc.

One hypothesis around the 20 amino acids relates to optimization of a quantum search algorithm [0]. Some recent work has added support to this idea [1].

[0] https://arxiv.org/abs/quant-ph/0002037 [1] https://www.technologyreview.com/2019/09/12/133081/an-import...

That does not seem a good paper. The only references on biology are quite generic books. Also the stated hypotheses are quite unfounded. As far as I know, there is barely any evidence of biology using quantum phenomena directly, except perhaps regarding photosynthesis and enzymatic activity:

https://en.m.wikipedia.org/wiki/Quantum_biology

If selection of basepairs came first (which is implied by RNA world hypothesis), then base 64 is the minimal encoding length for 21 amino acids + signalling. The question of why isn't it more complex becomes slightly moot - it's as simple as it can be given constraints. The question is why it isn't simpler then becomes asking why we have 21 amino acids.

Also not all DNA mutations produced by DNA damage are equally common, and some of the redundancy lessens their deleterious effect.

> A similar mystery is why DNA chose a base64 nucleic coding system for 21 amino acids and a few punctuations.

This implies a kind of Cambrian explosion of coding systems for nature to "choose" from. Maybe xenogenetics will help us recognize its remnants.

As my father pointed out to a younger me, 'Evolution is lazy. It would always rather repurpose something already created than build something new from scratch.'

So, why base64? Because there was already a nice base64 library in the evolutionary repo...

One unmentioned reason is due to DNA needing a proper folding shape and polarity for the accompanying enzymes to work correctly. Imagine a sequence where one codon causes a kink or part of the DNA strand to get stuck to another section. The working version would be another variant (with a different polarity, say, due to different nucleobases) resolving the earlier mechanical issue.

Huh. DNA is like an x86.

base216? Not base256?

Probably 6^3 compared to the original 4^3, by introducing two new nucleic acids beyond A,C,G,T, and keeping the codon size fixed at 3.

Reply on news.ycombinator.com