4 ms·
So, then, protein folding is likely also possible to make Turing complete, no?
by kian 3y ago
So, then, protein folding is likely also possible to make Turing complete, no?
- bananabiscuit 3y agoProtein folding is Turing complete because it can simulate a Turing machine by folding itself into a homo sapien, taking computation theory in college, and stepping through a Turing machine with a pen and paper for homework.
- grow2grow 3y agoThank you for this excellent example, which I'll be using when defending my unpopular opinion that Occam's razor is nonsense.
- kian 3y agoI love this response. On a slightly smaller scale, there are also ideas of protein interaction-network computational 'circuitry' -- I think that showing that folding can also compute is a nice addendum to this.
- andrewflnr 3y agoIf they are, it will be for completely unrelated reasons. Protein folding has very little in common with paper folding beyond the name. In particular, proteins are basically 1-dimensional, whereas paper folding is inseparable from the 2d nature of paper.
- dekhn 3y agoprotein are 1-dimensional only if you throw away large numbers of degrees of freedom, and every detail of their folding is determined by their three-dimensionality.
- andrewflnr 3y agoThat's equally (almost vacuously) true of origami, but the inherent topology of the building blocks still has a huge effect on what those degrees of freedom actually are. Do you think they're similar between a peptide chain and a sheet of paper? If no, then what are you actually arguing?
- dekhn 3y agoproteins themselves are turing complete; the folding is not really the interesting bit. just like there is a turing tarpit, trying to do anything reliable with protein folding (the process) is not likely to be effective. Instead, merely use existing functions provided by evolution to manage a DNA tape. You don't even need proteins; you can do it with just RNA and DNA. If you're really clever, with just RNA, and if you're super clever, just DNA.