6 ms·
DNA Digital Data Storage
- ausbah 7y agoIs there anything significant that could be learned from DNA and applied to traditional computer storage methods? Or is this efficient something intrinsic to DNA that can't simply "be transferred"?
- AstralStorm 7y agoNot really. The offset reads given specific primers producing entirely different results, perhaps, for steganography.
- ken 7y agoDNA storage for steganography was, in fact, the basis for a Star Trek episode!
- wespiser_2018 7y agoThe only significant thing about DNA, as applied to information storage, is just that DNA is nearly as energy efficient as physically possibly[1] for storing bits of information, and therefore, it has incredible potential to encode everything. The issue, at least from the last time I heard George Church speak (~Fall 2016), is that the interface for DNA storage systems would have to be radically re-imagined, since we can't go between Digital <-> DNA anywhere near the order of magnitude as we can go DNA <-> DNA, or Environment -> DNA. Something like genetically engineered bacteria that live on the walls and have special proteins that sense and encode the environment, and count the number of unique human genomes they encounter, like a time series database (think PipelineDB) may be possible. [1] I'm pretty sure it's based off this paper, but I'm not 100% sure. https://www.sciencedaily.com/releases/2009/07/090715131447.htm https://www.sciencedaily.com/releases/2009/07/090715131447.h... Basically, the idea is that for the energy used to copy DNA, DNA polymerase is very good at only inserting the "correct" nucleotide.
- jerf 7y agoIf you read the metaphorical marketing data sheet, DNA can look really interesting. If you get down into what reading it actually entails, it honestly starts looking less enticing. Biology optimizes pretty heavily for constructing proteins with it and the ability to duplicate it, not generic data storage. There's a lot of details you don't learn in school about how transcription works that make it less and less appealing as a general-purpose data store the more you learn, IMHO. If nothing else, it looks more like tape storage than any sort of random access storage, which right off the bat is a pretty significant hit nowadays. DNA mastery is important on its own terms as biology mastery, but I wouldn't be surprised there is no point in the future in which it is the optimal mechanism for storing generic data.
- waiquoo 7y agoThe other aspect is that the shelf life of DNA can be significantly longer than other mediums. I haven't personally done anything with DNA data storage, but the idea is that tape storage is good for 30-50 yrs, but we can recover intact DNA from mammoths (~10,000 yrs)
- DoctorOetker 7y agoI believe that DNA recovery after 10k years is not very representative of DNA durability.
- waiquoo 7y agoYou are correct, but the variable of interest is recoverability
- cdirkx 7y agoIsn't this mostly due to the huge redundancy? All trillion cells of that mammoth carried duplicates of the encoded information, and recovering that information involves a lot of matching incomplete degraded strings against each other. The estimated half-life for DNA is apparently 521 years [1] (still quite a bit longer than 50 years). [1] https://www.livescience.com/38150-dna-degradation-rate.html https://www.livescience.com/38150-dna-degradation-rate.html
- dsalzman 7y ago"In June 2019, scientists reported that all 16 GB of Wikipedia have been encoded into synthetic DNA.[4]"