5 ms·

Data storage on DNA can keep it safe for centuries

steve19 11y ago
Does this work because there is massive redundancy in the DNA soup? So if some strands of part of strands are contaminated, there are 100s or 1000s (millions?) of cloned DNA molecules floating around? Or is a single DNA molecule capable of long term data data storage and retrieval? I realize of course that scientists pull DNA from long dead things all the time, but for example there is of course hundreds of trillions of cells in a large mammal with copies of the DNA.
- wanda 11y ago
  https://dl.dropboxusercontent.com/u/32029178/Science-2012-Church-1628.pdf https://dl.dropboxusercontent.com/u/32029178/Science-2012-Ch...
ghshephard 11y ago
It's painful to read the NYT write about something you know about: It would also address a glaring Achilles’ heel at the heart of microelectronic data storage systems: Magnetic disks, tape and even optical storage systems safely store information at most for only a handful of decades. It's really not a glaring Achilles heel. This is a solved (for the most part) problem - see http://www.mdisc.com/ http://www.mdisc.com/ for a product which can archive your data a reasonable cost for centuries, and much more conveniently than trying to store/recall from DNA. Here are some of the endurance tests they put the mdic through - http://www.esystor.com/images/China_Lake_Full_Report.pdf http://www.esystor.com/images/China_Lake_Full_Report.pdf. I would be interested in seeing how DNA storage stood up to that sort of testing. The ability to store data in DNA is cool, and it's neat that the teams are making progress - but it's important to put the progress into context right now.
- theandrewbailey 11y ago
  Data integrity is not inherent to optical media, but it apparently is with DNA. While special archival grade writable DVDs and BluRays exist, people are cheap, and for the most part will not pay $2+ per disc. > a product which can archive your data a reasonable cost for centuries, and much more conveniently than trying to store/recall from DNA. If you're going to store media that long, you will also need the equipment to read data off it in n years/decades/centuries. DNA has been around for a very long time compared to optical media. The need to read DNA will last much longer than optical media. In that time, optical drives are likely to become obsolete and scarce (see the last 5 years), and DNA readers might be extremely cheap by comparison.
wuschel 11y ago
I wonder what the failsafe mechanisms are in regard to infection of the storage medium with fungi/bacteria that might alter the stored information. In my humble opinion that might a problem that should be considered when it comes the actual application of the concept.
- ende 11y ago
  Well, the danger of contamination would be in any host vectors used to clone the DNA. Its conceivable that a virus could infect the host vector and introduce its own chromosome into the host chromosome, but this probably wouldn't be a major problem. Any interuption in the encoded data should be easily detectable, highly unlikely in the first place, and there would be many copies of the dna. But the actual storage would be dna outside of a cellular context, in which case ot wouldn't be 'active'. The only dangers in that state would be water, heat, etc.
Myrmornis 11y ago
Why DNA out of interest, could one imagine any other polymer being chosen for the purpose? Is the reason: - Stability? - The dynamics and specificity of ligation between complementary sequences? - Alphabet size -- is 4 a good choice? - Enzymes exist that can carry out a lot of the necessary operations? - Humans have developed a lot of expertise in manipulating DNA?
- ende 11y ago
  All of the above! Dna is very stable (compared to, say, many proteins).. it is essentially the cell's harddrive, while mRNA is more like the memory. DNA is able to express complex information through sequences of a small number of states, is replicated and maintained through a highly conserved set of molecular machinery throughout life, and the human experience in manipulating it is approaching that of pushing electrons across silicon boards. The similarities are quite startling!

Does this work because there is massive redundancy in the DNA soup? So if some strands of part of strands are contaminated, there are 100s or 1000s (millions?) of cloned DNA molecules floating around? Or is a single DNA molecule capable of long term data data storage and retrieval? I realize of course that scientists pull DNA from long dead things all the time, but for example there is of course hundreds of trillions of cells in a large mammal with copies of the DNA.

wanda 11y ago
https://dl.dropboxusercontent.com/u/32029178/Science-2012-Church-1628.pdf https://dl.dropboxusercontent.com/u/32029178/Science-2012-Ch...

ghshephard 11y ago

It's painful to read the NYT write about something you know about: It would also address a glaring Achilles’ heel at the heart of microelectronic data storage systems: Magnetic disks, tape and even optical storage systems safely store information at most for only a handful of decades. It's really not a glaring Achilles heel. This is a solved (for the most part) problem - see http://www.mdisc.com/ http://www.mdisc.com/ for a product which can archive your data a reasonable cost for centuries, and much more conveniently than trying to store/recall from DNA. Here are some of the endurance tests they put the mdic through - http://www.esystor.com/images/China_Lake_Full_Report.pdf http://www.esystor.com/images/China_Lake_Full_Report.pdf. I would be interested in seeing how DNA storage stood up to that sort of testing. The ability to store data in DNA is cool, and it's neat that the teams are making progress - but it's important to put the progress into context right now.

theandrewbailey 11y ago
Data integrity is not inherent to optical media, but it apparently is with DNA. While special archival grade writable DVDs and BluRays exist, people are cheap, and for the most part will not pay $2+ per disc. > a product which can archive your data a reasonable cost for centuries, and much more conveniently than trying to store/recall from DNA. If you're going to store media that long, you will also need the equipment to read data off it in n years/decades/centuries. DNA has been around for a very long time compared to optical media. The need to read DNA will last much longer than optical media. In that time, optical drives are likely to become obsolete and scarce (see the last 5 years), and DNA readers might be extremely cheap by comparison.

wuschel 11y ago

I wonder what the failsafe mechanisms are in regard to infection of the storage medium with fungi/bacteria that might alter the stored information. In my humble opinion that might a problem that should be considered when it comes the actual application of the concept.

ende 11y ago
Well, the danger of contamination would be in any host vectors used to clone the DNA. Its conceivable that a virus could infect the host vector and introduce its own chromosome into the host chromosome, but this probably wouldn't be a major problem. Any interuption in the encoded data should be easily detectable, highly unlikely in the first place, and there would be many copies of the dna. But the actual storage would be dna outside of a cellular context, in which case ot wouldn't be 'active'. The only dangers in that state would be water, heat, etc.

Myrmornis 11y ago

Why DNA out of interest, could one imagine any other polymer being chosen for the purpose? Is the reason: - Stability? - The dynamics and specificity of ligation between complementary sequences? - Alphabet size -- is 4 a good choice? - Enzymes exist that can carry out a lot of the necessary operations? - Humans have developed a lot of expertise in manipulating DNA?

ende 11y ago
All of the above! Dna is very stable (compared to, say, many proteins).. it is essentially the cell's harddrive, while mRNA is more like the memory. DNA is able to express complex information through sequences of a small number of states, is replicated and maintained through a highly conserved set of molecular machinery throughout life, and the human experience in manipulating it is approaching that of pushing electrons across silicon boards. The similarities are quite startling!