Recommended Posts

I ran across this yesterday.  Very interesting stuff.  Considering the advances made on this already, I see a viable technology made available for use just around the corner - as in soon.  Investment-wise, I'm waiting and watching to see who picks up the mantle.

With a 'hello,' researchers demonstrate first fully automated DNA data storage

Share this post


Link to post
Share on other sites
Posted (edited)

How is this a better solution than collecting data in a 1-atom thick graphene layer?

Edited by Cassel

Share this post


Link to post
Share on other sites
44 minutes ago, Cassel said:

How is this a better solution than collecting data in a 1-atom thick graphene layer?

I don’t know anything about that graphene technology, so this is a bit speculative but:

Fabrication. We already have a lot of experience with microfluidics, DNA synthesis,  DNA sequencing, plus a bunch of other molecular biology techniques that might be handy for storing and indexing large amounts of DNA encoded data. My understanding is that producing significant areas of good quality (as in defect free or defect-managed if you want to use them to locally tweak the conductivity of your graphene) graphene is still challenging.

Backup. Almost trivial with DNA. Take your  DNA stored data, run it through a polymerase chain reaction and presto - more copies than you’ll ever need.

Or, for long term storage, splice your data into a standard E. coli strain and stick a couple of vials worth into a standard lab freezer. Added bonus - your backup is self replicating although you’d want to re-sequence it every so often to check for mutations in your data. If you get those, it’s nothing that a bit of CRISPR can’t fix.

Data compression. Viruses do some really funky data-compression tricks with their genomes. We could learn from that.

All told though, I think my first point is probably most important. Graphene might be technically better but DNA technology works (to proof of concept level) now and we have the tools to quickly iterate with it. That’s a big hurdle for graphene to get over. It’s kind of similar to the way that we have better semiconductors than silicon available.... but we use silicon anyway because it’s what we know and what our infrastructure is set up for.

Share this post


Link to post
Share on other sites
Posted (edited)
1 hour ago, KSK said:

I don’t know anything about that graphene technology, so this is a bit speculative but:

Fabrication. We already have a lot of experience with microfluidics, DNA synthesis,  DNA sequencing, plus a bunch of other molecular biology techniques that might be handy for storing and indexing large amounts of DNA encoded data. My understanding is that producing significant areas of good quality (as in defect free or defect-managed if you want to use them to locally tweak the conductivity of your graphene) graphene is still challenging.

Backup. Almost trivial with DNA. Take your  DNA stored data, run it through a polymerase chain reaction and presto - more copies than you’ll ever need.

Or, for long term storage, splice your data into a standard E. coli strain and stick a couple of vials worth into a standard lab freezer. Added bonus - your backup is self replicating although you’d want to re-sequence it every so often to check for mutations in your data. If you get those, it’s nothing that a bit of CRISPR can’t fix.

Data compression. Viruses do some really funky data-compression tricks with their genomes. We could learn from that.

All told though, I think my first point is probably most important. Graphene might be technically better but DNA technology works (to proof of concept level) now and we have the tools to quickly iterate with it. That’s a big hurdle for graphene to get over. It’s kind of similar to the way that we have better semiconductors than silicon available.... but we use silicon anyway because it’s what we know and what our infrastructure is set up for.

If you write that it is almost trivial, but to this day nobody uses it, it means that it is not that trivial :-)

The problem with DNA is that you need a laboratory clear environment. If you drop the device on the floor at home and something breaks, will you be in contact with the DNA of the virus or bacteria?

DNA for replication does not need any substances/resources?
What about UV radiation and mutations?

A few years ago I read about the possibility of producing a 1 m² graphene layer. With 1 cm² should be enough to build a small disk. Anyone knows how many graphene atoms can be placed on such a surface?
You just put it in vacuum box and there you go, you have storage device, that can last long. Copying can be done in the same way that optical drives work, you insert a blank graphene layer, and the device burns holes/atoms on layer to mark zeros and ones.

 

Edited by Cassel

Share this post


Link to post
Share on other sites

My thoughts, though not an expert:

It looks like the main problems that keeps us from mass using of DNA storage are cost (it is expensive per MB) and speed (it is slow). But it would last longer than magnetic techniques that last years at most, especially if stored spooled on a reel, or than optical stuff, that degrades in 10-20 years and must tediously be copied every now and then. I would assume that copying of DNA is less of a hazzle than copying 1000s of disks, especially in case of coding errors or partly degraded data.

And, of course, like all storage techniques, the ability to read and decode must still be there when and if the stored data must be read again one day. That's a huge problem inherent to all archiving systems.

Another technique could be manipulating single molecules in a 3d crystalline structure, or holographic storage, but i read we aren't that far yet.

If DNA storage is not too expensive and fast enough to read and write it'll be used. And it needs less valuable resources for the storage media.

Share this post


Link to post
Share on other sites
3 hours ago, Cassel said:

If you write that it is almost trivial, but to this day nobody uses it, it means that it is not that trivial :-)

The problem with DNA is that you need a laboratory clear environment. If you drop the device on the floor at home and something breaks, will you be in contact with the DNA of the virus or bacteria?

DNA for replication does not need any substances/resources?
What about UV radiation and mutations?

A few years ago I read about the possibility of producing a 1 m² graphene layer. With 1 cm² should be enough to build a small disk. Anyone knows how many graphene atoms can be placed on such a surface?
You just put it in vacuum box and there you go, you have storage device, that can last long. Copying can be done in the same way that optical drives work, you insert a blank graphene layer, and the device burns holes/atoms on layer to mark zeros and ones.

PCR itself is absolutely almost trivial. It’s a standard technique and all the kit and reagents needed to do it are commercially available. PCR for backing up DNA data storage - probably not done because I doubt development is at the stage where they’re worrying much about backup solutions.

DNA needing resources for replication? Depends how you do it. With PCR you’d need to supply the nucleotides, if you’re replicating in a cell, you’d need a suitable nutrient source (broth or agar).

UV radiation - keep it in a box. Mutations - this is purely guess work but I’m thinking you could get around this by having multiple copies of your data (which you’d likely have anyway by the nature of the system) so that your data is represented by the consensus sequence of those multiple copies and by having each bit represented by a long enough piece of DNA that you need a lot of point mutations to cause bitrot. 

As for needing a clean room - my reading of the article was that it was all about producing an automated system that didn’t need a lab? You’re right though, as long as this remains a lab-based technique it’ll just be an interesting curiosity rather than a practical technique.

As for getting in contact with the DNA of the virus or bacteria? The data store DNA isn’t going to code for anything biological (at least I assume not - that would seem to  be a basic safety precaution) so that’s not going to do anything to me. The actual virus or bacteria - I’m probably at greater risk every time I use the bathroom. :)

Besides, I’m thinking this will be some kind of lab on a chip arrangement (typing this on my phone so linking is awkward - go check it out on Wikipedia if you’re curious) so it’s likely to be quite robust. We’re not talking about a box of test tubes on your desk here. :) 

Share this post


Link to post
Share on other sites
Posted (edited)

Continued...

Producing a square metre of graphene is one thing - producing a square metre of graphene monolayer is another. At that sort of area I think you’re probably going be using some kind of vapour deposition technique and using that to grow monolayers strikes me as being a bit hit or miss. There will likely be patches of multilayer graphene, maybe even specks of graphite or amorphous carbon, on your monolayer.

But that’s an an engineering problem and one that I can imagine being solved given the current interest in graphene. 

I’m less sure about your read/write mechanism. I would have thought that the smaller the area of graphene you’re using to represent a bit, the more complicated and error prone your read/write optical system is going to be and I’d wonder if you could make it robust enough for consumer (edit, or routine use outside of an optics lab)

Also, the smaller your bits, the more at risk they’ll be from reactions with stray oxygen molecules in your system. Making a sufficiently high quality vacuum to avoid that might not be an easy or routine task.

TL: DR - both systems have challenges to overcome. I don’t think the graphene based system would be so much more straightforward than the DNA system as to make it the obvious winner. On the other hand I’m almost certainly biased due to greater familiarity (conceptually if not practically) with the technologies that a DNA system would or could be based on.

Edited by KSK

Share this post


Link to post
Share on other sites
Posted (edited)
3 hours ago, Green Baron said:

Another technique could be manipulating single molecules in a 3d crystalline structure, or holographic storage, but i read we aren't that far yet.

Bell Labs was working on that a long time ago,  perhaps even pre-1970's?  A one-inch square cube, Quartz?, dropped into a device that used a laser to read/write data.  Don't know what happened to that, but the memory boards featured (in 2001 A Space Odyssey) for HAL's brain were supposedly based on such a technology.  This same technology became reality for a short while, I remember seeing this device in use for video recording; https://www.technologyreview.com/s/404603/holographic-memory/

 

Edit:
Turns out this idea is still under development.  Google 'microsoft project silica'.

Edited by LordFerret

Share this post


Link to post
Share on other sites

An interesting downstream effect of this becoming viable will be that since data needs to be accurate, repair technologies to maintain this long term storage will be economically viable. Repairing mistranscribed DNA has other utility...

Repairing DNA damage by, say, aging.

 

Share this post


Link to post
Share on other sites

Considering it's been said our DNA contains a lot of 'junk', as in segments not used, I can't help but wonder if the idea of incorporating such data DNA into ones own will be looked at.  As in, you'll become your own storage device.

Share this post


Link to post
Share on other sites
Posted (edited)

Wasn't that a theme in one of those Star Treks, with a Klingon ? Well, their DNA probably isn't all that complicated anyway :-)

Seriously, it is not yet in all detail known which parts code what and how they interact and if a change somewhere may have side effects. Though geneticists are getting better, experiments like those recently shown off from a Chinese geneticist still have a lot of uncertainty in them.

Also, i fear, if somebody wants that data, they'll simply graze your bathroom for a hair or scale ....

 

Edited by Green Baron

Share this post


Link to post
Share on other sites

@LordFerret I commend ‘Blood Music’ by Greg Bear to you. :)

@Green Baron - I believe there are ‘safe harbour’ loci which can accept transgenes without risk of creating an oncogene in the process. Whether they can safely accommodate a big enough segment of DNA to be useful for data storage, is another matter.

Share this post


Link to post
Share on other sites
15 hours ago, LordFerret said:

Considering it's been said our DNA contains a lot of 'junk', as in segments not used, I can't help but wonder if the idea of incorporating such data DNA into ones own will be looked at.  As in, you'll become your own storage device.

It's not junk DNA, we do not know what it's for.

Share this post


Link to post
Share on other sites
Posted (edited)
32 minutes ago, Cassel said:

It's not junk DNA, we do not know what it's for.

When we remove the "junk DNA" from cells, the cells tend to die. So yeah, it's not really junk.

Edited by mikegarrison

Share this post


Link to post
Share on other sites

I'm sure this was the plot of a decent episode of Star Trek TNG. The Drumhead.

Share this post


Link to post
Share on other sites
Posted (edited)

They should develop a friendly virus performing DNA background defragmentation and optimization.

P.S.
Maybe also with a DNA compressing utility, DNA checksum verifier, etc.

Edited by kerbiloid

Share this post


Link to post
Share on other sites

That ‘junk’ DNA could be where the instinctive memories are coded. Instincts have to be coded somewhere. 

Share this post


Link to post
Share on other sites
Posted (edited)

I understand you, @Cassel. Keep in mind that Scientific American is a popular science magazine. I am by no means an expert for these things i only follow the parts that deal with paleo-anthropology, and that is complicated enough ... in most parts too hard to understand in detail. In this application, findings are still overthrown too quickly to care about the details ;-)

Knowledge about which parts code what and how will probably be improved and defined more sharply in the future, when we understand how it codes a living organism in detail.

Until then, seen from a high level, DNA is "only" a highly sophisticated coding mechanism. Nature uses it to code life in a way that nothing changes too quickly but yet stays adaptable. We can also use it to code things. We will see if it is better than other encodings, digital or analogue. Either way, to transmit DNA-encoded things over wire or ether, it still would have to be digitized or made analogue, requiring a de- and encoder on both ends, if i am not mistaken.

Edited by Green Baron

Share this post


Link to post
Share on other sites
2 hours ago, Cassel said:

This term is misleading, why scientists will not remove such nonsensical concepts?


Indeed, and they have been so successful in getting rid of other nonsensical terms like ATM Machine, PIN Number, DC Comics, UPC Code, HIV Virus, and LCD Display.

Once a term gets into the public mind, scientists have no control over it's popular usage.

Share this post


Link to post
Share on other sites

Scientists should take care that the terms they use to describe discoveries are clear even to ordinary people. They are getting salaries from the government just to work for the good of all of us. If such terms are created, it means that someone did his job wrong.

Share this post


Link to post
Share on other sites
This thread is quite old. Please consider starting a new thread rather than reviving this one.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.