back to article DNA as storage? Old and boring. Boffins now chaining monomers

Monomer-manipulating researchers writing in Nature Communications this month say they have read multi-byte sequences using mass spectroscopy and bit-storing monomers. The research paper, titled "Mass spectrometry sequencing of long digital polymers facilitated by programmed inter-byte fragmentation" claims it sets the stage …

  1. AceRimmer
    Mushroom

    Eh?

    DNA is already storage!

    1. Pompous Git Silver badge

      Re: Eh?

      But it's not "bioligical", though what it's tied to remains a mystery at this juncture.

    2. Doctor Syntax Silver badge

      Re: Eh?

      "DNA is already storage!"

      And also a polymer.

  2. Anonymous Coward
    Anonymous Coward

    This is pure hype

    This article is a very good example how the hype, driven by the cut-throat competition for funding is gradually destroying the science as we watch, a bit by imperceptible bit.

    The scientific work reported in the article appears perfectly fine, and will be useful for improving mass-spectroscopic analysis of polymers - something what is obviously useful and needed. However, if the authors were to write it up like an incremental technical advance of the already perfectly well-known and understood techniques [which it is], it would have never made it to a flashy journal [OK, Nature Comm is the third-rank of the shiny in the Nature family - but it is still plenty flashy]. Instead, it would have been a good, solid, and with luck useful work in one of the more technical journals.

    Unfortunately, this would not be much help in competition for the grants and faculty jobs. To succeed there, you need flashy publications. So the authors have to wrap their (once again, good and possibly excellent) work in hype, holding out, but not quite making, promises which could not possibly be realized, and in the process becoming a little bit more like a used-car salesmen.

    I should know - I did have to do the same thing often enough myself, and for the same reasons - and it always leaves me feeling just that little bit dirty and grubby.

    1. find users who cut cat tail

      Re: This is pure hype

      Well said.

      My thought after reading the articles: And that is why I will probably never get anything published in Nature (or even one of the many Nature FooBars). I might or might not do sufficiently excellent science -- but I am definitely incapable of such excellent bullshit.

    2. Anonymous Coward
      Anonymous Coward

      Re: To succeed there, you need flashy publications.

      This conflates two linked things. To get faculty job, you tend to need publications in certain "high impact" journals (e.g. Nature)[1], and to get published in journals like Nature, you need a certain kind of (typically) "flashy" publication. So what is happening, to an extent, is that hiring committees are in effect partly outsourcing their decision making to academic journals (and Nature being a /commercial/ journal, is even less suitable for this than those from academic societies).

      This may or may not be more significant than biases due to any "old-profs network" that might exist, weird assumptions about the "career trajectory" of applicants, or an ability to write plausible bollocks into boxes which ask for "three ways in which you would contribute to academic excellence at the University of X".

      I am less in agreement that flashy publications help with grant funding, but these things may well be rather field-dependent.

      [1] Note that the citation performance of the paper(s) is irrelevant, and only where it is published counts (this is not hyperbole, "in high impact journals" is even explicitly written down in job advertisements)

    3. John Smith 19 Gold badge
      Unhappy

      "To succeed there, you need flashy publications. "

      Indeed.

      Thus does Science embrace celebrity culture and "superstar" researchers with a talent for very positive PR--> raises lots of cash --> gets more PR.

      Anyone remember the much publicized MIT "Micro Engines" project and it's "Micro rocket" engine sub-project?

      No? I'm not surprised.

      Targeted at a rocket engine the size of an SD card with a planned T/W ratio of 1000:1.

      Completely failed to deliver the pumps to do so.

  3. Schultz
    Boffin

    Similarities and differences to DNA data storage ...

    In both cases, you need a lot of identical data-storing molecules to read out the data. For DNA, there is a well established copying mechanism, the 'polymerase chain reaction' (PCR). So you can arguably store data in a single DNA molecule and just make some 10^(large number) copies via PCR before reading the data. I don't know the current state of the art, but for some kBs of DNA data, we probably talk about milliliters of chemical solutions being processed over the course of hours to get there.

    For other molecules, there won't be such a convenient amplification mechanism, hence the chemists will have to store a big chunk of identical molecules to start with. And the stuff is truly destroyed when being read by mass spectrometry. Theoretically, you might read one bit from each molecule, but in practice you will need some 10^(decent number) extra copies to get the 'full sequence coverage' required to read out all stored data. Suggesting this method for data storage is so impractical that it should be in line for the next ignoble price.

  4. This post has been deleted by its author

  5. John Smith 19 Gold badge
    IT Angle

    "chemically interesting but digitally useless."

    Which (since El Reg is about IT) is kind of the point.

    Technically it's sort of a block co-polymer (different side chains).

    But yes the MS size relative to the polymer is enormous and I very strongly doubt it can do that analysis off 1 copy of the chain only. And of course it's a destructive read out process, and mfg is probably using a mod'ed version of the Merrifield protein synthesis method. Not exactly compact either.

    Basically all these "chemistry/biology + storage --> super duper new storage medium" suffer from the same problem. Good (possibly very good) science but p**s poor engineering. :-(

    Hints.

    It's only useful if the read/write package is even close to the same size, cost and speed of existing systems to the end user, and microSD is pretty small.

    Non destructive read out is a good thing.

    No chemical reagents needed is a good thing.

    Basic physical structure unaltered by the "drive."

    But engineering a "tape" (for example)a few atoms wide and a "write head" that can change the side groups on it as it moves (literally 1 bit being a few atoms) and a drive that can position the tape to that accuracy are very formidable multi-disciplinary tasks.

  6. Korev Silver badge
    Pint

    Citation

    Thanks for a proper title and link to the paper. ->

    Any chance you could use the DOI too in future too please. That should mean that even if $JOURNAL rejig their website then there will still be a proper identifier.

  7. John Smith 19 Gold badge
    Unhappy

    A final point on storage systems.

    At the molecular level light is huge 100s of nm, when the molecules are a few nm, or 10s of nm at most.

    That is part of what makes the readout problem so tricky. Kelvin Probe Force Microscopy might get the job done.

    There have been various projects to build miniature mass spectrometers but again, they all destroy the input medium by definition.

  8. Doctor Syntax Silver badge

    Whilst this is as wildly impractical as using DNA for data storage* it's worth pointing out that RAM, from magnetic core days onwards has generally had destructive reads. Reading has to be combined with rewriting.

    * Yes, life has been using it for this for a very long time. And, although life is good at replicating and reading data the mechanism for adding new content relies on making errors in replication and then throwing most of the results out as useless.

    1. John Smith 19 Gold badge
      Unhappy

      "worth pointing out,,RAM, from magnetic core days onwards..generally had destructive reads"

      Although IIRC most core systems did not.

      Personally I see this for archival purposes, hence like tape or backup disk, not real time updates or live DB storage.

      That said if you wanted to make multiple copies of really large chunks of the entire human knowledge base you're going to need something very compact.

  9. Roj Blake Silver badge
    Headmaster

    Mass Spectrometer, not Mass Spectroscope

    See title.

  10. Sssss

    Not just the durability under use and speed, but what about archive durability, meet Mold.

    This storage is far too big. DNA sounded wow, because it sounded like a biological cross over, very excitingly fashionable, although it suited biological environments better, with low durability and write issues in organisms. But even this new more exciting technology is not enough. By the time they sort out the issues sufficiently, we may be into better technology again. Anybody remember racetrack memory, it hadn't won the race yet. We can store on atomic level at least. The whole universe was said to be able to fit into a thimble.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon