6 comments

  • jcims 5 hours ago
    One name you'll find associated with many of these animations is Drew Berry.

    If I had these when I was in high school in the 80s I truly think I would have gone into molecular biology. They are obviously have flaws in terms of a true representation of the process, but it makes the machine much more apparent and that's always been the thing that kept it at bay for me.

    More of this style of animation can be found in the WEHImovies channel on YouTube

    https://www.youtube.com/@WEHImovies/videos

    • rolph 43 minutes ago
      one of the most informative moments, was when instructor broke ranks and referred to figures as "cartoons", and that reset the context for a lot of things.

      figures are very sparse, for brevity. the real situation is buried in a mantle of molecules.

      the animations dont quite capture what individual molecules are doing, but give snapshots of cannonical points in the process. its a very busy bunch of reaction intermediates, and resonance structures, facilitating the exchange of functionalities.

      most important was the concept of a function-repair equilibrium machine, as action cycle of the machine is damaging, and requires immediate repair, in addition to the environmental onslaught of damages.

      picture having to check a file for corruption every time its accessed.

  • bonyt 5 hours ago
    Hey, working at the DNALC was my first job when I was in high school. I made a port of their iOS 3D brain app for Android, based on pre-rendered images (which was the style at the time - 2009-ish). It looks like it has since been taken down, which makes sense - I targeted my G1 at the time for acceptable performance, and Android broke things as it moved on. I also helped out on some web apps at the time. Great experience.

    https://dnalc.cshl.edu/resources/products/3d-brain-app.html

    https://web.archive.org/web/20230307055457/https://play.goog...

  • timschmidt 8 days ago
    • dataviz1000 5 hours ago
      > https://dnalc.cshl.edu/resources/3d/central-dogma.html

      I stepped over people huddled on the sidewalk, dirty, splicing the fiber optic cable yesterday. I wonder how long before there are little robots that do the splicing without humans?

      • jcims 4 hours ago
        My brother does that exact work.

        From what I’ve gathered the actual splicing is partly automated today and relatively straightforward if somewhat tedious. The big variable is the context. New construction should have relatively few variables.

        With repair, everything goes out the window. I just talked to him last night and he was out on a cable cut repair all night Friday. Middle of a snowstorm, maps were not accurate, repair site was very difficult to work in.

        To answer your actual question, the big barrier to adoption (in the US anyway) will be the CWA. :D

  • HPsquared 6 hours ago
    With how massively parallel the human body is, this process is copying DNA at an average rate of around 1 million miles per hour if you put all the DNA into a single string. (Consider that each human cell contains about 2 metres worth of DNA)
    • af78 4 hours ago
      A rate of 10 000 (ten thousand) RPM is mentioned in the video for certain bacteria. My background is in mechanical engineering, does RPM stand for revolutions per minute here? Sounds unbelievably fast for biochemical processes.
      • jcims 4 hours ago
        Yep

        https://en.wikipedia.org/wiki/Helicase

        The wild thing is that it doesn't have a 'gas tank' of ATP to drive the reaction, it goes this fast while being fueled one molecule at a time from the environment.

        Where does the ATP come from?

        Buckle up my mechanical engineer friend - https://www.youtube.com/watch?v=OT5AXGS1aL8

        I've watched that video a hundred times and it still gives me chills haha.

        • rolph 35 minutes ago
          the reactant molecules themselves, are primed with an ATP like a one use capacitor, it provides threshold energy, and is "consumed" as part of the reaction.
      • timschmidt 4 hours ago
        I find that a good rule is that the smaller the system, the faster the interactions.

        Also you may be interested in flagellar motors: https://www.youtube.com/watch?v=VPSm9gJkPxU

    • pama 6 hours ago
      I’d be curious about this global replication rate as a function of age.
      • my-toe-siz 10 minutes ago
        There isn't a global rate, mitotis rate depends on the cell type and many other inputs.

        For example, see Table 1: https://book.bionumbers.org/how-quickly-do-different-cells-i...

        You /could/ compute a global mean or median mitosis rate, and show how it changes/doesn't change with age, but it wouldn't say very much biologically. A narrower analysis that considers cell type and other context could be meaningful.

  • the__alchemist 4 hours ago
    It would be so cool if we could (Maybe it's been done?) do this with a simulation!
    • timschmidt 4 hours ago
      I spoke to some researchers about this while working for a Science and Technology Research center affiliated with the regional supercomputer center. I was told that there are still far too many molecules in a single cell to simulate fully, but that simulations had been run with state of the art quantum physics simulation software for some dozens or hundreds of molecules over several femtoseconds. The researcher told me that this took several weeks of supercomputer time, and that when the results were examined one take-away was that "around biological molecules, water seems to behave in an exceedingly ordered manner" as if the water molecules themselves are an integral part of the machinery, not just a medium they're suspended in.
      • the__alchemist 4 hours ago
        So cool!

        I can see how a cell is far too complicated to contemplate at this time. But, if focusing on the video of the DNA replication complex. (DNA strand + a few enzymes), I wonder if it could be in the realm of doable within the coming years or decades.

        Re water... yea... I suspect explicit solvents are the way to go. So, you are not just simulating the protein and DNA molecules, but also each water.

        • timschmidt 20 minutes ago
          I believe the simulation I spoke to the researcher about was around 100 atoms cubed. So 1,000,000 atoms. Numbers vary wildly with cell size, but a typical cell might contain 100 trillion atoms. So, a factor of 100,000 difference in scale. Which would be between 16 and 17 doublings. Around 25 years given Moore's law. The conversation happened probably 8 years ago, so only 17 years to go! lol
  • christoph123 6 hours ago
    "Intricate as this mechanism appears, numerous components have been deliberately left out to avoid complete confusion" :D
    • N_Lens 5 hours ago
      Wise choice