zlacker

[parent] [thread] 7 comments
1. mnchar+(OP)[view] [source] 2020-04-27 02:47:14
Oh my that facepalm dreadful. Thank you! That gives me a new high-water mark for misleading biomolecular visualization computer graphics content. Snagged a copy.

When most everything is unmoving, it's "obvious"... well no, not to students, but... there's no pretense of doing anything other than stitching together an extremely selective set of "snapshots", to tell a completely bogus narrative of smooth motion.

Here it seems something like a Maya "jiggle all the things" option has been turned on. Making it sort of kind of look like you're being shown more realistic motion. But you're so not. It's the same bogus smooth narrative, now with a bit of utterly bogus jiggle. Those kinesin legs still aren't flailing around randomly. Nor only probabilistically making forward progress. And the thing it's towing still isn't randomly exploring the entire bloody space it can reach given the tether, between each and every "step". It still looks like a donkey towing a barge, rather than frog clinging to rope holding a balloon in a hurricane.

And given that the big vacuole or whatever should be flailing at the timescale defined by the kinesin feet, consider all those many much smaller proteins scattered about, just hanging out, in place, with a tiny bit of jiggle. Wow - you can't even rationalize that as being selective in "snapshots" - those proteins should just be blurs and gone.

And that's just the bogosity of motions, there's also... Oh well.

So compared with older renders, these new jiggles made it even harder to recognize that all the motion shown is bogus. And not satisfied with the old bogus motion, we've added even more. Which I suggest is dreadful from the standpoint of creating and reinforcing widespread student misconceptions. Sigh.

replies(2): >>satori+i2 >>dnauti+Rh
2. satori+i2[view] [source] 2020-04-27 03:16:14
>>mnchar+(OP)
Well, OP did say "even if it's impossible to understand" so if it is in fact in any way misleading, then my lawyers assure me that I may claim the full privileges of a contextual get-out-of-jail-free card for linking to it, and am hereby fully absolved of any intellectual harm caused to any and all individuals who may have viewed it.
replies(1): >>mnchar+nS
3. dnauti+Rh[view] [source] 2020-04-27 06:54:20
>>mnchar+(OP)
you might like this render better:

https://www.youtube.com/watch?v=DR80Huxp4y8

here's the artistic director for the inner life of the cell (the worse one) going on and on about how "beautiful" the science of biology is:

https://www.ted.com/talks/david_bolinsky_visualizing_the_won...

replies(2): >>mnchar+R01 >>marcos+Bn1
◧◩
4. mnchar+nS[view] [source] [discussion] 2020-04-27 13:52:00
>>satori+i2
Ha. I've wondered if increasing embarrassment might reduce long-term stable misconceptions in education content. Like astronomy texts getting the color of the Sun wrong. Or wing lift discussed elsewhere. But making textbooks liable for intellectual harm... wow. What might the internet, media, politics, thought and conversation look like, if we were all liable for negligent intellectual harm?
◧◩
5. mnchar+R01[view] [source] [discussion] 2020-04-27 14:47:17
>>dnauti+Rh
> artistic

Yeah. One might for example reduce reinforcement of the big-empty-cell misconception by briefly showing more realistically dense packing, eg [1], before fading out most of it to what can be easily rendered and seen. But that would be less "pretty". Prioritizing "pretty" over learning outcomes... is perhaps a suboptimal for education content.

> better

But still painful. Consider those quiet molecules in proteins, compared with surrounding motion. A metal nanoparticle might be that rigid, but not a protein.

One widespread issue with educational graphics, is mixing aspects done with great care for correctness, with aspects that are artistic license and utter bogosity. Where the student or viewer has no idea which aspects are which. "Just take away the learning objectives, and forget the rest" doesn't happen. More like "you are now unsalvageably soaked in a stew of misconceptions, toxic to transferable understanding and intuition - too bad, so sad".

So in what ways can samplings of a protein's configuration space be shown? And how can the surround and dynamics be shown, to avoid misrepresenting that sampling by implication?

It can be fun to picture what better might look like. After an expertise-and-resource intensive iterative process of "ok, what misconceptions will this cause? What can we show to inoculate against them? Repeat...". Perhaps implausibly intensive. I don't know of any group with that focus.

[1] https://www.flickr.com/photos/argonne/8592248739

replies(1): >>dnauti+Ax1
◧◩
6. marcos+Bn1[view] [source] [discussion] 2020-04-27 17:11:37
>>dnauti+Rh
At least there is some water there. But what strange force is that holding proteins together when they are completely out of alignment, and keeping the water away from everything else?
◧◩◪
7. dnauti+Ax1[view] [source] [discussion] 2020-04-27 18:13:18
>>mnchar+R01
david goodsell's pictures are fantastic. I used to work down the hall from him!
replies(1): >>mnchar+0Q1
◧◩◪◨
8. mnchar+0Q1[view] [source] [discussion] 2020-04-27 20:11:04
>>dnauti+Ax1
Agreed; cool, seems a neat guy. And much of his work is CC-BY, thus great for open education content. Hmm, the Wikimedia Commons capture of his work seems to be missing quite a bit. Oh nifty, there's now an interactive version of his 2014 "Molecular Machinery: A Tour of the PDB".[1]

[1] https://cdn.rcsb.org/pdb101/molecular-machinery/ [] http://pdb101.rcsb.org/sci-art/goodsell-gallery [] http://pdb101.rcsb.org/motm/motm-by-date [] https://cdn.rcsb.org/pdb101/molecular-machinery/

[go to top]