Nature v Computers the GluR2 AMPA Inhibitor2 complex switch

  • 1
  • Idea
  • Updated 2 years ago
Nature wins again! Nature v Vienna2, Vienna, NuPack & the GluR2 AMPA inhibitor complex tri-switch; a picture.  Click to see more clearly :)

Photo of eternacac

eternacac

  • 274 Posts
  • 19 Reply Likes

Posted 2 years ago

  • 1
Photo of Pi

Pi

  • 19 Posts
  • 2 Reply Likes
Would anyone care to explain what we are looking at?
Photo of eternacac

eternacac

  • 274 Posts
  • 19 Reply Likes
I think it important to remember our computer models are just that, models of reality and subject to error.

The top row are the actual native folds and energies as determined and published in research journals. Nature does this RNA tri-switch this way. I put it in puzzlemaker tri-switch and and got the values for each engine, Vienna2, Nupack, Vienna and they are listed in the lefthand column for ease of identification. I colored in the white fold representation for Nature, since nature gets it right and I printed the energies actually obtained by scientists underneath. The rows underneath Nature show the target fold and what each engine says it cannot account for (mixed red/white icons). The target fold in these rows is identical to the native fold of nature. The energies shown in these rows is the one calculated by each engine for the target state which it claims cannot exist but the large fold shown is the incorrect fold of the native state that the engine actually produces, thus showing the error of the engine compared to reality. You can see which engine comes closest to the correct energy and fold and wonder why each fails or succeeds as it does.

So I made another picture showing the areas I find interesting and the closest any of the engines gets to nature:

Photo of Pi

Pi

  • 19 Posts
  • 2 Reply Likes
Oh, thanks. I think I get it now. So you are comparing the predicted and actual energies. Interesting. Though the first image doesn't really show that. You may have made a mistake and not show the forced predicted structure (target mode), but instead the best engine guess (natural mode). Dunno.
Photo of Pi

Pi

  • 19 Posts
  • 2 Reply Likes
I meant something like this:

GGGCGAAUUCAACUGCCAUCUAGGCAGUAACCAGGAGUUAGUAGGACAAGUUUCGUCC
(((((((....((((((.....))))))........(((.....)))....)))))))
(((((((....((((((.....)))))).((.....)).............)))))))
((((((((((.(((((.......)))))..(.....)......))).....)))))))
(Edited)
Photo of spvincent

spvincent

  • 48 Posts
  • 9 Reply Likes
This, and watching the RNA flip around when changing engines in the last lab puzzle set, makes me wonder about the TB lab. How many good solutions were never submitted because they didn't satisfy NuPack?