Closed since almost 3 years ago
Novice Overall Prediction Electron DensityThe structure of this protein has already been solved and published, but close inspection suggests that there are some problems with the published solution. We'd like to see if Foldit players can use the same electron density data to reconstruct a better model. There's a lot of copies of the same protein here, making quite the protein architecture. Trim tool highly recommended!
Memory usage:
-base shake or wiggle sidechains: about 1100 Mo
-auto structure: +1000 Mo
-running a recipe using recentbest: + 5000 Mo
(https://fold.it/recipes/108253)
-limitting graph length and mem usage has no effect so far (I didn't reach more than 3-4 undos yet)
-working on a trimmed zone keeps the base memory (about 1500 Mo) and doesn't explose mem usage when running a recipe.
If you discover other tips, please write them here ;)
In the last time the price of the memory was falling significantly. It is not a joke, I should by memory, because that 32GB is not enough in case of very big proteins (using 16 cores).
:) :) :)
A 3D viewer test for this very large protein:
Just a note on the puzzle size and issues that it causes, just a reminder that the Trim tool is built to help in these cases, as we anticipate wiggling and shaking all will cause problems if it's completely untrimmed. Now if these crashes happen on small trimmed portions, that's another issue…
