Manual Wiggle refusing to work at low CIs.

Started by alcor29

alcor29 Lv 1

This used to happen before once in a long while, but now it's really annoying. I want to wiggle at CI 0.02 etc. and it wont move. Yes, I get around it but that's not the point. Something is amiss.

bkoep Staff Lv 1

Hmm, can you share a solution where this happens, and upload it for Scientists?

What happens with the Wiggle timer and iterator when you wiggle (under the stop button in the bottom left corner of the window)? Do the timer and iterator continue to count up while the protein remains still, or do they also freeze?

bkoep Staff Lv 1

Ah, I think I see the issue with @alcor29's solution.

Wiggle fails in extreme circumstances when you have atoms sitting very close together. This is rarely encountered when running normal Foldit actions, but will happen occasionally with Neural Net Mutate. If you try to use Wiggle immediately after Neural Net Mutate, then sometimes you will see this behavior with Wiggle "spinning its wheels."

I think we can probably update Neural Net Mutate so that this happens less often. In the meantime, I would avoid using Wiggle immediately after Neural Net Mutate, and instead do a quick Shake first; that should avert the issue.

alcor29 Lv 1

Actually I do shake after Neural Net. But, it's possible however that I forgo this time. I wonder though if this happens more often if atoms are placed close together regardless of using the NN?

bkoep Staff Lv 1

I wonder though if this happens more often if atoms are placed close together regardless of using the NN?

It could be, but I suspect this will mainly occur when atoms are extremely close (possibly with identical coordinates), in a way that would be difficult to do manually.