653 wiggle lock

Started by Rav3n_pl

Angus Lv 1

The segment was obviously moving around while being wiggled. You are using a strange screen, so I can't tell if it was a local wiggle, but I expect it was.

Where did you expect the segment to go, over to the other side of the puzzle? It's connected where it is by cut points.

Still don't see a bug, or you aren't explaining it very clearly.

Rav3n_pl Lv 1

I will try to repeat it from some start position and post another video to demonstrate.
You can check it in selection interface:

  • reset and thread one of poistions
  • select part that is cut on 2 ends
  • press wiggle
  • manipulate CI slider/constratins check box
  • put bands to space from one end or another
    If some residue is locked in space you will be unable to move it by any tool. Only global wiggle is moving it.

brow42 Lv 1

This is the "normal" behavior of local wiggle and cut points. I showed it to jflat before.

In local wiggle of a cut segment, one end is locked. If no segment is frozen, this is the low (high?) number end. If you freeze the high number end, or anywhere in the middle, the low end moves.

This does make it impossible to use local wiggle to fix two long blue cuts. A helix will be ripped apart on the first iteration. The alternative is to use freezing and zero length bands with global wiggle. Or, use the move tool and pull to get it mostly correct and global wiggle.

There's always a fixed segment in local wiggle somewhere. Try to local wiggle with a band to an all-loop protein…Just tried it, the last segment is locked.

This is part of the algorithm and I don't think it's fixable without a lot of cleverness. Perhaps they can apply the global wiggle algorithm to a limited number of segments if one of the end points is a cut.

Rav3n_pl Lv 1

I`m not sure how it really works, but maybe solution will be cheat wiggle that rest of protein is locked and anchor wiggle to segment that is not selected?