Great thing would be "bond" somehow part of guide to certain segments/sidechains.
That way we can align guide to parts that we "see" are in good place.
I ignored guide because I cant rotate/move it to "right" position and "align" button just making more mess.

I really liked the electron density and manually realigned it a several times doing early work on the loops, then used the button to get it tight - eventually didn't bother to realign manually but would have liked to have seen how close it was at the finish.

Also the 'align protein to map' is very slow and must really tax the compute, my laptop fan was going crazy and you would probably think twice before using on slower computers or having anything else running at the same time :-).

I used the guide for a bit, but I ended up turning it off because I wasn't sure what I was doing with it. Perhaps if I stuck at it a bit more I might have got a better sense of how the guide and protein were interacting. I'm not sure.

The first instance of "Mr. ED" was fairly instructive and points to the importance of taking the Quest To The Native (QTTN) puzzle guides somewhat seriously as practice opportunities. Still not obvious just what to do with a complete or partial band set, as just yanking furiously on the whole set is usually counterproductive and breaks the model. I'm looking forward to future versions of this sort of challenge.