Add missing sidechains

[Tristan Croll]

Would be great to have this as a command. Could be a simple wrapper around swapaa - simply "mutating" the residue to its current name should do the trick.

[Tom Goddard]

Eric can look into this when he gets around to adding rotamer support.

[tic20@…]

OK. I should be able to put together something usable for ISOLDE's 
purposes in the meantime. I've been meaning to add a method to grab 
instantaneous non-bonded and map energies from OpenMM (prior to running 
any minimisation/equilibration steps), which should give a nice 
objective measure for choosing rotamers (or at least weeding out those 
that clash). It gets a little tricky because often the correct rotamer 
can nevertheless end up pointing a surprisingly long way from the true 
position when naively generated based on existing coordinates (because 
the backbone atoms also need to twist slightly to shift the CA-CB axis - 
what the Richardsons refer to when they talk about their "backrub 
rotamers" method). I'm thinking of using the nonbonded energy to narrow 
down to a shortlist of possible conformations, then relax each one and 
choose the best from the results. All the necessary machinery is in 
place, so I just need to do it... but first I need to finish slogging 
through the documentation for my first serious release.

On 2018-04-11 20:06, ChimeraX wrote:

[pett]

You can certainly now "swapaa" to the same residue type.