[gmx-users] Setting multiple coordination distances in constrained system

[Alexander Björling]

Dear users,

I have a dimeric protein system, where I would like to define two pairwise
sidechain-sidechain distances, one in each monomer. I'd like to do this
back and forth, so that I can alternately enforce long (~0.6nm) and short
(~0.3nm) distances, and follow the response of the protein. I have reason
to believe that this will be meaningful.

I'm running with "constraints = all-bonds", and so introducing new bonds or
constraints causes LINCS-related crashes. I then have to minimize and
equilibrate between runs, and this would make the above "experiment" less
easy to interpret.

I would really like to use the pull code to apply a harmonic potential on
top of all the constraints, but as I understand it that doesn't allow
defining two distances with respect to two reference groups.

Any suggestions for how to accomplish this would be most appreciated.

Cheers,
Alex