issues
search
openmm
/
openmm
OpenMM is a toolkit for molecular simulation using high performance GPU code.
1.5k
stars
525
forks
source link
NonbondedForce parameter offsets: Negative scale factors should be allowed
#3319
Closed
jchodera
closed
2 years ago
jchodera
commented
2 years ago
3286 added a number of helpful checks for valid parameter values, but it also appears to have
restricted
sigmaScale
and
epsilonScale
NonbondedForce
parameter offset scaling factors to be nonnegative
. This does not appear to be necessary, since it should be perfectly reasonable to reduce the initial parameter values (provided the final value doesn't end up negative, but this cannot be checked ahead of time). We in fact use this in perses, since this is necessary for relative alchemical free energy calculations. (cc
https://github.com/choderalab/perses/issues/890
).
peastman
commented
2 years ago
Good catch. I'll fix that.
3286 added a number of helpful checks for valid parameter values, but it also appears to have restricted
sigmaScale
andepsilonScale
NonbondedForce
parameter offset scaling factors to be nonnegative. This does not appear to be necessary, since it should be perfectly reasonable to reduce the initial parameter values (provided the final value doesn't end up negative, but this cannot be checked ahead of time). We in fact use this in perses, since this is necessary for relative alchemical free energy calculations. (cc https://github.com/choderalab/perses/issues/890).