[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [IMP-dev] adding periodic boundary conditions?



Actually that is already doable using the IMP::core::TransformationSymmetry score state (or particle state?).
For your XYZ box, copy your system into 8 clones, and apply a different TransformationSymmetry to each of them,
corresponding to the different translation. Just note that to make the calculation efficient, the score should not
include the inter clone score terms. Also the optimizer/samplers must be applied only to the original copy.


On Fri, Nov 15, 2013 at 2:06 PM, Barak Raveh <" target="_blank">> wrote:
I am considering the addition of periodic boundary conditions support to the BD simulation. I guess it would apply to any optimizer in IMP - what would be the technical complications? I guess naively,what is needed is mainly to reformulate the restraints in a periodic way, but I wonder if it affects any sort of score evaluator optimizations (ClosePairs, etc. - the latter is probably easy to handle  as it mainly affects particles at the boundary). In the future it will also allow periodically optimized restraints (such as ones using FFT).

Barak

p.s. in my particular case, I thought of PBC over the x,y plane only, if that matter for anything, which I believe it doesn't

_______________________________________________
IMP-dev mailing list
">
https://salilab.org/mailman/listinfo/imp-dev