That part should be OK (I had the same in Brownian optimizer). I noticed that in the MD the mass should be not optimizable, as I did as follow:
mk = IMP.FloatKey("mass")
p.set_is_optimized(mk, True)
I will try to follow the charm examples.
Thanks!
Davide
On Nov 11, 2013, at 11:07 PM, Daniel Russel wrote:
Or just not marking the particles as optimizable? At least if you use a thermostat, things should move even with no forces. It is a bit confusing (making sure particles are optimized).... You can also pass the list of particles to be optimized explicitly to the optimizer, I kind of prefer that way.
I think I was missing the forcefield (particles were not being optimized).
On Nov 11, 2013, at 4:16 PM, Daniel Russel wrote:
What aspect of MD would you like an example of? The MD class basically has two functions: one to set the time step and one to optimize. For the force field side there are already the charmm_forcefield examples. The bencmark_md_charmm.cpp provides a pretty simple example of using them together.
Yes, you make a pair score from WLC (using either SphereDistancePairScore or DistancePairScore) and then create a restraint. The first part you would have to do with the create_restraint function anyway. Although it might be cool to add a create_distance_restraint function. But we don't have such a thing.
Do I need to create a new PairScore class to use a WLC score with, e.g. container::PairsRestraint?
I'm asking because the code works with existing PairScores, but they are mainly distance scores.
Davide
On Nov 7, 2013, at 5:21 PM, Daniel Russel wrote:
Just to add, for now just create a core::PairRestraint or containerPairsRestraint. The functionality is identical to what create_restraint would do in python (in C++ create_restraint can do cooler things with compile time binding).
There's something I don't understand: if I use IMP.container.create_restraint(), I get the attribute error "AttributeError: 'module' object has no attribute 'create_restraint'", which I wasn't expecting.
Davide
On Nov 4, 2013, at 6:42 PM, Daniel Russel wrote:
It is used like any other pair score. core:: create_restraint and container::create_restraint or core::PairRrestraint/container::PairsRestraint can be used to bind it to particle pairs to make a Restraint. The container::ConsecutivePairContainer is probably useful for generating a particle pairs list.
On Monday, November 4, 2013, Davide Baù wrote:
Thanks Daniel!
The WLC class hasn't been used in forever as far as I know, but worked fine last I looked at it.
Any possibility of having a example of how to use the WLC? I looked around but did not find any.