Added handling of virtual sites + Fixed an issue in BI #7
Conversation
…types_mass) that must be used in case there is no definition of the atom mass in the itp file of the molecule we want to optimize via fetch_cg_mass_itp(ns)
swarmcg/swarmCG.py
Outdated
| @@ -20,113 +20,36 @@ | |||
| from swarmcg.shared import exceptions | |||
| from swarmcg.simulations.potentials import (gmx_bonds_func_1, gmx_angles_func_1, gmx_angles_func_2, | |||
| gmx_dihedrals_func_1, gmx_dihedrals_func_2) | |||
| from swarmcg.simulations.vs_functions import (vs2_func_1, vs2_func_2, vs3_func_1, vs3_func_2, vs3_func_3, vs3_func_4, vs4_func_2, vsn_func_1, vsn_func_2, vsn_func_3) | |||
There was a problem hiding this comment.
How about
import swarmcg.simulations.vs_functions as f
and then referencing all fucntions as f.[function_name]?
There was a problem hiding this comment.
right :)
…orrectly make CG traj whole during optimization runs analysis + Fix dihedrals handling
swarmcg/config.py
Outdated
| 'angle': [1, 2], # tested and verified: 1, 2 | ||
| 'dihedral': [1, 2, 4], # tested and verified: 1, 2, 4 -- ongoing: 9 (need to merge the 1+ dihedrals groups on plots) | ||
| 'virtual_sites2': [1], # tested and verified: 1 -- ongoing: 2 (need GMX 2020) | ||
| 'virtual_sites3': [1, 2], # tested and verified: 1, 2 -- ongoing: 3, 4 |
There was a problem hiding this comment.
So here I let only the verified ones and the program checks that the ITP uses functions we have validated as working. For evolving the code you thus need to put into these arrays the functions you want to work on, if they are not there already.
swarmcg/evaluate_model.py
Outdated
| sys.exit() | ||
| # if len(sys.argv) == 1: | ||
| # args_parser.print_help() | ||
| # sys.exit() |
There was a problem hiding this comment.
Let's no do this in fact, because we might want to execute the script from within a directory in which all files are already present
|
|
||
| coord[:, bead_id, :] = traj | ||
|
|
||
| ns.aa2cg_universe.load_new(coord, format=mda.coordinates.memory.MemoryReader) |
There was a problem hiding this comment.
Mapping the trajectory like this is a whole lot cleaner and faster than what we were doing previously
CharlyEmpereurmot
force-pushed
the
vs_suite
branch
from
00c7340 to
de23669
Compare
CharlyEmpereurmot
force-pushed
the
vs_suite
branch
from
c4bc993 to
140271f
Compare
So in principle this handles all functions for virtual sites, except 3out and 4fd that I have left for later (mostly because I don't get the input/parameters/gromacs manual).
There was also a small issue in the direct boltzmann inversion that involved utils.sma() and utils.ewma(), which are rolled back and this seems to be all good.
I did not refactorize much of the previous code, but what has been added should be ~better compliant.