You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@mallanos and @mattholc weren't as enthusiastic about this as me and @nbruciaferri, so we decided to not separate the coordinates from the molsetups, unless it really is necessary to address #131
Arguments in favor of separating coords from molsetups:
similarity to the design of openmm (topology and positions are separate) and RDKit (conformer objects are option in molecule objects)
docking results are small trajectories
could parameterize molecules without calculating their 3D coordinates
Against:
not enough use cases for coord-less (pun not intended) molsetups to justify the risk of API user mistakes when molsetup and coords can be manipulated separately
the positions of CG/G pseudo atom and generally offsite atoms depend on the positions of other atoms, even though this already partially mitigated because calculating the positions of offsite atoms is a two step process and the first step does not depend on coordinates.
Evaluate if coords can be eliminated from MoleculeSetups and remove them if possible.
The text was updated successfully, but these errors were encountered: