Membrane proteins can easily be placed in the vesicle simulation. They just need to be oriented with the membrane on the xy-plane of their coordinate system. The user should also provide an input for the z-height of the membrane relative to the macromolecules coordinates.
Biggest issues for implementation:
- memory concerns for sampling of atoms to electrostatic potential for GPUs (and time for CPU)
- ground truth annotations of the particles: class mask needs to be generated per vesicle and classes need to remain unique for each simulation
Membrane proteins can easily be placed in the vesicle simulation. They just need to be oriented with the membrane on the xy-plane of their coordinate system. The user should also provide an input for the z-height of the membrane relative to the macromolecules coordinates.
Biggest issues for implementation: