mesh ply files from masif preprocessing scripts were not fed into atoms_to_points_normals

[jiyanbio]

atoms_to_points_normals() function was decrbied clearly in your paper. But the trainning process by using mesh ply files from masif preprocessing scripts skipped it.
My question is that, in the inference step, If such model directly took a pdb as the input which would be convert to points by atoms_to_points_normals() function, does it make sense when input files are different between training and inference?

[FreyrS]

Hi @jiyanbio,

dMaSIF calls atoms_to_points_normals() also during training and only extracts the interface labels from the mesh ply files.
During training we simplified the process such that the surfaces are only generated before the first epoch and not during every epoch, see

dMaSIF/main_training.py

Line 55 in 0dcc26c

train_dataset = iterate_surface_precompute(train_loader, net, args)

, this line eventually leads to

dMaSIF/data_iteration.py

Line 126 in 0dcc26c

net.preprocess_surface(P1)

where the surface of each protein is generated.