new pdb -> new pkl

[fslee62]

i was able to run preprocess_pdb_to_pkl.py to make corresponding pickle files using the sample pdb data. however, i was not able to using preprocess_pdb_to_pkl.py for new pdb data. as far as i know, preprocess_pdb_to_pkl.py needs to read the *.error files, which were provided in the sample data. how can i make these *.error files for new pdb's. i saw that the *.json files were made in the process, but not thes *.error files. i must be missing something simple here. please help.

[fslee62]

i modified preprocess_pdb_to_pkl.py to run cases where there were no native.pdb's and *.error files. subsequent scoring using the pretrained proteinGCN model ran to completion.