I am currently working on a research project, where I need to process quite large amount of .cif files to remove both free solvent molecules and part of the bound solvent molecules (defined) to further use the "modified" crystallographic data for set of theoretical calculations. I am guessing that such procedure requires modeling of a "grid" and then initially sorting components by their weight (lighter matrix should be related to the non-bound solvent molecules). However, this has turned out to be a very daunting task for me to achieve.
Maybe someone has already done something similar or at least can provide some ideas to deal with the task at hand?
Any help would be much appreciated!