I am performing molecular docking experiments but find that the results are "pre-determined" by the setting used. Therefore, I am looking for a general guideline on performing docking and subsequent interpretation. (I know docking is all hypothetical and needs further experimental validation, but in this question please assume, although might not be good to have such an assumption, the docking is a good method for an in silico screen.)
First, I found that choosing the correct pdb structure seems critical. In general, should we use the structure with bound ligand (either endogenous/ synthetic)? The follow-up questions for this are: should we always set flexible residue during docking to account for potential differences between the ligand to be docked and the experimentally determined structure with a ligand that has already "induced fit"? I am using vina, is there other software that is particularly useful for accounting for the potential structural differences of difference-bound states of the receptor?
Next, I would like to know what is the proper definition of a binding pocket. I acknowledge that there is software, such as AutoSite, to guess the binding pocket. However, I wonder if there are any rules/ conventions in the field to define a binding pocket. More specifically, sometimes, that software guesses any indented surface as pockets. In my current task, I am trying to dock something in an orphan receptor, the "catalytic" residue is not present/ established, while I can for sure refer back to some related orthologs. I would like to know what is a proper way to choose where to dock; or after I dock to all possible regions, how to interpret the results? Similarly, what is the proper way to define a docking grid?
Lastly, using Autodock vina as an example, I wonder whether the binding affinity/ docking score etc is correlated with the size of the molecule (or MW, or number of atoms, number of carbon, etc). One of my tasks is to try to dock some natural small metabolites, such as glucose, citrate, and lactate. I found that when I docked these molecules, the smaller molecules always has a worse score (higher kcal/ mol). Thus, I would like to know are there any precautions when interpreting the docking result with a ligand library with a wide range of MW.