Collections of light programs for molecular docking simulations

• gold_docking_parallel.py: parallelly run GOLD docking using concurrent feature of Python module. For example, if the computer have 64 threads, then there will be 64 docking jobs running at the same time.
• molecular_selector.py: select molecules according to molecular similarity with Tanimoto index. This is to reduce the number of molecules for further docking simulation. For example, after 1 Million ligands being screened, select top 50 as the template and search for similar molecules in the remaining database. Then docking on the selected molecules can be performed using gold_docking_parallel.py described above. This approach has been proven effective.
• pocket_finder.py: give a protein structure (in PDB) and one/more ligands (in mol2), find the binding pocket using GOLD docking. The idea is to do a brute-force scan of the predefined grid points around/in the protein.
• print_docking_summary.py: print the docking results either with bestgrid or bestligand function.

• Only the GOLD docking is supported for now. Future development will involve other docking programs especially the free-of-charge ones.
• Protein and ligand files used for docking are vital in molecular docking. Utilities such as gold_utils, openbabel, pka prediction tools etc will be wrapped for preparing the files.