Why Structure-based Pharmacophores

The Data Source Clean-up and Interpretation of PDB Ligand Molecules 

Data formats in the Brookhaven Protein Databank have become an intensively discussed topic in the last few years. The original PDB file format [17] was created in the late 1970s and maintained by the Research Collaboratory for Structural Bioinformatics [13]. In order to improve the organization of bibliographic 

XML technologies [22, 23] have become a general standard for storing and converting all kinds of data and technically more robust solutions than mmCif based on XML, such as PROXIML [24], were proposed but still not accepted by the Research Collaboratory for Structural Bioinformatics. 

The first part of this section describes how this interpretation is done and which assumptions have been made to retrieve plausible results. The second part will then describe the algorithms used to compute bond characteristics from geometric information, because the PDB file format and its successors include no means to specify hybridization states or bond orders, which are essential for the characterization of properties of small organic ligands. Ligand bond characteristics interpretation is a prerequisite for the step to follow the detailed description of protein-ligand interactions by pharmacophore models. 

---