Back to Table of Contents
When we encode a search of the text and numerical items, or a search of the 2D chemical connectivity, we are searching data which is stored explicitly in the database, and which is identified by some pre-defined name.
Thus the crystallographic R-factor is called *RFAC in the system, and its value (e.g. 0.037) is stored explicitly in the CSD.
Similarly, when we search for a 2D substructural fragment, we encode the query in terms of formal chemical descriptors (element types, number of attached terminal H atoms, bond types etc.). Again, these items are stored directly in the database as sub-fields of the 2D chemical connection table.
The situation is very different in the case of a 3D search. Here the database description of the 3D structure consists only of the atomic x,y,z-coordinates determined in the diffraction experiment, whilst the search is most sensibly conducted in terms of derived geometrical parameters as discussed earlier.
For this reason, we must first choose a set of geometrical parameters that describe the 3D structure or some feature of it. This choice is often obvious or can be deduced from previous publications on the 3D characteristics of a given fragment. In some cases, however, it is necessary to calculate a range of parameters, and examine the results to see which of them are most effective in characterising the fragment for comparison or search purposes.
Having made this choice, we assign a name to each parameter and QUEST3D calculates its numerical value for each of the fragments that pass the 2D search process. In this way, the chosen subset of geometrical parameters now become explicit and searchable under the direct control of the user.
Volume 1 Chapter 7 Selecting Suitable 3D Search Constraints.