Chain-Anchored Systems
Chain-anchored systems are unbranched fragments specially annotated to describe the ring/chain topology of a molecule. An advantage of these fragments is that the structural similarity of molecules sharing a given fragment is significantly greater than that of molecules grouped using a standard, non-annotated fragment. The Chain-anchored Systems method of class generation is selected by choosing the appropriate dialog option after selecting the menu item CLASSES>>Generate.
You can specify an upper and lower bound on the number of atoms in each fragment; note that memory usage and processing time increase significantly as the maximum fragment size increases. Additional parameters include the [homogeneity and redundancy settings] and options to force closed rings (see below) and to incorporate fuzzy atoms and bonds.
The Force closed option in the Rings field is available on the Class generation options dialog, which is displayed when you click on the Change General Options button in the Generate classes dialog, assuming that Chain-anchored Systems is the selected method of class generation. Force closed rings specifies that class scaffolds will not contain partial rings: if the scaffold captures any portion of a ring, it will always capture the entire ring. The use of this option tends to produce classes with less structural diversity, since only molecules having identical ring structures will be placed into the same class. Note that the effect of lowering class diversity can be mitigated if desired by using the fuzzy atoms/bonds option discussed in a subsequent subsection. Activating the closed rings option tends to result in lower memory usage, but it may cause class generation to take longer.
The number of fragments generated in the Chain-anchored Systems approach can become very large even for data sets of modest size, particularly when the Force closed Rings option is turned off. You should always ensure that adequate memory resources are available to support any such change.