Sgroups

In Sgroups, repeating units are enclosed in square brackets and a subscript n is placed to the right of the closing bracket. The subscript blk indicates a block copolymer and mon a monomer in a source-based representation. Superscripts indicate the orientation of the repeating units (hh = head-to-head, ht = head-to-tail, eu = either unknown) in a structure-based description. Crossing bonds (bonds... 

Fig. 8 (a) Structure-based and (b) source-based Sgroup representation of a polystyrene/ polyethylene oxide block-copolymer... 

Sgroup Data. In MDL structure storage, the attachment of structure-differentiating data directly to the structure. Such data may relate to the structure as a whole, or to atoms, bonds, fragments, or collections of atoms and bonds. Examples would include atomic partial charges on 3D models or percent composition attached to components of a formulation. 

The heats of formation of phenylalkynes were obtained from combustion by Rossini s group . For ethylbenzene they used i/f = 29.94 0.79 kJ mol and obtained the enthalpy of formation for phenylacetylene equal to 306.76 1.71 kJ mol Using the combustion results ofRossini sgroup " , for -propylbenzene 7.91 0.79 J mol Rogers obtained the enthalpy of formation for 1-phenyl-1-propyne 268.33 2.18 kJ mol The same method used for 1-phenyl-1-butyne, assuming that for -butylbenzene equals... 

The precise composition of a mixture will be recorded in fields in the registration database, but it is convenient also to display this information along with the structure. There is no established convention for such mark-up and it depends on the abilities of the underlying software. For Molfile-based systems, SGroup brackets can be used for this purpose. For stoichiometric mixtures with integral multipliers, the multiple SGroup bracket (MUL) can be used (Scheme 6.17). 

If the structure is partially known, it is valuable to record that part of the structure which is known, together with some descriptive chemically significant text (CST). Where the structure is attached to a resin or biomolecule, then the point of attachment can be indicated. The CST is attached to the whole structure as a labelled SGroup. 

Taylor Francis Group an informa business www.iaylorandfraiici.sgroup.coni... 

We have therefore introduced the notion of an SGroup as an extension to chemical structure representation, and have implemented SGroups in a MACCS-II... 

Data SGroups can be defined by any arbitrary substructure. A data value is associated with this substructural piece only, and not with the entire structure. While data SGroups can also be defined in terms of crossing bonds, it is convenient to think of them as arbitrary substructures carrying their own data. There is no hierarchy limitation on data SGroups arbitrary overlap is permitted. 

The chemical SGroups outlined above are now considered in turn. 

With a polymer SGroup, we associate the following internally supported attributes ... 

A user can search for an exact match of a query by considering or ignoring the polymer SGroups. If the SGroup definitions are considered, the user can additionally choose whether to consider or ignore (a) the polymer type, subtype, connectivity, and (b) the endgroups. 

Substructure search generally works intuitively, in that substructures which match must have corresponding polymer SGroup properties as well. For substructure search we recognise phase shift , and we traverse crossing bonds in all possible combinations. However, we do not currently allow for constructive substructure search for example, a query with two instances of a repeating polymer unit will not hit the structure in the database which includes just one instance. 

We have implemented two additional t3q>es of chemical SGroups to simplify structure display and input. 

MultGroups are chemical SGroups which are to be repeated a specified integral number of times. These SGroups may have zero or two crossing bonds. These can be used both as drawing short cuts and for display simplification (see Figure 9). 

The definition of a data SGroup (set of atoms and bonds) can be either tied directly to a chemical SGroup, or to an arbitrary set of atoms and bonds (except that once a bond is included, the two atoms of which it is composed are automatically included). Each data SGroup has associated with it a field and a value of data. The data fields to be allowed are defined by a database administrator as a part of the data dictionary. In the MACCS-II implementation, the program allows the same data field control as for data attached to the structure (text, numeric, and formatted). 

Examples of data that one might wish to tie directly to a chemical SGroup include percentage composition (e.g., of a component in a mixture or formulation, or of an SRU in a copolymer), additional information about how a polymerisation was conducted, or average molecular weight of a poljmier (see Figures 4,5, and 6). 

Data attached to arbitrary atoms (and bonds) can be used to (a) annotate a particular set of atoms, e.g., as part of a pharmacophore or as reactive sites (b) annotate fragments, e.g., with stoichiometric multipliers of a salt or solvate, as major or minor, or as active or inert (c) describe an unknown portion of a structure by attaching descriptive data to a nvdl or atom (d) explain more fully the nature of a particular site (atom), e.g., stereochemical purity, isotopic purity (see Figure 10). In short, the use of SGroup data permits a user-extensible chemical structure representation. The user can define new data fields and attach values to atoms. These values are fully a part of the connection table and are searchable both with exact match and with substructure searching. 

Within the operation of a chemical information computer program, the representation of a chemical structure will take various forms (see Figure 11). Any structural feature must be representable in each of these forms, and must have appropriate manipulation operations available with each. Central to the processing is the main connection table (generally data structures defined in a computer programming language). However, there are other major embodiments of the structure. We have implemented SGroups in all of these environments. 

In MACCS-II this representation is in the form of a Molfile. Extensions have been added to the Molfile description that allow SGroup information to be stored and transferred to other programs. 

Besides having a graphical user interface, MACCS-II also provides access to structures through MEDIT, a structure editor which is available from the command line or from a MACCS-II program. The MEDIT language has been extended to allow for the definition, identification, modification, and deletion of SGroups. 

A new side menu for drawing and manipulating structures with SGroups has been added to the MACCS-II structure drawing editor (see Figure 12). 

The display shows traditional brackets with identifiers in the lower right corner. The brackets positions are determined automatically when a chemical SGroup is defined, but may be altered by the user. The position of SGroup data is also determined at time of definition using defaults, but the user has a variety of methods which affords a great deal of flexibility in modif3dng how the data is to be displayed. 

The hierarchy of components and mixture or formulation SGroups is enforced by the user interface. The user interface also includes simplifications such that component SGroups are automatically defined when a mixture or formulation SGroup is defined but the user has the option of defining components, e.g., for the purpose of collecting multiple fragments as one component (as in the case of a salt). 

In this paper we have shown how the representation of chemical structures in computer programs can be enhanced by the addition of SGroups to the representation. We have implemented specialised handUng for classes of polymers, mixtures, and superatoms. The implementation of data SGroup allows users to add properties to the connection table and have them searchable as an integral part of the structure. 

---