Retrieval structure

Thus, the presence or absence of a given structure frequently needs to be checked and the retrieved structure (if any) has to be further processed. 

Figure 6.6. Relibase home page. Relibase provides resources for searching/retrieving structures, binding sites of receptor proteins, and chemical diagram of ligands. All nonprotein moieties (except water) in the PDB complexes are considered as ligands.

Dubois and colleagues studied the problem of overlapping fragments. They developed the program DARC-EPIOS, which can retrieve structural formulas from overlapping C-NMR data [29]. The COMBINE software uses similar techniques. 

DARC-EPIOS is an automated software for retrieving structural formulas from overlapping C-NMR data. 

Fig. 15.1-6 Selected fragment screening experiment applied to proteases and kinases. In their landmark study, Fesik et al. equilibrated hydrophobic molecules with stromelysin and detected binding by shift of NMR signals, retrieving structural information from the initial study [66]. Other studies screened fragment collections using...

Contribute strHcture data STATUS Find entries awaiting release DOWNLOAD Retrieve structure files FTP LINKS Srowse related information PREVIEW Beta-test new feati res ... 

ChemSketch is a software program developed in-house to draw chemical structures and to retrieve structures from MACCS. It is written in VAX C and will run on any of the VAX CPUs as long as they are running the VMS operating system. ChemSketch currently supports Tektronix 401x, VT-640, and VT-240 terminal emulation. 

This is used within SOCRATES in two ways either as a browsing search, an alternative to a conventional substructure search, on a full file, or as a means of making a large output from a substructure search more intelligible, by ranking the retrieved structures in order of similarity to an ideal target structure. 

Consider the case of a text search where we combine several keywords to form a specific query to meet our requirements, and so is the case here each fragment is like a keyword, which can be combined to perform a specific stracture search. When we use a particular fragment as our query or as a part of our query, the retrieved structure must contain that fragment. The list of retrieved stractures will include all those structures that contain the fragment in the specified manner in their stracture. 

The third point to note is that a substructure search results in a simple partition of the database into two discrete subsets, namely, those structures that contain the query and those that do not. All of the retrieved records are thus presumed to be of equal usefulness to the searcher, and there is no direct mechanism by which they can be ranked in order of decreasing utility, e.g., in order of decreasing probability of activity. In fact, the problem is still worse, in that not only can one not differentiate between retrieved structures, but also one cannot differentiate between different parts of the query, since a substructure search implicitly assumes that all parts are of equal importance. In other words, a substructural query implies that all features contained within it have a weight of one (or some other constant value) and that all other possible features (e.g., the screens in a screening system that have not been assigned to that query) have a weight of zero. 

The Cambridge Structural Database (CSD) is a repository for information about small molecule crystal structures. Scientists use single-crystal X-ray crystallography to determine the crystal structure of a compound. Once the structure is determined, information about the structure is deposited in electronic form in the CSD. Other scientists can search and retrieve structures from the database. 

If the retrieved structure in step (3) is not identical to the proposed structure, or if the proposed structure is not among the retrieved structures in step (5), it is best to determine whether the proposed structure is present in the databases (6) before concluding that it is invalid. If found, its spectral data are retrieved and compared with the observed spectral data (7). If there are significant spectral differences, it can be concluded that the proposed structure is incorrect. (If the spectral data are the same, this represents conflicting evidence. The verification attempt is flawed.) However, it should be noted that, at this stage (6), the absence of the proposed structure in the databases does not necessarily invalidate the chemist s assignment. The spectral properties of the proposed structure could conceivably be very similar to those of the compound(s) retrieved from the databases. 

Similarity searches have been applied to the interpretation of H and C NMR spectra and IR spectra. In one approach, a C NMR similarity search, using the spectrum of an unknown as the query, is followed by application of a maximal common substructure algorithm to identify the most common substructures in the set of structures corresponding to the retrieved spectra, If known, the molecular formula of the query can be used to reduce the number of retrieved structures prior to the latter step. The study demonstrated that the substructures predicted in this fashion usually reveal important structural features of the unknown, which can be used directly as constraints in structure generation. 

Offline preparation and presentation of chemical structure graphics is now possible on a PC using a variety of query editor programs (7), such as those listed in Figure 3. In most cases these are specific to a single system, and use a proprietary format for transmission of the query structure between the PC and the mainframe computer. For example, STN Express allows query structures to be built offline and then, after connection to STN, uploaded and searched against the STN online structure files. Retrieved structures can be downloaded onto the PC and browsed offline in STN Express. Similarly, CHEMLINK is now available for Telesystemes-DARC, while in the case of in-house systems, ChemBase provides similar capabilities for MACCS. 

---