Automatic structure identification

The catalyst powder was characterized by atomic absorption (chemical composition), X-ray powder diffraction (structure identification and degree of crystallinity) and nitrogen adsorption/desorption. For the latter method, an automatic Micromeretics ASAP 2000 apparatus was used, which also allowed the determination of the pore size distribution in the mesopore and macropore region (2 nm to 300 nm). 

For structural identification of the fractions, the XH-NMR spectrometer was directly coupled via capillary tubing to the chromatograph. The injection of the sample into the HPLC system was automatically initiated by the NMR console via a trigger pulse when starting the acquisition of NMR data. Using an appropriate pulse sequence, both solvent resonances (ACN at 2.4 ppm and water at 4.4 ppm) could be suppressed simultaneously. As a result of the on-line HPLC-NMR experiment, a contour plot XH chemical shift vs. retention time was generated (see Fig. 39). Due to the efficient solvent suppression, the obtainable structural information relates to the entire chemical shift region. From the contour... 

Automatic computer identification of alkanes was reported in 1989 by Davidson. His FORTRAN program has implemented, although with many restrictions and limitations, the Recommendations A-1 through A-4 for nomenclature of acyclic hydrocarbons. The program identified all the alkane chains in a structure and was able to select a master chain. Any introduction of functionalities containing heteroatoms or any, even the simplest, cyclic characteristics led to rejection by the algorithm... 

At this point it should be noted that similar models have been used for the structural identification and damage detection issues in a number of recent studies (Functionally Pooled ARX models see Kopsaftopoulos and Fassois 2013 and the references therein) and automatic control within the Linear Parameter Varying (LPV) models framework (Toth 2010, Zhao et al. 2012). 

We are developing an expert system to automate the first step of this process, the interpretation of molecular spectra and identification of substructures present in the molecule. The automatic interpretation of spectra would by itself provide a useful tool for an organic chemist who may not be an expert spectroscopist. Also, reported algorithms for the assembly of candidate structures from known substructures, such as the GENOA program. (3-6) rely on the input of accurate and specific substructures in order to function correctly and efficiently. Identification of substructures is thus a logical starting point. 

Electron crystallography of textured samples can benefit from the introduction of automatic or semi-automatic pattern indexing methods for the reconstruction of the three-dimensional reciprocal lattice from two-dimensional data and fitting procedures to model the observed diffraction pattern. Such automatic procedures had not been developed previously, but it is the purpose of this study to develop them now. All these features can contribute to extending the limits of traditional applications such as identification procedures, structure determination etc. 

SD structure file or a file containing a list of identification numbers [Excel or Comma Separated Value (CSV) format] can be uploaded. In the Excel or CSV cases the protocol will automatically look up the structure that corresponds to the identification number, as it does for the text box input. A useful feature (see Databases under Parameters in Fig. 7) is the ability to search multiple databases in a single search. This is made possible by the use of Perl in the Pipeline Pilot protocol. The Perl code is very general and easily allows for the addition of new databases as they become available, thereby further increasing the versatility of this protocol. 

In 1991, we first introduced the one-bead one-compound (OBOC ) combinatorial library method.1 Since then, it has been successfully applied to the identification of ligands for a large number of biological targets.2,3 Using well-established on-bead binding or functional assays, the OBOC method is highly efficient and practical. A random library of millions of beads can be rapidly screened in parallel for a specific acceptor molecule (receptor, antibody, enzyme, virus, etc.). The amount of acceptor needed is minute compared to solution phase assay in microtiter plates. The positive beads with active compounds are easily isolated and subjected to structural determination. For peptides that contain natural amino acids and have a free N-terminus, we routinely use an automatic protein sequencer with Edman chemistry, which converts each a-amino acid sequentially to its phenylthiohydantoin (PTH) derivatives, to determine the structure of peptide on the positive beads. 

This review focuses on the structural aspects and discusses several approaches to computerized structural feature analysis and automatic identification of structural similarity of organic molecules with some illustrative examples relating to the structure-activity problems. 

This review discusses several approaches for the automatic identification of common structural features or structural similarity of organic molecules. The organization of the chapter is as follows. Section 2 gives an overview of the methods for structural feature analysis. Identification of common structural features is discussed in Sect. 3 with a few applications in structure-activity studies, which is subsequently followed by the identification of structural similarity in Sect. 4. The quantification of structural similarity is discussed in Sect. 5. The basic algorithms of these approaches and the relative software systems are also referred to with some illustrative examples. 

They generally lack sensitivity and a direct relation to molecular structure. GC-MS is fast, direct, and very sensitive, and the spectrum provides a result which puts identification beyond dispute. Computer-assisted systems are now available which embody extensive drug reference libraries and can be automatically searched to identify unknown spectra. The further development of chemical ionization and mass fragmentography methods using stable isotopes now permits very accurate quantitative work. 

The UniProt KB is an automatically and manually annotated protein database drawn from translation of DDBJ/EMBL-Bank/GenBank coding sequences and directly sequenced proteins. Each sequence receives a imique, stable identifier allowing unambiguous identification of any protein across datasets. The KB also provides cross-references to external data collections such as the underlying DNA sequence entries in the DDBJ/EMBL-Bank/GenBank nucleotide sequence databases, 2D PAGE and 3D protein structure databases, various protein domain... 

The standard low-resolution mass spectrum (Fig. 30.3) is computer generated, which allows easy comparison with known spectra in a computer database for identification. The peak at the highest mass number is the molecular ion (M ), the mass of the molecule minus an electron. The peak at RA = 100%, the base peak, is the most abundant fragment in the spectrum and the computer automatically scales the spectrum to give the most abundant ion as 100%. The mass spectrum of a compound gives the following information about its chemical structure ... 

Takahashi, Y, Sukekawa, M. and Sasaki, S.I. (1992). Automatic Identification Molecular Similarity Using Reduced-Graph Representation of Chemical Structure. J.Chem.Inf.Comput.ScL, 32, 639-643. 

---