Skip to main content

Über dieses Buch

This volume contains the full text of twenty ofthe twenty-one papers given at the Montreux 1990 International Chemical Information Conference in Mon­ treux, Switzerland between 24 and 26 September 1990. The one paper that is omitted was not received in time for incorporation in these Proceedings. The papers reflect the diverse nature of chemical information, an information field that has usually been in the forefront of applying new technology to solving information problems. In many ways, the electronic information revolution is still in its infancy; during the Montreux conferences, we intend to chart the dynamic interaction between chemical information and new technology. One publishing problem with an information field that moves so rapidly is the constant need to make printed information available within weeks or months of it being written. The majority of papers in this volume were written during the period May - July 1990. Conventional publishing, of course, allows authors time to proof-read their texts, to make changes and corrections and allows time for the contents to be indexed extensively. Time, however, is a luxury in the case of conference proceedings in the area of chemical, pharmaceutical and patent information at the beginning of the 1990s. We hope readers will appreciate the necessary trade-off that has had to take place between text thoroughly prepared, revised, indexed and corrected; and text that is available for general readership soon after it was written.



Development of computer tools for machine learning of generic reactions starting with specific reactions

The reasonings in organic synthesis often are of analogical type: try to adapt to a new situation, processes already described in previous cases. This is easily conceivable, if one considers there are, in chemistry, few general rules without exceptions which can provide a widely usable theoretic model.
Gérard Kaufmann, Philippe Jauffret, Christian Tonnelier, Thierry Hanser

Recent developments in reaction indexing

Chemists have access to large amounts of information about chemical reactions through online and inhouse reaction indexing systems. Existing databases continue to grow rapidly, and new ones appear — the most significant recently being the large VINITI database which has become available to the West. Users need the most effective ways to find information relevant to a query from these large and varied information sources.
Philip N. Judson

Chemical reaction databases: status and potential

The last decade saw the development of computer-readable chemical reaction databases with many useful features for chemists, particularly the ability to search by structure and reaction site. An overview of user needs, as well as the current status of major reaction databases, including ORAC, REACCS, and CASREACT, is presented. Some possibilities for future developments in the area are also suggested.
E. M. Langstaff, B. K. Sobala, B. C. Zahm

Automatic extraction of reaction information from databases using classification and learning techniques

Work under development at RISC-Linz is outlined. By means of a set of tools for manipulating synthetic information from a reaction database, learn and classification algorithms are used to extract and organise synthetic reaction class descriptions (i.e. reaction patterns). This information is to be used either in an automatic synthetic planning system or for database organisation. The general principle and philosophy of the methods used are explained and two example procedures are suggested which use these principles.
Edward S. Blurock

Statistical analysis of patents applied to a separation process

The statistical use of patents to determine the trend of research fields in an industry, country or firm is based on the postulate that there is a correlation between research and patents, and that patents can be considered as a valid indicator of technology. The truth, however, may disprove this postulate. A company may have gained such an advance in a given field that it prefers to keep its knowledge secret rather than risking, by applying for patents, to reveal a significant portion to its leading competitors. A company may, on the other hand, apply first for patents to cover technologies and products that it intends to promote, and then for ‘smoke-screen’ patents to camouflage its true choices. As such the survey analyst must crosscheck information coming from different sources so as to give the data acquired their true position.
A. Girard, M. Moureau

The impact of optical storage media on patent documentation and information

Today everyone in a documentation department seems to know quite well what patent documentation is, where and how to search patent information and how to find original patent specifications. Patent documents are more easily accessible than in former years but that there is still much specialist knowledge needed to find and to order them.
Martin A. Lobeck

Market-driven research and emerging technology

Market-driven research has become a much talked-about concept. More and more articles appearing in business sources and indeed in many trade journals are discussing strategies employed by companies as being ever more ‘market-driven’.
Anne Chapman

The integration of chemical information systems into in-house systems in a modern computer environment

Chemical information systems are concerned with storing, searching and retrieving information about chemical structures. The information stored may include graphical structural descriptions at a number of levels, and, in addition, a wide variety of numeric and textual data associated with the chemical structure. Chemical information systems are differentiated from other types of system by their ability to perform searches in which the search query is wholly or partly defined by a graphical representation. This ability to search by whole structure or partial structure (substructure searching) has ensured the evolution of several generations of specialised software systems which latterly, even if not in the earliest implementations, have enabled the chemist to communicate with the system in the natural language of chemistry, the chemical structure diagram.
William G. Town

Long term needs of large users of patent information

When the American Chemical Society published the first edition of ‘Chemical Abstracts’ in 1907, patent information had already been recognised as an important source of published information for over 50 years. Thomas Kelly’s ‘History of Public Libraries in Great Britain’ [1] refers to the enthusiastic pioneering work initiated by Bennet Woodcroft, Superintendent of Specifications at the British Patent Office as early as 1854.
J. L. Wales

User-driven database creation: the CHEMLIST file

The CHEMLIST database was created by the Central Abstracting & Indexing Service (CAIS) of the American Petroleum Institute (API) to provide a consolidated source of information useful to industry for complying with the provisions of the U.S. Toxic Substances Control Act (TSCA). CHEMLIST is the result of cooperation among potential users, a database producer (CAIS), and an online host (STN International).
Irving Zarember, Rosemarie Parker, Everett H. Brenner

A stereochemically accurate chemical substance database based on the systematic names of organic compounds

As the chemical nomenclature of organic compounds of the IUPAC type is a kind of artificial language, it is possible to analyse the components and syntax of nomenclatures using a dictionary and computer programs to build their total structure. Such techniques were first developed by Vander-Stouw of CAS in the 1970s (1) and then at the Beilstein Institute and recently by Cooke-Fox (2). However their details are not known. Moreover those programs seem not to process stereochemistry. On the other hand, Petrarca et al (3) and then Wipke (4,5) reported methods to handle stereochemistry in computers in earlier works. But these methods seemed to require direct structural diagram input and the algorithm to obtain stereo-specifically standardised connection table preserving its chemical meaning in relation to the structural diagram seems not be sufficient.
Keisuke Araki, Masanori Kaji

Interpreting the results of cancer-related tests: two online services which can help

Researchers in industry, government and academia who are concerned with the assessment of risk from potential carcinogens must rely on an experimentally diverse and qualitatively uneven body of knowledge. Studies which can contribute to the evaluation process include reports of human exposure, experiments involving whole animals and in vitro assays in bacteria, cells or tissues. Though they serve as convincing evidence, reports based on the exposure of humans to carcinogens exist for relatively few substances. Therefore, in most cases, the evaluation of human risk from exposure is based on tests in surrogate systems.
Mildred R. Green

A comparison of searching the Beilstein database on different online vendors

I am honoured to have been invited to speak at the Montreux 1990 International Chemical Information Conference. My co-author, Mary Ann Palma, and I are pleased to share with you some of the results of our evaluation of the various online loadings of the Beilstein File. Our research has led to two other presentations besides this one: a poster presentation at the Molecular Design Inc. (MDL) Software Users’ Group Meeting, May 3,1990 (‘Beilstein: the Other Chemical Reactions Database’) and a paper, ‘Searching the Beilstein Database Online: A Comparison of Systems’, which will appear in the December 1990 Database magazine. We thank the editor of Database, Ms. Paula Hane, for commissioning that article and prodding us down the path to an evaluation of the online versions of Beilstein.
Robert E. Buntrock, Mary Ann Palma

Comparison of scope and applications of currently available drug information files

Nowadays pharmaceutical research is a highly competitive as well as a very demanding business area. It has been estimated that over $ 200 million have to be invested in order to fully develop a new drug, requiring twelve years from synthesis to launch. The consequence of this prolongation of the development period is that there is only a short time in which to market the product under patent protection.
T. Busch, M. Blunck, A. Mullen, E. Müller

Improving access to drug information by linking nomenclatures and classifications

This paper discusses the possibility of automated retrieval of drug related information from a variety of databases. We have presumed that recognition of identical characterstrings in different systems does not present major problems. We have however, not looked at options to further develop semantic rules for attaining some degree of contents relevance. We demonstrate that identifying conceptual equalities through analysis of characteristics used to define these concepts is a rewarding method to match (100% correspondence) and map (a large degree of overlap) drug related information between systems.
J. D. de Vries

Integrated access to medical and pharmacological information: the unified medical language system at the National Library of Medicine

Of the many challenges facing those in the information science field today, none is more pressing than the inexorable growth of the information itself. This growth is manifested not only in the total volume of published material, but in the variety of information sources, the heterogeneity of information content, and the complexity of information formats.
Peri Schuyler

The impact of computers in biotechnology drug research

Biotechnology has informational needs which overlap, but are different from, the ‘chemical information’ required for traditional pharmaceutical drug development. As with chemical information, computers play a crucial role in organizing, storing and analyzing biotechnology data. This paper will address the main forms of information relevant to biotechnology drug development, and the role of the computer in managing it.
Polly Moore

The integration of chemical and biological information

New drug discovery remains an urgent yet inefficient discipline. The need for new therapies has never been greater, with an agingpopulation worldwide and new diseases like AIDS highlighting the limitations of our current medical armamentarium. Yet despite the expenditure of tens of billions of dollars per year in medical research in the United States alone, only about two dozen new chemical entities are introduced into medicine each year. Research and development of a new drug takes 10–20 years and well over $200 million, and can fail at any step along the way. Clearly we need to be as efficient and effective as possible in discovering and developing new drugs.
Peter Gund

Protein conformational analysis and online sequence searching: an application to drug research

Protein structure is generally described at a primary or three-dimensional level. The primary structure is the amino acid sequence of the protein, the order that the constituent amino acids occur in a chain. The amino acid sequence of a protein is determined precisely by genes and is the key to understanding the biochemistry of these macromolecules. The three-dimensional structure or the conformation of proteins is determined variously by hydrophobic interactions among the amino acids and other intramolecular forces, such as van der Waals interactions. These interactions affect the secondary, tertiary, and quaternary structure. The pattern of folding of a peptide chain into α helix or β-sheet is referred to as the secondary structure. This is determined by the interaction of closely grouped amino acids. Further folding, which involves interactions between groups distant in the protein, determines the tertiary structure. Finally, the aggregation of monomelic protein subunits into oligomers yields the quaternary structure.
Anish Mohindru, Barbara J. B. Ambrose, James F. Corning


Weitere Informationen