Metadata primary

Metadata are most easily provided in pick lists that appear as dropdown menns in the nser interface. New files in the system wonld typically receive mnltiple metadata. To stay with the previous example, each new metabolite would receive metadata for Species, Matrix, Tissue, and so forth. There are certain cases in which previously selected metadata (primary metadata) may affect the number of entries in a pick list of other metadata (secondary metadata). Selecting the primary metadata would then lead to exclusion of several nonapplicable secondary metadata or exclusion of an entire meta-key. 

Web in the life of the medicinal chemist. One may see the development of alerting services for the primary medicinal chemistry journals. The Web-based information search process could be replaced by a much more structured one based on metadata, derived by automated processing of the original full-text article. To discover new and potentially interesting articles, the user subscribes to the RSS feeds of relevant publishers and can simply search the latest items that appear automatically for keywords of interest. The article download is still necessary, but it may be possible for the client software to automatically invoke bibliographic tools to store the found references. Another application of the Chemical Semantic Web may be as alerting services for new additions to chemical databases where users get alerts for the new additions of structures or reactions. 

The Protein Data Bank (PDB http //www.pdb.org) is the worldwide repository of three-dimensional structural data of biological macromolecules, such as proteins and nucleic acids (Berman et al. 2003). The Protein Data Bank uses several text file-based formats for data deposition, processing, and archiving. The oldest of these is the Protein Data Bank format (Bernstein 1977), which is used both for deposition and for retrieval of results. It is a plain-text format whose main part, a so-called primary structure section, contains the atomic coordinates within the sequence of residues (e.g., nucleotides or amino acids) in each chain of the macromolecule. Embedded in these records are chain identifiers and sequence numbers that allow other records to reference parts of the sequence. Apart from structural data, the PDB format also allows for storing of various metadata such as bibliographic data, experimental conditions, additional stereochemistry information, and so on. However, the amount of metadata types available is rather limited owing to the age of the PDB format and to its relatively strict syntax rules. 

Some of these secondary metadata may be shared by both primary metadata, and some of them are individually used for just one primary metadata. For instance, a blood sample can be taken from a living animal, and there might be a chance to take a sample of liver tissue without significantly harming the animal however, taking a sample of nerve tissue is usually not possible without killing the animal. That is, the combination of In Vivo with Whole Blood and Liver makes sense, whereas for In Vitro all the tissues apply. 

A solution is the definition of dependencies between metadata, which has consequences on the input of metadata if an object is created or uploaded. If a value is selected from the pick list for any metadata defined as primary data, available values for the secondary metadata are modified — usually reduced — according to the definition of dependencies. If only a single value remains to be valid for secondary data, the corresponding value is automatically selected. 

The BioMart data model is a simple modular schema composed of a central table, linked to its satellite tables by primary/foreign key relations. The schema can be normalized but typically includes denormalizations to achieve maximum query response optimizations. All the BioMart metadata is stored in XML configuration files on the database servers. The metadata files can be readily created and modified using MartEditor, a Java-based configuration editor. 

In any consideration of metabolomics, one must not forget that data handling is of para-moimt importance, both primary data and metadata. Collection of large datasets generated by the instrumentation required for metabolomics necessitates both comprehensive databases and sophisticated statistics to extract maximum biological relevance from the data. For metabolomics, unsupervised multivariate statistics (Fiehn, 2002 Ward et al, 2007 Weckwerth, 2008),... 

HathiTrust Digital Library http //www.hathitrust.org/home (accessed November 12, 2010). This collection includes an extensive number of full-text documents from the Middle Ages to modem times, and in all languages. It makes accessible primary documents, and is open to the public. It represents a conglomerate of metadata and virtual documents collected by a large number of academic libraries. 

---