Example Data Sources

Furthermore, a wide range of other information products has been issued by UNEP Chemicals, often with partner organizations like the International Programme on Chemical Safety (IPCS). For example, data sources about persistent organic pollutants include the UNEP Chemicals extensive website... 

The comparison of both data sources qualitatively shows a similar picture. Regions of high mobflity are located especially between the secondary structure elements, which are marked on the abscissa of the plot in Figure 7-17. Please remember that the fluctuations plotted in this example also include the amino acid side chains, not only the protein backbone. This is the reason why the side chains of large and flexible amino acids like lysine or arginine can increase the fluctuations dramatically, although the corresponding backbone remains almost immobile. In these cases, it is useful to analyze the fluctuations of the protein backbone and side chains individually. 

Groundwater monitoring is a necessary component in any investigation of subsurface contamination. A wide variety of information can be gleaned from the data including groundwater velocity and direction, and contaminant identification and concentration. These data can be combined with other observations to infer various characteristics of the contamination. Examples are source and timing of the release, and future location of the contaminant plume. 

A client can research the above topics by using publicly available data sources or commercial data collection services to obtain details about a site. The commercial data collection services can compile public records of remediation activities, citations, or fines from governmental entities and other data of interest for any property. Examples of these types of firms are Vistainfo, lAO Environmental Services and Lexis/Nexis. 

Figure 10.1 An example of the kinds of information needed by a scientist to make an effective decision about chemical follow-up. Note the variety of data sources, some of which might be unavailable to the scientist or which he or she may not be aware exists.

A variety of data sources are available to inform interactive programs, including prospective data sets, retrospective databases, expert opinion, and unpub-lished/published literature. Time horizon, that is, the length of time into the future considered in the analysis over which costs and outcomes are projected, is very important here [26]. For example, if a clinical trial or the published literature only report short-term results for a chronic condition, the outcomes may come into question. This is where decision-analytic models may come... 

One eritical factor that has been neglected in considering mechanisms of cardiac fatalities is the timeframe for various types of toxicities. For example, a majority of cocaine-related fatalities and near fatalities reported from emergency rooms are attributed to one or more types of cardiac ischemic or hypertensive episodes (Isner et al. 1986). Thus, these studies may discount the cocaine-induced arrhythmias and conduction defects as important direct causes of fatalities. Yet, if coroner reports are used as data sources (Virmani et al. 1988 Wetli and Wright 1979 Mittleman and Wetli 1984), there are great numbers of deaths in which pulmonary effusion and lack of evidence for coronary occlusion, acute myocardial infarction, or... 

The identified precursors will be sorted according their perceived risk , which means their likelihood and consequences. The likelihood can often be established from the data sources, although the possible consequences for safety are estimated and are therefore always subjective. The likelihood of the precursors is identified by establishing the aggregation level and relative frequency of the identified precursors, as discussed in Chapter 5. Table 17 shows an example of the retrieved likelihood from precursors identified in company A during a specified time period of a year. 

Figure 11.6 Principal construction of a neural network. The input and output data are examples of potential data sources of interest.

Data sources contained incomplete and sometimes inaccurate incident information-for example, on numbers of injuries and community impacts. Descriptions of incidents and causal information were sometimes vague and incomplete. 

Physical integration is mainly a matter of technical or practical import. It may be necessary to make local copies of some data sources in order to ensure reliability or to reduce access times. For example, in the WWW (World Wide Web), Internet service providers often use local caches of popular web sites to improve access time. Because of the way certain servers operate, it may be necessary to have local copies of all data. 

In the bioinformatics realm, SRS (Sequence Retrieval System) [2] is a popular system, which uses a centralized collection of data resources primarily in flat text file form and, more recently, handles XML (Extensible Markup Language) files as well. Data resources are treated in a federated manner since each is maintained in its original form. However, SRS contains a large number of cross-references between corresponding fields in various data sources, so that keyword searches can be done across them. SRS thus performs more structured searches across the information than what a simple text search provides (such as web indexes perform, for example). Even though the data model implicit in the cross-reference tables is not very deep, SRS provides a useful way for users to browse and do simple queries across a large number of data sources as well as to integrate results from some computational methods. 

Both DIVA and RS3 provide some functionality in terms of substructure searches (SSS), although it is somewhat limited. For example, DIVA searches can only be performed on data that have already been queried from the database ). This pre-queried data need to be readily available to DIVA either via RS3 or as an SD file. In the case of RS3, the inclusion of multiple data sources (e.g., searching the corporate database and an external vendor library) is not trivial. As a result, while DIVA and RS3 are very useful for SSS under certain conditions, they are not as robust when compared to the Pipeline Pilot protocol. 

---