Information extraction

Scientific insti uments can easily pi oduce and collect huge amounts of data. Nevertheless, it is seldom possible to read the relevant information in these data directly. Therefore more elaborate method. have to be applied for information extraction. 

An important issue, the significance of which is sometime underestimated, is the analysis of the resulting molecular dynamics trajectories. Clearly, the value of any computer simulation lies in the quality of the information extracted from it. In fact, it is good practice to plan the analysis procedure before starting the simulation, as the goals of the analysis will often detennine the character of the simulation to be performed. 

Table 5.11 Seqnence information extracted from the LC -MS -MS spectrum shown in Fignre 5.20...

Figure 4.4 The general protocol for information extraction from an herbal text (A-E) is paired with case examples from our work with the Ambonese Herbal by Rumphius. (A) Text is digitized. (B) Through either manual reading or automated extraction the plant name(s), plant part(s), and symptoms or disorders are identified. (C) These extracted data are then updated (as necessary) to reflect current names of the plants, using the International Plant Names Index (IPNI), and the pharmacological function(s) of the described medicinal plants are extrapolated from the mentioned symptoms and disorders. (D) The current botanical names are queried against a natural products database such as the NAPRALERT database to determine whether the plant has been previously examined. (E) Differential tables are generated that separate the plants examined in the literature from plants that may warrant further examination for bioactivity. (Adapted from Trends in Pharmacological Sciences, with permission.) See color plate.

Incorporating the Kirtas system with the International Plant Names Index and SNOW-MED allows movement of the historic text into an electronic format, identihcation of current plant names, and identihcation of the symptoms treated with the plants. To complete the mining of historic herbal texts for novel drug leads we use the Natural Products Alert (NAPRALERT ) database to compare the information extracted from the historic herbal text to the reports of plant use in the current literature. The NAPRALERT database provides a summary of plants ethnopharmacological use, biochemical activities, and isolated compounds [27]. By querying each plant (with the current plant name) it is possible to identify any reports in the current literature regarding the plant. As an example, Table 4.1 shows the NAPRALERT output for Cycas rumphii. 

The next steps consist of the extraction and normalization of terms from the zoned input document. To this end, we apply standard natural language processing techniques and normalize the extracted terms to their canonical form with string manipulations and morphological analysis. The former refers to the treatment of symbols (e.g., dashes), and the latter refers to variations of words due to inflection (e.g., plurals). These steps of information extraction rely on, and make extensive use of, our terminologies and ontologies. 

The UltraLink component then generates a list of value pairs with the form (SOURCE, LOCAL TERM) using the information extracted from the Metastore. It should be noted that an UltraLink is only generated if the data source contains information about the term under consideration. 

The ontology that underlies the information extraction and annotation process is solely based on taxonomic relationships. We intend to enrich our ontology with typed relationships. We are currently evaluating how typed relationships can extend the functionality of the UltraLink and how the expressivity for our ontology impacts the computational complexity of formal reasoning [9]. 

Figure 4.4 The general protocol for information extraction from an herbal text (A-E) is paired with case examples from our work with the Ambonese Herbal by Rumphius. For full caption see page 112.

Chemical structural information is one of the missing pieces in the great effort to bring biomedical research into the realm of twenty-first century information extraction and knowledge discovery paradigms. Proteins, genes, diseases, and chemical compounds constitute the major entities extracted in the biomedical domain. The ability to read structure information and substructure information and their association to other entities could have a major impact on toxicity information in particular and ADMET data in general. 

The ZEKE-PFI practitioner relies on the existence of a narrow band of long-lived, high-n Rydberg states ( ZEKE states )15 lying just below each true cation threshold. If the shift of this band of states relative to the true threshold is essentially constant, then spectroscopic information extracted from differences in band frequencies will faithfully reflect the true cation energy levels. This assumption seems to hold to an accuracy of 1 cm-1 or better in the many molecules studied, as judged... 

The procedure of processing the images of the microarray consists of addressing, segmentation, and information extraction. 

Protein-based methods rely upon the structural information extracted from the X-ray crystallographic and/or homology protein structures. These also include docking techniques for the exploration of possible binding modes of a ligand to a given transporter protein. 

CCR is easily measured by heteronuclear NMR experiments of isotopically labeled molecules. The information extracted from these experiments will significantly improve the resolution of NMR structures, especially bound conformations of weakly bound ligands, since /-couplings cannot be used in this case. The reason is the fact that the nonbound conformation significantly contributes to the averaged values of the coupling constant. 

Zero filling the FID more than a factor of two does not contribute to information extraction and any features revealed by this are artefacts. In most instances, zero filling by a factor of two amounts to an interpolation procedure benefiting primarily peak-picking. There are other procedures which can allow peak-picking interpolation between data points and the one used by the author is a simple equation to fit the maximum intensity and one point either side to a parabola and compute the position of its maximum. Bruker peak-pick table positions for instance are not separated by a multiple of the digital resolution and it would seem that they use the same or an analogous procedure. 

Information extraction Chemometrics tools are used to unlock hidden information already present in information-rich multivariate analytical instruments... 

Regarding point 2, the information extraction function of chemometrics is a very valnable one that is often overlooked, especially in the industrial world. It will be mentioned later in this chapter that this function can be nsed concurrently with the instrument specialization function, rather than relying npon additional exploratory data analysis. 

As a resnlt, process analytical applications tend to focus more on automation, minimization of maintenance, and long-term reliabihty. In this more pragmatic context, the instrument specialization function of chemometrics is used even more exclusively than the information extraction function. 

With these considerations in mind, it becomes clear that the information extraction capability of chemomet-rics can be quite useful in several PAT-relevant situations, especially during the research and development phase of a project, or when one is trying to improve customer confidence in the capability of a calibration. As a result, a short discussion of several common exploratory tools will be made here. 

The two systems analyzed were compared and contrasted on many accoimts. The infrared spectra of Figures 5 6 showed the unreacted materials on top and the reacted species below. The strong absorption band at 2200 cm-1 was attributed to the cyanate group and disappeared after reaction. The new absorption bands at 1370 cm-1 and 1570 cm-1 were due to the triazine ring. The percent conversion was monitored easily in this way. In both cases it was very high, almost 100J5 for I. Other important information extracted from the spectra was that no significant cross-reaction occurred. It was required that only physical interaction between the separate species should occiir if a true SIPN was to be obtained. 

Zimmermann M, Fluck J, Thi le TB, et al. (2005) Information extraction in the life sciences Perspectives for medicinal chemistry, pharmacology and toxicology. Curr. Top. Med. Chem. 5 785-796. 

Figure 12.5 SGX-CAT crystal status display. The webpage summarizes the status of up to 380 crystals using information extracted from the SGX LIMS. Each row represents a carousel of 19 crystals, each sguare in the row corresponding to a specific sample. Every square provides both an overlay and a dynamic link to further Information on each crystal, including the screening images. The web page is updated every 15 min.

---