Array dataset

Fig. 13 3-D cutaway image showing the extent of conversion of the esterification occurring within the fixed bed considered in Figs. 11 and 12. The conversion was calculated from the chemical shift of the OH peak in a 4-D chemical shift image. The chemical shift image was acquired with an isotropic spatial resolution of 625 pm. The RARE image of the structure of the bed was acquired at an isotropic spatial resolution of 78 pm. Both datasets have been reinterpolated on to a common array giving an effective isotropic spatial resolution of 156 pm. The direction of flow is in the negative z direction. The grey scale indicates the fractional conversion within the bed.

The value of the BioPrint dataset is achieved from a combination of high quality in vitro data generated for each compound, and in vivo data extracted from public medical literature (see below). Relating both types of information supports the bioinformatics applications of the database. Also of value is the diversity of compounds, both chemical and biological, which are indicated for a large array of therapeutic areas. This diversity provides a good training set to develop and test various QSAR methods, and supports the cheminformatics applications of the database (Fig. 1). 

Cross comparisons across toxicogenomic datasets provides new statistical questions and subsequent challenges in data analysis. Meta-analyses may integrate datasets from multiple experimental studies consisting of different models (species, source), platforms (array type), statistical techniques (normalization) and design. The first challenge that needs to be addressed is how to properly make comparisons across datasets. To normalize datasets, better results may be achieved when data is first normalized internally and then externally (88). Secondly, equivalent and current annotation is needed to identify common genes across platforms, models, etc. 

Human Proteinpedia provides a list of phosphopeptides identified in mass spectrometry-based phosphoproteomic studies. In addition, phosphorylation and dephosphorylation data curated from the literature are mapped to corresponding site and residue of sequences provided in HPRD. Using PhosphoMotif Finder one can analyze the presence of phosphorylation-based motifs, which are derived from the literature, in any protein of interest. This is a valuable data for biomedical investigators in the development of phospho-spedfic antibodies and peptide arrays. Availability of many raw MS datasets deposited in Human Proteinpedia... 

One of the greatest challenges that accompanies the implementation of array detectors is the handling of the massive amounts of data that are generated. Since the detector is essentially a collection of many small detectors, each dataset is composed of data orders of magnitude more than from traditional experimentation. To make matters worse, this massive amount of data is collected in a time period that is... 

At present, two problems persist for the clinical application of IR spectral imaging methods. One of them is the enormous size of the datasets. In order to image an entire lymph node section, about 5x5 mm2 in size, on the order of 200 x 200, or 40000, individual spectra are collected. Although the data acquisition presents little problem and could be performed with existing array detector-based spectrometers within a few minutes, the data reduction by HCA takes over a day, even when using powerful personal computers such as the 64 bit AMD processor in our laboratory. The situation will undoubtedly improve tremendously when true 64 bit, WINDOWS-based operating system will be available to address the enormous... 

Lowess normalization methods are based on lowess (loess) scatterplot smoothing algorithms. The lowess smoother attempts to smooth contours within a dataset. Typically the lowess will be robust to genes which are active in treatment as they will be observed as outliers (45). Some normalization methods include a print-tip normalization (46), since physical location on the array and the print-tip may contribute some effect and variance beyond the biological and treatment variation. 

Recent times have seen the advent of high throughput assays such as array comparative genomic hybridization and cDNA microarray, which have lead to the rapid discovery of thousands of potential biomarkers. However, these need to be validated in tissue-based studies in large datasets to prove their potential utility. As these datasets are typically present in the form of formalin fixed paraffin-embedded tissue blocks, immunohistochemical (IHC) methods are ideal for validation. However, performing whole-section IHC on hundreds to thousands of blocks requires lot of resources in terms of reagents and time. In addition, an average block will yield less than 300 slides of 5 pm each. The tissue microarray (TMA) technique circumvents some of these problems. 

Most traditional chemometrics is concerned with two-way data, often represented by matrices. However, over the past decade there has been increasing interest in three-way chemical data. Instead of organising the information as a two-dimensional array [Figure 4.38(a)], it falls into a three-dimensional tensor or box [Figure 4.38(b)], Such datasets are surprisingly common. In Chapter 5 we discussed multiway PLS (Section 5.5.3), the discussion in this section being restricted to pattern recognition. 

Create an /V-clcment array, Label, that contains a cluster label for each of the N molecules in the dataset. Initialise Label by setting each element to its array position, thus assigning each molecule to its own initial cluster ... 

---