Sampling fidelity

Verification is the complement of calibration model predictions are compared to field observations that were not used in calibration or fidelity testing. This is usually the second half of split-sample testing procedures, where the universe of data is divided (either in space or time), with a portion of the data used for calibration/fidelity check and the remainder used for verification. In essence, verification is an independent test of how well the model (with its calibrated parameters) is representing the important processes occurring in the natural system. Although field and environmental conditions are often different during the verification step, parameters determined during calibration are not adjusted for verification. 

The effect of this expansion process is illustrated in Figure 4.4, in which all points in the sample database are drawn from a donut-shaped object. The outline of that object is quickly reproduced, with increasing fidelity as the number of units increases. It is notable that fewer units lie in the interior of the object once the map is complete than was the case when the SOM was trained on a set of points that defined a conical flask. This is a consequence of the flexibility of the geometry of the lattice in which the GCS grows. 

In a DNA array, gene-specific probes are created and immobilized on a chip (silicon wafer, nylon or glass array substrate). Biological samples are labeled with fluorescent dyes or radioactivity. These labeled samples are then incubated with the probes to allow hybridizations to take place in a high fidelity manner. After incubation, non-hybridized samples are washed away and spot fluorescent or radioactivity signals resulting from hybridization can be detected. 

Total RNA is isolated from the lymphocytes according to standard procedures and used as a template for radioactive labeled cDNA synthesis. The purified cDNA is used as probe for cDNA expression arrays. The advantages of this method as compared to other array systems are as follows (1) Radioactive-labeled probes are more sensitive than fluorescent-labeled probes and therefore need less sample RNA. (2) The primers used in the cDNA synthesis match the genes represented on the array. (3) The primer sequences are longer compared to other array systems, which increases the hybridization fidelity of RNA to the matching correct set of genes and therefore reduces mismatch reactions. 

In addition to the temporal correlation coefficient, the spatial correlation coefficient was calculated approximately for fixed values of time. Except for one of the mathematical models, all techniques showed a better temporal correlation than spatial correlation. The two correlation coefficients are cross plotted in Figure 5-6. Nappo stressed that correlation coefficients express fidelity in predicting tends, rather than accuracy in absolute concentration predictions. Another measure is used for assessing accuracy in predicting concentrations the ratio of predicted to observed concentration. Nappo averaged this ratio over space and over time and extracted the standard deviation of the data sample for each. The standard deviation expresses consistency of accuracy for each model. For example, a model might have a predicted observed ratio near unity,... 

This need is addressed in part by NIR-CI, in that it offers the ability to obtain high fidelity, spatially resolved pictures of the chemistry of the sample. The ability to visualize and assess the compositional heterogeneity and structure of the end product is invaluable for both the development and manufacture of solid dosage forms. NIR chemical images can be used to determine content uniformity, particle sizes and distributions of all the sample components, polymorph distributions, moisture content and location, contaminations, coating and layer thickness, and a host of other structural details. "... 

A common problem experienced under the current automation paradigm, with the demand for greater results in shorter iteration cycles (1-2 weeks), is that screeners are required to collect data from several single assay workstations, often run on different software platforms. With the demand for results on more compounds per week, the screeners must process more samples through the assays, each with fewer data points and replicates. This trade off in quality for quantity can result in lower overall data fidelity. 

The authors wish to acknowledge a number of individuals their efforts in collecting the samples for this study including Sarah Brooks, Fidel Fajardo Rios,... 

A third possibility for the synthesis of nanomaterials in constrained volumes is the use of molds (Figure 3.1c). Advantages of this method include its simplicity, versatility, and precise control over the shape of the solid, even with intricate forms. An elegant example of this strategy is the preparation of zeolites which precisely replicate the complex microstructure of wood. To do this, Dong et al. [43] infiltrated a zeolite synthesis solution into a wood sample. After the necessary hydrothermal treatment, and subsequent calcination to remove the template as well as the wood, a zeolitic structure was obtained that reproduced with full detail and fidelity the wooden sample used as a mold. 

Overfitting is the commonest problem in multivariate statistical procedures when the number of variables is greater than objects (samples) one can fit an elephant with enough variables. Tabachnick and Fidell (1983) have suggested minimum requirements for some multivariate procedures to avoid the overfitting or underfitting that can occur in a somewhat unpredictable manner, regardless of the multivariate procedure chosen. 

---