Applicability domains

Our main application domain is the steam generator tubes of pressurized water nuclear plant. These tubes have 22.22 mm outer diameter and 1 27 mm thickness. 

Polyarylates can be blended with a wide range of commercially available thermoplastics, including polyamides, polycarbonates, polyetherketones, polyesters, and poly(phenylene sulfide), thus broadening their application domain. 

The general conclusion is that the phosphazene macromolecules synthesized and characterized up to now show a large range of properties and that they can potentially be exploited in many different applicative domains. 

The tools for in silico toxicology are broadly applied in the drug development process. The particular use of the tools is clearly context-dependent, which includes the quality of the prediction and the applicability domain of the model. 

All predictions must be taken for what they are, namely, generalizations based on current knowledge and understanding. There is a temptation for a user to assume that a computer-generated answer must be correct. To determine whether this is in fact the case, a number of factors concerning the model must be addressed. The statistical evaluation of a model was addressed above. Another very important criterion is to ensure that a prediction is an interpolation within the model space, and not an extrapolation outside of it. To determine this, the concept of the applicability domain of a model has been introduced [106]. 

In the area of predictive toxicology the applicability domain is taken to express the scope and limitations of a model, that is, the range of chemical structures for which the model is considered to be applicable [106]. Although this issue has been fundamental to the use of QSAR (and indeed any predictive technique) since its conception, there remain few reliable methods to define and apply an applicability domain in predictive toxicology. The current status of methods to define the applicability domain for use in (Q)SAR has been assessed recently by Netzeva et al. [106]. 

There is currently debate on the best methods to define the applicability domain for a model in predictive toxicology. The ultimate solution is likely to be lacking for a number of years. However, there are some initiatives that are beginning to address the issue of applicability domain, which include the use of statistical measures and also mechanistic appreciation. 

There are a growing number of tools to assess applicability domain, and a number of expert systems, for example, TOPKAT and MultiCASE, have their own measures of fit. These need to be developed and their application to larger drug libraries demonstrated. 

Netzeva TI, Worth AP, Aldenberg T, Benigni R, Cronin MTD, Gramatica P et al. Current status of methods for defining the applicability domain of (quantitative) structure-activity relationships. The report and recommendations of ECVAM workshop 52. ATLA 2005 33 152-73. 

In this section we focus on methods for the quantitation of a compound in the presence of an unknown interference without the requirement that this interference should be identified first or its spectrum should be estimated. Hyphenated methods are the main application domain. The methods we discuss are generalized rank annihilation method (GRAM) and residual bilinearization (RBL). 

We also pay attention impact of preparation technique of polycrys-tal adsorbents and applicability domains of obtained semiconductor sensors. 

We have to mention that the third type of adsorbents - monocrystalline oxides - is also feasible. From our stand-point their applicability domain deals with the studies of elementary stages of gas - solid body interaction, the results obtained being useful to manufacture sensors sensitive for specific gases. 

The description of the applicability domain is needed to express the limitations in terms of the types of chemicals properties of mechanism can be generated by the model with an acceptable reliability. 

As a result, VEGA creates a PDF file that contains all the information about the prediction, including the final assessment of the prediction, the list of the six most similar compounds found in the training and test set of the model, the list of all Applicability Domain indices and a reasoning on SAs with a brief explanation of their meaning. 

Results from VEGA underline the presence of 13 compounds with carcinogenic activity and only two non-carcinogenic compounds (perfhiorobutanesulfonic acid and perfluorotetradecanoic acid). It is crucial to underline that for each selected PFCs, VEGA provides a remarks about the performance of the prediction in fact, the model indicates that all the analyzed compounds are out of the VEGA applicability domain. 

PFTA Non Positive Compound is out of model Applicability Domain No alerts for carcinogenic activity Positive SAs for the micronucleus assay... 

It is important to underline that the use of different models is strongly recommended in case of QSARs evaluation in order to face the limitation of their applicability domain and to produce a reliable judgment. 

The evaluation for aquatic toxicity on daphnids and fish is reported in Tables 12 and 13. Bold values indicate that compounds are out of the model applicability domain (ECOSAR) or that the prediction is not reliable. ECOSAR and ToxSuite are able to predict all the selected compounds while T.E.S.T. fails in prediction for the daphnia toxicity of perfluorinated compounds (PFOS and PFOA). Tables 12 and 13 include also a limited number of experimental results provided by the model training dataset (some data are extracted from USEPA Ecotox database). Predicted results are in agreement for five compounds only (2, 3, 5, 13 and 14) for both endpoints while the predictions for the other compounds are highly variable. 

Several hydrogen sensing technologies and detection schemes were discussed in this chapter. The success of each technology is determined by its performance in the target application domain. 

But applications usually need a different type of computing environment. The reasoning task, accomplished by AI techniques, often constitutes ten percent or less of the code of an application. The majority of the code is for conventional programming tasks, such as data acquisition, data base access, numerical calculations, and graphics. In each application domain, computer hardware and software has been selected to match the needs of its tasks. In... 

A special case is the application domain of discrete functions (e.g., measurements on some spatial grid). The Fourier transform of a discrete function can be computed quite efficiently by a special algorithm (Fast Fourier Transform) at discrete points in Fourier space [132]. 

It needs to be emphasized that no matter how robust, significant, and validated a QSAR may be, it cannot be expected to reliably predict the modeled property for the entire universe of chemicals. Therefore, before a QSAR model is put into use for screening chemicals, its domain of application must be defined and predictions for only those chemicals that fall in this domain should be considered reliable. Some approaches that aid in defining the applicability domain are described below. 

Extent of Extrapolation For a regression-like QSAR, a simple measure of a chemical being too far from the applicability domain of the model is its leverage, hi [36], which is defined as... 

Residual Standard Deviation Another important approach that can be used to evaluate the applicability domain is the degree-of-fit method developed originally by Undberg et al. [40] and modified recently by Cho et al. [6]. According to the original method, the predicted y values are considered to be reliable if the following condihon is met ... 

Fig. 16.5 Computer-aided drug discovery workflow based on combination of QSAR modeling and consensus database mining as applied to the discovery of novel anticonvulsants [10]. The workflow emphasizes the importance of model validation and applicability domain in ensuring high hit rates as a result of database mining with predictive QSAR models.

---