Predictive importance

A. N., Molecular hashkeys a novel method for molecular characterisation and its application for predicting important pharmaceutical properties of molecules, J. Med. Chem. 2000, 42, 1739-1748. 

While the resulting model is not quantitatively predictive, important observations can be made based on parametric simulation studies. It is proposed that changes in viscosity due to wafer temperature may be as large as 30%, and that such viscosity dependencies can have significant impact on fluid film thickness and transitively on removal rate. The importance of other process parameters, such as wafer curvature, is also indicated by the model. 

The water-promoted hydrolyses of a bicyclic amide, l-azabicyclo[2.2.2]octan-2-one (87), and a planar analogue, l,4-dimethylpiperidin-2-one (88), were studied using density functional theory in conjunction with a continuum dielectric method to introduce bulk solvent effects. The aim of these studies was to reveal how the twisting of the C-N bond affects the neutral hydrolysis of amides. The results predict important rate accelerations of the neutral hydrolysis of amides when the C-N bond is highly twisted, the corresponding barrier relaxation depending on the specific reaction pathway and transition state involved.85... 

Blending of chemical reactants is a common operation in the chemical process industries. Blend time predichons are usually based on empirical correlations. When a competitive side reaction is present, the final product distribution is often unknown until the reactor is built. The effects of the position of the feed stream on the reaction byproducts are usually unknown. Also, the scale-up of chemical reactors is not straightforward. Thus, there is a need for comprehensive, physical models that can be used to predict important information like blend time and reaction product distribution, especially as they relate to scale and feed position. 

Various methods have been proposed to measure the importance of inputs (Sarle, 1998) and are likely to be useful in different applications of neural nets. The two most common notions of importance are predictive importance and causal importance. Predictive importance is concerned with the increase in generalization error when an input is omitted from a network. Causal importance is concerned with situations in which an individual wants to quantify the relationship between input value manipulation and consequent output change. 

Although the localized electron model can account in a general way for metal-ligand bonds, it is rarely used today because it cannot predict important properties of complex ions, such as magnetism and color. Thus we will not pursue the model any further. 

We now know beyond a shadow of a doubt that, or general intelligence, is what carries the explanatory weight in individual differences in ability. It is that predicts important social variables and will be of vast importance in resolving many current social controversies. We know g is important. Unfortunately, we do not know whatsis. 

Without considering the stability of a PM model in an independent sample, it is possible to be unaware of the fact that some factors represent spurious associations with the outcome because of noise in the data or multiple comparisons. Furthermore, minor changes in the data set may result in the selection of different covariates. This might leave one in a quandary as to which covariates actually are of predictive importance. When statistical significance is the sole criterion for... 

The predictive importance of the Hammett relationship is impressive. It applies not only to hydrolysis reactions but also to substitution and oxidation reactions of aromatic compounds and even to enzyme-catalyzed reactions like the oxidation of phenols and aromatic amines by peroxidase (Job and Dunford, 1976). According to Exner (1972), data available in 1953 allowed prediction of 42,000 rate or equilibrium constants, of which only 3180 had been measured at the time. 

Table IX. Predicted Importance of Antibacterial Drug Discoveries 1970—1980 ...

Table XII. Predicted Importance of Microbial Metabolite Discoveries 1970—1980 by Drug Category...

The chemical natures of hormones play a predictably important role in their roles in cell signaling. Steroid hormones, for example, can enter the cell directly through the plasma membrane or can bind to plasma membrane receptors. Nonsteroid hormones enter the cell exclusively as a result of binding to plasma membrane receptors (Figure 24.8). 

Thermodynamic parameters such as A//°, A5 , and the dependence of K with T are useful for comparing reactions of different metal ions reacting with the same ligand or a series of different ligands reacting with the same metal ion. When these data are available for a set of related reactions, correlations between these thermodynamic parameters and the electronic structure of the complexes can sometimes be postulated. However, exclusive knowledge of the A//° and A.S° for a formation reaction is rarely sufficient to predict important characteristics of coordination complexes such as their structures or formulas. 

In this study the yam pullout test is applied to investigate internal mechanical properties of the plain woven fabrics. In the first step an analytical model was developed, inputs of which employs simple mechanical properties such as the fabric modtrlus, the weave angle, and the fabric deformation angles during the pullout test. This model predicts important mechanical parameters such as the weave angle variations, the yam-to-yam friction coefficient, the normal load in crossovers, the lateral forces, and the opposed yam strain within the fabric. This approach may be extended to other types of the woven fabrics. 

The technique chosen to perform the analysis was canonical variates analysis. Briefly, the method abstracts functions from combinations of PCs. It has been found with large sample sets ( 100 or more) that up to 25 PCs give optimal predictive ability. It is unwieldy to select the predictively important analytes from 25 PC dimensions. For this reason the PC-derived canonical varieties are computed wherein the dimensions available are one less than the munber of ascribed characteristics or groups. In the example noted above, therefore, the two CVs describe the 25 PC dimensions, since there are three groups in the analysis. In this example, a calibration discriminant analysis was calculated using 71 samples, and 23 prediction samples were used to test the calibration. Figure 6 illustrates the two-dimensional CVs calibration and prediction. 

These relatively simple models-Just lines and dots on a piece of paper-predict important properties of the substances around you every day. Our understanding of chemical bonding has allowed us to put together molecules that never existed before. Many of these molecules have impacted society and changed the way we live. For example, nylon, plastic, latex, and AIDS drugs were all synthesized by chemists who understood chemical bonding and knew how to put molecules together to achieve specific purposes. Flow would your life be different without them ... 

---