Quantitative predictive power

Equation (3) has good quantitative predictive power and is a successful extra-thermodynamic relationship like the Hammett sigma function. No other approach to modeling complex formation equilibria, including HSAB itself, can predict log values for unidentate ligands so accurately. 

Proton transfer at the surface of a protein or biomembrane is a cardinal reaction in the biosphere, yet its mechanism is far from clarification. The reaction, in principle, should be considered as a quantum chemistry event, and the reaction space as a narrow layer, 3-5 water molecules deep. What is more, local forces are intensive and vary rapidly with the precise molecular features of the domain. For this reason, approximate models that are based on pure chemical models or on continuum physical approximations are somewhat short of being satisfactory models with quantitative prediction power. 

In the development of MINDO/3, nearly all quantities that entered the Fock matrix and the energy expression were treated as free parameters. The orbital exponents that entered into the basis, set were varied, and it was found advantageous to allow the s and p exponents to be different. In addition, the core integral V (Eq. [12a]) were also freely varied. Finally, the resonance integral, 3, was chosen to be a pair parameter rather than the average of two atomic parameters. As a result of the introduction of these parameters, the MINDO/3 model achieved a rather impressive qualitative as well as quantitative predictive power. 

The A model for pathway R is unnecessarily complex, h.is weak predictive power, and is not consistent with experimental faets. The D model for pathway R is simpler, has quantitative predictive power, agrees quantitatively with a large body of data, and is not inconsistent with any experimental facts. 

Already we have mentioned that the van der Waals exponents are incorrect. This incorrectness is not just due to the modest quantitative predictive power of the van der Waals approach. Here the latter misses the underlying physical picture completely. 

This empirical principle is no more than a rule of thumb. It provides neither an explanation nor a quantitative prediction. However, we shall see that it is consistent with thermodynamics and, as we develop it, we shall see the powerful quantitative conclusions that can be drawn from thermodynamics. 

Considering a trade-off between knowledge that is required prior to the analysis and predictive power, stoichiometric network analysis must be regarded as the most successful computational approach to large-scale metabolic networks to date. It is computationally feasible even for large-scale networks, and it is nonetheless far more predictive that a simple graph-based analysis. Stoichiometric analysis has resulted in a vast number of applications [35,67,70 74], including quantitative predictions of metabolic network function [50, 64]. The two most well-known variants of stoichiometric analysis, namely, flux balance analysis and elementary flux modes, constitute the topic of Section V. 

In this and later regression equations, the 95% confidence limits are given in parentheses. Eqn. 8.1 quantitates the facilitation of hydrolysis by electron-withdrawing substituents. Because such substituents decrease the electron density on the carbonyl C-atom and render it more susceptible to nucleophilic attack, Eqn. 8.1 is compatible with a base-catalyzed reaction, as indeed shown. Eqn. 8.1, thus, leads to mechanistic insights, but its predictive power is narrow since the o parameter is available for only a few substituents. 

In summary, the above discussion illustrates how metabolic data for large series of analogous compounds may be amenable to quantitative structure-metabolism relationships. In ideal cases, such regression equations may even have some predictive power and can lead to mechanistic insights. 

Computational modeling can be a very powerful tool to understand the structure and dynamics of complex supramolecular assemblies in biological systems. We need to sharpen the definition of the term model somewhat, designating a procedure that allows us to quantitatively predict the physical properties of the system. In that sense, the simple geometrical illustrations in Fig. 1 only qualify if by some means experimentally accessible parameters can be calculated. As an example, a quantitative treatment of DNA bending in the solenoid model would only be possible if beyond the mechanical and charge properties of... 

Oprea, T. I., Waller, C. L., and Marshall, G. R. (1994). Three dimensional quantitative structure-activity relationship of human immunodeficiency virus (I) Protease inhibitors. 2. Predictive power using limited exploration of alternate binding modes../. Med. Chem. 37, 2206-2215. 

---