Geometrical representation structure prediction

Models that are used to predict transport of chemicals in soil can be grouped into two main categories those based on an assumed or empirical distribution of pore water velocities, and those derived from a particular geometric representation of the pore space. Velocity-based models are currently the most widely used predictive tools. However, they are unsatisfactory because their parameters generally cannot be measured independently and often depend upon the scale at which the transport experiment is conducted. The focus of this chapter is on pore geometry models for chemical transport. These models are not widely used today. However, recent advances in the characterization of complex pore structures means that they could provide an alternative to velocity based-models in the future. They are particularly attractive because their input parameters can be estimated from independent measurements of pore characteristics. They may also provide a method of inversely estimating pore characteristics from solute transport experiments. 

Highly simplified models of protein structure embedded into low coordination lattices have been used for tertiary structure prediction for almost 20 years [65, 66, 75]. For example, Covell and Jemigan [64] enumerated all possible conformations of five small proteins restricted to fee and bcc lattices. They found that the nativelike conformation always has an energy within 2% of the lowest energy. Virtually simultaneously. Hinds and Levitt [28] used a diamond lattice model where a single lattice unit represents several residues. While such a representation cannot reproduce the geometric details of helices or P-sheets, the topology of native folds could be recovered with moderate accuracy. 

This study demonstrated the development of linear model equations that relate NMR chemical shift data to atom-based struaural properties for a set of methyl-substituted norboman-2-ol compounds. The models were used to simulate chemical shift data for both the reference and prediction compounds. These simulated spectra provide excellent representations of the actual measured spectra. The new torsional descriptors proved valuable in the discrimination of the geometric structural environments of topologically identical carbon atoms for all six atom subsets. This allowed accurate identification of all 42 simulated spectra when compared with measured spectra. Therefore, speara can also be simulated for those mono-, di-, and trimethyl compounds for which measured NMR spectra do not currently exist. The accuracy of those simulations should approximate that of the simulated spectra for the 10 prediction compounds in this smdy. 

Botschwina [84] obtained two-dimensional ab initio electric DMSs of NH3" " with the eoupled electron-pair approximation (CEPA). This 2D representation included only the variation of the two vibrational coordinates that describe the ion as having structures of geometrical symmetry. With this ab initio information, Botschwina [84] predicted relative infrared intensities [84] for a number of bands with an accuracy equaling that of present-day calculations. 

---