The ability of the Gaussian-2, Gaussian-3, Complete Basis Set–QB3, and Complete Basis Set–APNO model chemistries to model the geometries of small water clusters

The Gaussian-2, Gaussian-3, Complete Basis Set-QB3, and Complete Basis Set-APNO methods have been used to calculate geometries of neutral clusters of water, (H2O)n, where n = 2–6. The structures are in excellent agreement with those determined from experiment and those predicted from previous high-level calculations. These methods also provide excellent thermochemical predictions for water clusters, and thus can be used with confidence in evaluating the structures and thermochemistry of water clusters.

Use of the Supermolecule Approach To Model the Syn and Anti Conformations of Solvated Cyclic 3‘,5‘-Adenosine Monophosphate

The supermolecule approach has been used to model the hydration of cyclic 3‘,5‘-adenosine monophosphate, cAMP. Model building combined with PM3 optimizations predict that the anti conformer of cAMP is capable of hydrogen bonding to an additional solvent water molecule compared to the syn conformer. The addition of one water to the syn superstructure with concurrent rotation of the base about the glycosyl bond to form the anti superstructure leads to an additional enthalpy of stabilization of approximately −6 kcal/mol at the PM3 level. This specific solute−solvent interaction is an example of a large solvent effect, as the method predicts that cAMP has a conformational preference for the anti isomer in solution. This conformational preference results from a change in the number of specific solute−solvent interactions in this system. This prediction could be tested by NMR techniques. The number of waters predicted to be in the first hydration sphere around cAMP is in agreement with the results of hydration studies of nucleotides in DNA. In addition, the detailed picture of solvation about this cyclic nucleotide is in agreement with infrared experimental results.

Study Of The Binding Of Substrate By The Phe557Val Mutant Soybean Lipoxygenase-1

Lipoxygenases are a class of enzymes which consist of non-heme iron dioxygenases that are produced by fungi, plants, and mammals and catalyze the oxygenation of polyunsaturated fatty acid substrates to unsaturated fatty acid hydroperoxide products. The unsaturated fatty acid hydroperoxide products are stereo- and regiospecific. One such lipoxygenase, soybean lipoxygenase-1 (SBLO-1), catalyzes the conversion of linoleate to 13-hydroperoxy-9(Z),11(E)-octadecadienoate (13-HPOD) and a small amount of 9-hydroperoxy-10(E),12(Z)-octadecadienoate (9-HPOD). Although the structure of SBLO-1 is known and it is the most widely studied lipoxygenase, how it binds to substrate is still poorly understood. Two competing binding hypotheses that have been used to understand and explain the binding are the head first binding model and the tail first binding model. The head first binding model predicts linoleate binds with its polar carboxylate group in the binding pocket and the methyl terminus at the surface of the binding pocket. The tail first binding model predicts that linoleate binds with its methyl terminus end in the binding pocket and the polar carboxylate group at the surface of the binding pocket. Both binding models have been used in the explanation of previous work. In previous work the replacement of phenylalanine with valine has been performed to produce the phe557val mutant SBLO-1. The mutant SBLO-1 was then used in the enzymatic oxygenation of linoleate. With this mutant, the amount of 9-HPOD that is formed increases. This result has been interpreted using the head-first binding model in which the smaller valine residue allows linoleate to bind with the polar carboxylate group of linoleate interacting with arginine-707. The work presented in this thesis confirms the regiochemical results of the previous work and further tests the head-first binding model. If head-first binding occurs, the 9-HPOD is expected to have primarily S configuration. Utilizing chiral-phase HPLC, it was found that the 9-HPOD produced by the phe557val mutant SBLO-1 is primarily S, consistent with head-first binding. The head-first binding model was also tested using linoleyl dimethylamine (LDMA), which has been shown to be a good substrate for SBLO-1 at pH 7.0, where LDMA is thought to be positively charged. This model predicts that less of the 9-peroxide should be produced with this substrate. Through the use of gas chromatography/mass spectrometry, it was found that the conversion of LDMA by the phe557val mutant SBLO-1 resulted in the formation of a 46:54 mixture of the 13-peroxide:9-peroxide. The higher amount of 9-peroxide is the opposite of what is expected for the currently proposed model suggesting that the proposed model may not be entirely correct. The results thus far have been consistent with reverse binding but not with the proposed interaction of the polar end of the substrate with arginine-707.

Advantages and Applications of Transforming Empirical Model Input Space with Dimensional Models

Model-based calibration of steady-state engine operation is commonly performed with highly parameterized empirical models that are accurate but not very robust, particularly when predicting highly nonlinear responses such as diesel smoke emissions. To address this problem, and to boost the accuracy of more robust non-parametric methods to the same level, GT-Power was used to transform the empirical model input space into multiple input spaces that simplified the input-output relationship and improved the accuracy and robustness of smoke predictions made by three commonly used empirical modeling methods: Multivariate Regression, Neural Networks and the k-Nearest Neighbor method. The availability of multiple input spaces allowed the development of two committee techniques: a 'Simple Committee' technique that used averaged predictions from a set of 10 pre-selected input spaces chosen by the training data and the "Minimum Variance Committee" technique where the input spaces for each prediction were chosen on the basis of disagreement between the three modeling methods. This latter technique equalized the performance of the three modeling methods. The successively increasing improvements resulting from the use of a single best transformed input space (Best Combination Technique), Simple Committee Technique and Minimum Variance Committee Technique were verified with hypothesis testing. The transformed input spaces were also shown to improve outlier detection and to improve k-Nearest Neighbor performance when predicting dynamic emissions with steady-state training data. An unexpected finding was that the benefits of input space transformation were unaffected by changes in the hardware or the calibration of the underlying GT-Power model.