Model building deletions

For peptides and nucleic acids, the system should provide rapid generation of a model from sequence data in any of the commonly observed conformations (e.g., a-helix, /J-sheet, /2-turn, B-DNA, Z-DNA). For peptides, it should be possible to make insertions or deletions in the sequence easily and to mutate side chains for homology model-building applications, where the sequence of the unknown structure is mapped onto the three-dimensional structure of a sequentially homologous protein whose structure has previously been determined by X-ray crystallography. 

Figure 7.11 Flow chart of forward stepwise model building to a full covariate model followed by backwards deletion to a reduced, final covariate model.

Step 4 Build the model using covariates using stepwise model building strategies (usually forward selection followed by backwards deletion). 

Validation without an independent test set. Each application of the adaptive wavelet algorithm has been applied to a training set and validated using an independent test set. If there are too few observations to allow for an independent testing and training data set, then cross validation could be used to assess the prediction performance of the statistical method. Should this be the situation, it is necessary to mention that it would be an extremely computational exercise to implement a full cross-validation routine for the AWA. That is. it would be too time consuming to leave out one observation, build the AWA model, predict the deleted observation, and then repeat this leave-one-out procedure separately. In the absence of an independent test set, a more realistic approach would be to perform cross-validation using the wavelet produced at termination of the AWA, but it is important to mention that this would not be a full validation. 

In separate calculations, build models of the two carbocations and submit them to AMI calculations of their energies. Use a geometry optimization. When you build the models, most programs will require you to build the skeleton of the hydrocarbon that is closest in structure to the carbocation and then to delete the required hydrogen and its free valence. 

We use the modeling language SMV and the model-checker NuSMV2 3]. SMV enables the declaration of integer variables and constraints on their behavior. NuSMV builds transparently the Cartesian product of the ranges of all variables. When no constraint is declared, all the combinations of variable values (i.e., states) are possible and all transitions between each pair of states are implicitly declared. Constraints are then added to delete undesired states and transitions. As for variables, time is discrete. It is modeled by the operator next (). NuSMV is well-adapted to our variable-oriented modeling approach. Moreover, the implicit transition declaration is convenient for modeling the whole physically possible behavior. 

---