It is important to note that, as sequencing methodologies can vary, for example, by the cleanup method, thermal cycling regime, or whether single or bidirectional sequencing is used for analysis, the validation scheme should be carefully tailored to the application.
The validation scheme thus allows determining conditions that could be used to obtain DNA-microarray data of improved quality.
This technique was examined using a validation scheme over four measurement sites with varied aerosol levels and landscapes, ranging from croplands to tundra ecosystems, and over extended time periods.
The contribution of each type and the combination of topological information was evaluated using a cross validation scheme where 10% of the 1,700 classified proteins in the All network with at least one classified neighbor were taken at random as the test set, their functions hidden and predicted from quantities computed on the remaining 90% of classified proteins (training set).
To test the accuracy of the classification we used n-fold validation scheme leaving out two random subjects for the test phase after the training with the remaining 26 subjects, repeating 14 times the procedure.
Classification methods were evaluated using the same cross-validation scheme but with the response vector of the original training sets dichotomised using a given threshold.
Sensitivity and specificity for the resulting models (using the above cross-validation scheme) was highest with eight miRNAs, the selected diagnostic variables being: hsa-miR-96, hsa-miR-450a, hsa-miR-183, hsa-miR-9, hsa-miR-577, hsa-Let-7i, hsa-miR-27b and hsa-miR-34a (sensitivity, 1; specificity, 0.978; misclassification rate, 0.011).
Since the size of the gold standard is rather modest (116 PGSC genes ids), we used an additional validation scheme based on gene expression data.
Using a cross-validation scheme, we found that the proposed approach displays better performance than chemical-structure-based methods for the prediction of drug side effects.
With an appropriate (lambda) selected using a cross-validation scheme, all redundant features are removed from the model by forcing their coefficients to be zero.
