Supervised and Unsupervised Metabonomic Techniques in Clinical Diagnosis: Classification of 677- MTHFR Mutations in Migraine Sufferers

The principal limitation of this study is that we relied on the relative paucity of the sample database. However, the results seem promising since even the first multivariate analysis we performed (unsupervised PCA; Figure 8.3) showed a natural tendency in clustering T/T677 mutations. The application of this method depicts a sort of "natural normalization" of the studied data set, given that PCs correspond to the eigenvectors of the correlation matrix, which in turn correspond by definition to the covariance matrix of the standardized variables. This method is convenient when dealing with heterogeneous variables defined by different measurement units, ruling out all questionable a priori defined standardization processes. Another limitation of the study is the small number of hematochemical data inserted into the database. We decided to develop a study based on exams that are routinely performed. Therefore, we did not specifically test any hematochemical variable other than the ones listed in Table 8.2 (first column), which reflects the list of blood examinations requested by the institution where the patients were followed up. To the best of our knowledge, this is the first attempt to apply metabonomic statistical techniques to a clinical data set consisting of metabolic and instrumental data. We believe that this approach is a pilot experience to bring metabonomic (and, in general, data mining) techniques into clinical practice, when tuned for specific and well-defined groups of patients. By using a metabonomic approach, we developed a classifier to detect genetic mutations of the 677-MTHFR gene in a population of women suffering from migraine. Our database consisted of biochemical and instrumental data. Transcranial Doppler ultrasonography was used to measure the baseline values of cerebral blood flow and their variations during voluntary BH. Unsupervised and supervised approaches were used to extract the variables correlating with mutation. Our classifier showed an overall satisfactory performance: 95.9% sensitivity and 89.0% specificity. We found that cerebral blood flow velocities and response to BH are important variables in characterizing mutations in migraine. This pilot study adapting metabonomic techniques to the profiling of pathologic subjects could be a basis for the development of a clinically applicable methodology devoted to the profiling, classification, and evolutionary modeling of migraine patients.