Support Vector Machines (SVM) review as a powerful class of supervised classification and clinical Proteomics example. The data set contain 216 subjects each of which presented with 100 quantified values (clinical Proteomics). Out of all the subjects, 121 are diagnosed as ovarian cancer and 95 are healthy subjects. Therefore, it is binary classification and the class labels are set as -1 and +1. Using "Kernel Trick" gives better accuracy (97,8%). Kernel means we project our data set to a higher dimension space where the data points are linearly separable. There are different functions for this projection such as Gaussian (Radial) basis function and polynomials.