This work established a classification model for gene selection using multiple sclerosis gene expression data. The distinction between the three feature selection algorithms and the classification models was that the feature selection algorithms were used to detect a group of discriminative genes from a large number of candidates, reducing the dimensionality of data sets, and the models were built and assessed based on the selected genes for sample predictions. In evaluating the performance of different models, four measures including Sensitivity, Specificity, Accuracy and F1 score were calculated based on the confusion matrix output by each classifier using total dataset. Sensitivity measures the proportion of true positives which are correctly identified, and Specificity measures the proportion of negatives which are correctly identified. Accuracy and F1 score measures a model��s prediction accuracy rate. All the four statistics reach their best values at 1 and worst score at 0. We assessed the four statistics, and determined a relative optimal classifier with highest Sensitivity, Specificity, Accuracy and F1 score. In this study, 8 genes were identified to be associated with multiple sclerosis. We built an SVM as the best model for sample prediction, having a predictive accuracy of around 86%. The SVM outperformed the other models as assessed by Sensitivity, Specificity, F1 score and Accuracy. The KEGG enrichment analysis suggested that the genes selected were statistically related to pathways involving apoptosis and T 5601640 cytokine-cytokine receptor interaction. Among the 8 genes, TNFSF10 had a close relationship with multiple sclerosis. Gene Ontology enrichment analysis revealed that TNFSF10 involved in the biological processes including protein kinase cascades, regulation of signal transduction and apoptosis, and the GPS1 and TRPS1 were primarily enriched in multiple sclerosis. Apoptosis is a common regulatory mechanism for maintaining normal development and homeostasis of the immune system. Because the process of eliminating auto-reactive T cells via apoptosis is impaired in multiple sclerosis, apoptosis signalingrelated genes may be strong candidate genes for TC-H 106 involvement in multiple sclerosis. According to the GeneCards database, there were six published studies referring to the relationship between TNFSF10 and multiple sclerosis, indicating TNFSF10 might have an important role in multiple sclerosis.