Development of a Pharmacogenomics Model based on Support Vector Regression with Optimal Features Selection Approach to Determine the Initial Therapeutic Dose of Warfarin Anticoagulant Drug

Introduction: Using artificial intelligence tools in pharmacogenomics is one of the latest bioinformatics research fields. One most important drugs that determining its initial therapeutic dose difficult anticoagulant warfarin. Warfarin an oral that, due to narrow window and complex interrelationships individual factors, selection optimal challenging . Method: Inaccuracy warfarin will simply lead thrombosis or severe bleeding ultimately, patient death. Among relatively successful methods kernel-based estimation, comparison identification suitable kernels have not been researched. In present research, while carefully examining this approach, different features algorithms were analyzed based on expert opinions, appropriate subset efficient predictor variables was identified for estimation. Results: current study, a dataset collected by International Consortium used. The results showed support vector machine with kernel proposed can successfully predict ideal significant percentage patients error approximately 0.7 mg per week Conclusion: estimation conducted using least squares version regression feature strategy. way, better approach predicting presented, which significantly reduce wrong consequences