Informative miRNAs Selection in Cancer Detection Using Adopted QGA

Cancer is the most popular reason of death worldwide that many people struggle with it. Although the cancer is dangerous, but if it detect in early stages increases the chance of patient survival. The miRNAs are one of the important ways for early cancer detection that it caused to return an interesting field for researches. All the miRNAs haven’t any role in cancer detection. The Quantum Genetic Algorithm (QGA) is a developed Genetic Algorithm (GA) that by using of quantum computing on top of the genetic algorithm to alleviate the pre convergence problem. The interest of this paper is to adopt the QGA for solving of informative miRNAs selection and irrelevant miRNAs removing problem. However, in the suggested algorithm, SVM classifier performance and the dimension of the selected feature vector are dependent on heuristic information for QGA. As a result, the proposed approach selects the adaptive feature subset with respect to the shortest feature dimension and the improved performance of the classifier. The performances of this method are evaluated on the popular data setwhich the experimental results show that since QGA-SVM is used as one of wrapper methods, as a result, its overall performance is better separation between normal and cancer expression for all types of cancer and better classification rate.