- 會議論文
An Optimization Approach to Tumor Classification of Gene Expression Data
並列摘要
Given thousands of genes, only a small number of them show strong correlation with a certain phenotype. To identify such an optimal subset of genes is complex, which play a crucial role when classify the multiple-class genes express models from tumor samples. This paper studies the most challenge simultaneously uses a powerful feature selection and sets classifier for accuracy tumor classification. This paper proposes an optimization approach for tumor classification of gene expression data. We use a powerful intelligent genetic algorithm IGA efficient to select the most relevant genes and the parameters of Support Vector Machine (SVM). The parameters of SVM are encoded in chromosome to perform classifier optimization. The effectiveness of the proposed method is demonstrated by 4 datasets 9_Tumors, 14_Tumors, Brain_Tumors 1 and Brain_Tumors2. It is shown that IGA-SVM outperforms the existing method in terms of both the number of features and accuracy classification.
延伸閱讀
- Du, M. G., Zhang, S. W., & Wang, H. (2009). Tumor Classification Using High-Order Gene Expression Profiles Based on Multilinear ICA. Advances in Bioinformatics, 2009(), 181-189. https://doi.org/10.1155/2009/926450
- Chiu, H. S. (2004). Identifying Significant Transcriptional Modules by Genetic Algorithms from Gene Expression Data [master's thesis, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU.2004.00675
- Lin, T. C. (2006). Using Genetic Algorithm for Biomarker Discovery in Multiclass Tumor Classification [doctoral dissertation, Yuan Ze University]. Airiti Library. https://doi.org/10.6838/YZU.2006.00243
- Lo, H. Y. (2011). Characterization of two candidate tumor suppressor genes [doctoral dissertation, The University of Hong Kong]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0029-1812201200018827
- Ulfenborg, B., Levan, K. K., & Olsson, B. (2013). Classification of Tumor Samples from Expression Data Using Decision Trunks. Cancer Informatics, (2013), 53-66. https://doi.org/10.4137/CIN.S10356