Epithelial-Mesenchymal Transition Regulatory Network-Based Feature Selection in Lung Cancer Prognosis Prediction
Feature selection technique is often applied in identifying cancer prognosis biomarkers. However, many feature selection methods are prone to over-fitting or poor biological interpretation when applied on biological high-dimensional data. Network-based feature selection and data integration approaches are proposed to identify more robust biomarkers. We conducted experiments to investigate the advantages of the two approaches using epithelial mesenchymal transition regulatory network, which is demonstrated as highly relevant to cancer prognosis. We obtained data from The Cancer Genome Atlas. Prognosis prediction was made using Support Vector Machine. Under our experimental settings, the results showed that network-based features gave significantly more accurate predictions than individual molecular features, and features selected from integrated data (RNA-Seq and micro-RNA data) gave significantly more accurate predictions than features selected from single source data (RNA-Seq data). Our study indicated that biological network-based feature transformation and data integration are two useful approaches to identify robust cancer biomarkers.
KeywordsCancer prognosis prediction Epithelial mesenchymal transition Feature selection Data integration Network motif
This study was funded by the German Ministry of Research and Education (BMBF) Project Grant 3FO18501 (Forschungscampus MODAL).
- 5.Thousands of Samples are Needed to Generate a Robust Gene List for Predicting Outcome in Cancer, vol. 103. National Academy Sciences (2006)Google Scholar
- 12.Huang, H.-L., Wu, Y.-C., Su, L.-J., Huang, Y.-J., Charoenkwan, P., Chen, W.-Li., Lee, H.-C., Chu, W.C.-C., Ho, S.-Y.: Discovery of prognostic biomarkers for predicting lung cancer metastasis using microarray and survival data. BMC Bioinform. 16(1) (2015)Google Scholar
- 14.Schliekelman, M.J., Taguchi, A., Zhu, J., Dai, X., Rodriguez, J., Celiktas, M., Zhang, Q., Chin, A., Wong, C.-H., Wang, H., et al.: Molecular portraits of epithelial, mesenchymal, and hybrid states in lung adenocarcinoma and their relevance to survival. Cancer Res. 75(9), 1789–1800 (2015)CrossRefGoogle Scholar
- 16.Elsevier. EMT as the Ultimate Survival Mechanism of Cancer Cells, vol. 22 (2012)Google Scholar
- 19.Amin, E.M., Oltean, S., Hua, J., Gammons, M.V.R., Hamdollah-Zadeh, M., Welsh, G.I., Cheung, M.-K., Ni, L., Kase, S., Rennel, E.S., Symonds, K.E., Nowak, D.G., Royer-Pokora, B., Saleem, M.A., Hagiwara, M., Schumacher, V.A., Harper, S.J., Hinton, D.R., Bates, D.O., Ladomery, M.R.: WT1 mutants reveal SRPK1 to be a downstream angiogenesis target by altering VEGF splicing. Cancer Cell 20(6), 768–780 (2011)CrossRefGoogle Scholar
- 20.Berx, G., De Craene, B.: Regulatory networks defining EMT during cancer initiation and progression. Nat. Rev. Cancer 13(6), 97–110 (2013)Google Scholar
- 24.Chang, C.-C., Lin, C.-J.: LIBSVM: a library for support vector machines. ACM Trans. Intell. Syst. Technol. (TIST) 2(3), 27 (2011)Google Scholar
- 25.World Scientific. Integrative Network Analysis to Identify Aberrant Pathway Networks in Ovarian Cancer (2012)Google Scholar