A Privacy Preserving and Safety-Aware Semi-supervised Model for Dissecting Cancer Samples
Research in cancer genomics has proliferated with the advent of microarray technologies. These technologies facilitate monitoring of thousands of genes in parallel, thus providing insight into disease subtypes and gene functions. Gene expression data obtained from microarray chips are typified by few samples and a large number of genes. Supervised classifiers such as support vector machines (SVM) have been deployed for prediction task. However, insufficient labeled data have resulted in a paradigm shift to semi-supervised learning, in particular, transductive SVM (TSVM). Analysis of gene expression data using TSVM revealed that the performance of the model degenerates in the presence of unlabeled data. We address this issue by using a representative sampling strategy which ensures safety of the classifier even in the presence of unlabeled data. We also address the issue of privacy violation when classifier is shipped to other medical institutes for analysis of shared data. We propose a safety aware and privacy preserving TSVM for classifying cancer subtypes. Performance of TSVM with SVM and accuracy loss of the proposed TSVM are also analyzed.
KeywordsGene expression Transductive support vector machine Safety Privacy
The research was financially supported by Department of Information Technology, Government of Kerala and the facilities were provided by Indian Institute of Information Technology and Management - Kerala.
- 2.Camara, F., Samb, M.L., Ndiaye, S., Slimani, Y.: Privacy preserving RFE-SVM for distributed gene selection. Int. J. Comput. Sci. Issues 154–159 (2012)Google Scholar
- 6.Guo, S., Zhong, S., Zhang, A.: Privacy preserving calculation of fisher criterion score for informative gene selection. In: 2014 IEEE International Conference on Bioinformatics and Bioengineering (BIBE), pp. 90–96. IEEE (2014)Google Scholar
- 7.Haferlach, T., Kohlmann, A., Wieczorek, L., Basso, G., Te Kronnie, G., Bn, M.C., De Vos, J., Hernndez, J.M., Hofmann, W.K., Mills, K.I., Gilkes, A.: Clinical utility of microarray-based gene expression profling in the diagnosis and subclassifcation of leukemia: report from the International Microarray Innovations in Leukemia Study Group. J. Clin. Oncol. 28(15), 2529–2537 (2010)CrossRefGoogle Scholar
- 8.Hruschka, E.R., Covoes, T.F.: Feature selection for cluster analysis: an approach based on the simplified Silhouette criterion. In: 2005 International Conference on Intelligent Agents, Web Technologies and Internet Commerce, pp. 32–38 (2005)Google Scholar
- 9.http://orange.biolab.si.datasets.psp. Accessed 6 June 2014
- 11.Li, X.: Privacy preserving clustering for distributed homogeneous gene expression data sets. In: Innovations in Data Methodologies and Computational Algorithms for Medical Applications, pp. 184–207. IGI Global (2012)Google Scholar