Semi-supervised Feature Extraction for RNA-Seq Data Analysis
- 2k Downloads
It is of urgency to effectively identify differentially expressed genes from RNA-Seq data. In this paper, we propose a novel method, semi-supervised feature extraction, to analyze RNA-Seq data. Our scheme is shown as follows. Firstly, we construct a graph Laplacian matrix and refine it by using labeled samples. Secondly, we find semi-supervised optimal maps by solving a generalized eigenvalue problem. Thirdly, we solve an optimal problem via joint L2,1-norm constraint to obtain a projection matrix. Finally, we identify differentially expressed genes based on the projection matrix. The results on real RNA-Seq data sets demonstrate the feasibility and effectiveness of our method.
KeywordsFeature extraction L2,1-norm constraint Spectral regression RNA-Seq data analysis
This work was supported in part by the NSFC under grant Nos. 61370163 and 61272339; China Postdoctoral Science Foundation funded project, No. 2014M560264; Shandong Provincial Natural Science Foundation, under grant Nos. ZR2013FL016 and BS2014DX004; Shenzhen Municipal Science and Technology Innovation Council (Nos. JCYJ20140417172417174, CXZZ20140904154910774 and JCYJ20140904154645958).
- 1.Jolliffe, I.T.: Principal Component Analysis. Springer, New York (2002)Google Scholar
- 8.Cai, D., He, X., Han, J.: Spectral regression for efficient regularized subspace learning. In: IEEE 11th International Conference on Computer Vision, 2007. ICCV 2007, pp. 1–8 (2007)Google Scholar
- 9.Belkin, M., Niyogi, P.: Laplacian eigenmaps and spectral techniques for embedding and clustering. In: NIPS, pp. 585–591 (2001)Google Scholar
- 10.Nie, F., Huang, H., Cai, X., Ding, C.: Efficient and robust feature selection via joint l2, 1-norms minimization. Adv. Neural Inf. Process. Syst. 23, 1813–1821 (2010)Google Scholar