Name
Abstract | ||
|---|---|---|
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. © Springer International Publishing Switzerland 2015. |
Year | DOI | Venue |
|---|---|---|
2015 | 10.1007/978-3-319-22053-6_70 | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Keywords | Field | DocType |
Feature extraction,L-2,L-1-norm constraint,Spectral regression,RNA-Seq data analysis | Laplacian matrix,Data set,RNA-Seq,Pattern recognition,Computer science,Matrix (mathematics),Projection (linear algebra),Feature extraction,Artificial intelligence,Eigendecomposition of a matrix,Spectral regression | Conference |
Volume | ISSN | Citations |
9227 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 8 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Liu Jin-Xing | 1 | 40 | 16.11 |
| Xu Yong | 2 | 2119 | 73.51 |
| Gao Ying-Lian | 3 | 29 | 18.73 |
| Wang Dong | 4 | 9 | 2.23 |
| Chun-hou Zheng | 5 | 732 | 71.79 |
| Junliang Shang | 6 | 42 | 14.78 |