Name
Title | ||
|---|---|---|
An efficient nonnegative matrix factorization model for finding cancer associated genes by integrating data from genome, transcriptome and interactome |
Abstract | ||
|---|---|---|
The recent development of sequencing technology has offered us the opportunity of regarding cancer genome as a biological system and finding cancer associated genes computationally. Since some cancer associated genes may be omitted when using only mutation frequencies of genes, many recent studies use both mutation data from genome and gene interaction network from interactome to detect cancer associated genes. However, transcriptome information is not exploited in the task of finding cancer associated genes, which is also highly related to cancer. In this article, we introduce an nonnegative matrix factorization based model to find cancer associated genes, which can efficiently integrate information from genome, transcriptome and interactome via graph Laplacian regularization. When we compare our method with two existing methods and apply these methods on three independent cancer datasets, our method outperforms the existing methods for the evaluation of two well-curated known cancer genes. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/CISS.2018.8362246 | 2018 52nd Annual Conference on Information Sciences and Systems (CISS) |
Keywords | Field | DocType |
biological system,data integration,nonnegative matrix factorization,Laplacian regularization | Genome,Mathematical optimization,Interactome,Gene,Computer science,Transcriptome,Interaction network,Genomics,Non-negative matrix factorization,Computational biology,Cancer | Conference |
ISBN | Citations | PageRank |
978-1-5386-0580-6 | 0 | 0.34 |
References | Authors | |
0 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jianing Xi | 1 | 4 | 0.79 |
| Ao Li | 2 | 60 | 7.89 |
| Minghui Wang | 3 | 54 | 8.18 |