Abstract
We present a novel algorithm to estimate genome-wide gene networks consisting of more than 20,000 genes from gene expression data using nonparametric Bayesian networks. Due to the difficulty of learning Bayesian network structures, existing algorithms cannot be applied to more than a few thousand genes. Our algorithm overcomes this limitation by repeatedly estimating subnetworks in parallel for genes selected by neighbor node sampling. Through numerical simulation, we confirmed that our algorithm outperformed a heuristic algorithm in a shorter time. We applied our algorithm to microarray data from human umbilical vein endothelial cells (HUVECs) treated with siRNAs, to construct a human genome-wide gene network, which we compared to a small gene network estimated for the genes extracted using a traditional bioinformatics method. The results showed that our genome-wide gene network contains many features of the small network, as well as others that could not be captured during the small network estimation. The results also revealed master-regulator genes that are not in the small network but that control many of the genes in the small network. These analyses were impossible to realize without our proposed algorithm.
| Original language | English |
|---|---|
| Article number | 5551118 |
| Pages (from-to) | 683-697 |
| Number of pages | 15 |
| Journal | IEEE/ACM Transactions on Computational Biology and Bioinformatics |
| Volume | 8 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - Mar 25 2011 |
| Externally published | Yes |
Fingerprint
All Science Journal Classification (ASJC) codes
- Biotechnology
- Genetics
- Applied Mathematics
Cite this
Estimating genome-wide gene networks using nonparametric bayesian network models on massively parallel computers. / Tamada, Yoshinori; Imoto, Seiya; Araki, Hiromitsu; Nagasaki, Masao; Print, Cristin; Charnock-Jones, D. Stephen; Miyano, Satoru.
In: IEEE/ACM Transactions on Computational Biology and Bioinformatics, Vol. 8, No. 3, 5551118, 25.03.2011, p. 683-697.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Estimating genome-wide gene networks using nonparametric bayesian network models on massively parallel computers
AU - Tamada, Yoshinori
AU - Imoto, Seiya
AU - Araki, Hiromitsu
AU - Nagasaki, Masao
AU - Print, Cristin
AU - Charnock-Jones, D. Stephen
AU - Miyano, Satoru
PY - 2011/3/25
Y1 - 2011/3/25
N2 - We present a novel algorithm to estimate genome-wide gene networks consisting of more than 20,000 genes from gene expression data using nonparametric Bayesian networks. Due to the difficulty of learning Bayesian network structures, existing algorithms cannot be applied to more than a few thousand genes. Our algorithm overcomes this limitation by repeatedly estimating subnetworks in parallel for genes selected by neighbor node sampling. Through numerical simulation, we confirmed that our algorithm outperformed a heuristic algorithm in a shorter time. We applied our algorithm to microarray data from human umbilical vein endothelial cells (HUVECs) treated with siRNAs, to construct a human genome-wide gene network, which we compared to a small gene network estimated for the genes extracted using a traditional bioinformatics method. The results showed that our genome-wide gene network contains many features of the small network, as well as others that could not be captured during the small network estimation. The results also revealed master-regulator genes that are not in the small network but that control many of the genes in the small network. These analyses were impossible to realize without our proposed algorithm.
AB - We present a novel algorithm to estimate genome-wide gene networks consisting of more than 20,000 genes from gene expression data using nonparametric Bayesian networks. Due to the difficulty of learning Bayesian network structures, existing algorithms cannot be applied to more than a few thousand genes. Our algorithm overcomes this limitation by repeatedly estimating subnetworks in parallel for genes selected by neighbor node sampling. Through numerical simulation, we confirmed that our algorithm outperformed a heuristic algorithm in a shorter time. We applied our algorithm to microarray data from human umbilical vein endothelial cells (HUVECs) treated with siRNAs, to construct a human genome-wide gene network, which we compared to a small gene network estimated for the genes extracted using a traditional bioinformatics method. The results showed that our genome-wide gene network contains many features of the small network, as well as others that could not be captured during the small network estimation. The results also revealed master-regulator genes that are not in the small network but that control many of the genes in the small network. These analyses were impossible to realize without our proposed algorithm.
UR - http://www.scopus.com/inward/record.url?scp=79952856971&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79952856971&partnerID=8YFLogxK
U2 - 10.1109/TCBB.2010.68
DO - 10.1109/TCBB.2010.68
M3 - Article
C2 - 20714027
AN - SCOPUS:79952856971
VL - 8
SP - 683
EP - 697
JO - IEEE/ACM Transactions on Computational Biology and Bioinformatics
JF - IEEE/ACM Transactions on Computational Biology and Bioinformatics
SN - 1545-5963
IS - 3
M1 - 5551118
ER -