Limits...
Gene network inference by fusing data from diverse distributions.

Žitnik M, Zupan B - Bioinformatics (2015)

Bottom Line: In a simulation study, we demonstrate good predictive performance of FuseNet in comparison to several popular graphical models.Fusion of datasets offers substantial gains relative to inference of separate networks for each dataset.Our results demonstrate that network inference methods for non-Gaussian data can help in accurate modeling of the data generated by emergent high-throughput technologies.

View Article: PubMed Central - PubMed

Affiliation: Faculty of Computer and Information Science, University of Ljubljana, Ljubljana, Slovenia and Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA.

Show MeSH

Related in: MedlinePlus

Application of gene network inference algorithms to multinomial-distributed simulated data. Simulation studies on four network types were performed: random (see Supplementary Fig. S2), hub, scale-free and small world. For each graph type, we generated n = 300 observations at a high signal-to-noise ratio (SNR) with P = 50 variables (nodes) taking values from an alphabet of size m = 3. Boxplots are shown for multinomial FuseNet (proposed here) and the multinomial graphical model (Mult-GM) (Jalali et al., 2011)
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4542780&req=5

btv258-F2: Application of gene network inference algorithms to multinomial-distributed simulated data. Simulation studies on four network types were performed: random (see Supplementary Fig. S2), hub, scale-free and small world. For each graph type, we generated n = 300 observations at a high signal-to-noise ratio (SNR) with P = 50 variables (nodes) taking values from an alphabet of size m = 3. Boxplots are shown for multinomial FuseNet (proposed here) and the multinomial graphical model (Mult-GM) (Jalali et al., 2011)

Mentions: Experimental evidence indicates that FuseNet outperforms Gaussian-based competitors (GLASSO, Log-GLASSO and NPN-Copula) as well as existing methods that are designed specifically for the Poisson and the multinomial data (LPGM in Fig. 2 and Mult-GM in Fig. 3). The overall good performance of FuseNet is consistent across the four types of network structure and the two data distributions that we considered in experiments.Fig. 2.


Gene network inference by fusing data from diverse distributions.

Žitnik M, Zupan B - Bioinformatics (2015)

Application of gene network inference algorithms to multinomial-distributed simulated data. Simulation studies on four network types were performed: random (see Supplementary Fig. S2), hub, scale-free and small world. For each graph type, we generated n = 300 observations at a high signal-to-noise ratio (SNR) with P = 50 variables (nodes) taking values from an alphabet of size m = 3. Boxplots are shown for multinomial FuseNet (proposed here) and the multinomial graphical model (Mult-GM) (Jalali et al., 2011)
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4542780&req=5

btv258-F2: Application of gene network inference algorithms to multinomial-distributed simulated data. Simulation studies on four network types were performed: random (see Supplementary Fig. S2), hub, scale-free and small world. For each graph type, we generated n = 300 observations at a high signal-to-noise ratio (SNR) with P = 50 variables (nodes) taking values from an alphabet of size m = 3. Boxplots are shown for multinomial FuseNet (proposed here) and the multinomial graphical model (Mult-GM) (Jalali et al., 2011)
Mentions: Experimental evidence indicates that FuseNet outperforms Gaussian-based competitors (GLASSO, Log-GLASSO and NPN-Copula) as well as existing methods that are designed specifically for the Poisson and the multinomial data (LPGM in Fig. 2 and Mult-GM in Fig. 3). The overall good performance of FuseNet is consistent across the four types of network structure and the two data distributions that we considered in experiments.Fig. 2.

Bottom Line: In a simulation study, we demonstrate good predictive performance of FuseNet in comparison to several popular graphical models.Fusion of datasets offers substantial gains relative to inference of separate networks for each dataset.Our results demonstrate that network inference methods for non-Gaussian data can help in accurate modeling of the data generated by emergent high-throughput technologies.

View Article: PubMed Central - PubMed

Affiliation: Faculty of Computer and Information Science, University of Ljubljana, Ljubljana, Slovenia and Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA.

Show MeSH
Related in: MedlinePlus