Limits...
Directed mammalian gene regulatory networks using expression and comparative genomic hybridization microarray data from radiation hybrids.

Ahn S, Wang RT, Park CC, Lin A, Leahy RM, Lange K, Smith DJ - PLoS Comput. Biol. (2009)

Bottom Line: Upregulated edges in the RH network had significantly more overlap than downregulated.Consistent with their regulatory role, transcription factors had significantly more outgoing edges (regulating) than incoming (regulated) in the RH network, a feature hidden by conventional undirected networks.Directed RH genetic networks thus showed concordance with pre-existing networks while also yielding information inaccessible to current undirected approaches.

View Article: PubMed Central - PubMed

Affiliation: Signal and Image Processing Institute, University of Southern California, Los Angeles, California, USA.

ABSTRACT
Meiotic mapping of quantitative trait loci regulating expression (eQTLs) has allowed the construction of gene networks. However, the limited mapping resolution of these studies has meant that genotype data are largely ignored, leading to undirected networks that fail to capture regulatory hierarchies. Here we use high resolution mapping of copy number eQTLs (ceQTLs) in a mouse-hamster radiation hybrid (RH) panel to construct directed genetic networks in the mammalian cell. The RH network covering 20,145 mouse genes had significant overlap with, and similar topological structures to, existing biological networks. Upregulated edges in the RH network had significantly more overlap than downregulated. This suggests repressive relationships between genes are missed by existing approaches, perhaps because the corresponding proteins are not present in the cell at the same time and therefore unlikely to interact. Gene essentiality was positively correlated with connectivity and betweenness centrality in the RH network, strengthening the centrality-lethality principle in mammals. Consistent with their regulatory role, transcription factors had significantly more outgoing edges (regulating) than incoming (regulated) in the RH network, a feature hidden by conventional undirected networks. Directed RH genetic networks thus showed concordance with pre-existing networks while also yielding information inaccessible to current undirected approaches.

Show MeSH
Comparison of RH networks and existing datasets.(A) Overlap between 10 randomly permuted RH networks and HPRD network. The RH networks were constructed from right-tailed thresholding and one-sided Fisher's exact and chi-square tests used to assess significance. (B) Averaged  values for overlap between randomly permuted RH networks and different existing datasets (HPRD, KEGG, SymAtlas coexpression, GO, GO-molecular function, GO-cellular component and GO-biological process annotation networks). (C) Overlap between RH networks constructed from a subset of randomly selected RH clones and HPRD network. Mean of overlap significance (solid line) over 50 random subsets shown with standard errors calculated by bootstrapping (dash-dot line). (D) Same as (C) except averaged  values over different existing datasets. (E) Comparing different thresholding approaches. Maximum  over varying correlation coefficient thresholds shown. (F) Comparing betweenness centralities of RH and HPRD networks. P-values of Spearman correlation coefficients (one-sided, positive direction) between the betweenness centralities of RH and HPRD networks shown.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2690838&req=5

pcbi-1000407-g002: Comparison of RH networks and existing datasets.(A) Overlap between 10 randomly permuted RH networks and HPRD network. The RH networks were constructed from right-tailed thresholding and one-sided Fisher's exact and chi-square tests used to assess significance. (B) Averaged values for overlap between randomly permuted RH networks and different existing datasets (HPRD, KEGG, SymAtlas coexpression, GO, GO-molecular function, GO-cellular component and GO-biological process annotation networks). (C) Overlap between RH networks constructed from a subset of randomly selected RH clones and HPRD network. Mean of overlap significance (solid line) over 50 random subsets shown with standard errors calculated by bootstrapping (dash-dot line). (D) Same as (C) except averaged values over different existing datasets. (E) Comparing different thresholding approaches. Maximum over varying correlation coefficient thresholds shown. (F) Comparing betweenness centralities of RH and HPRD networks. P-values of Spearman correlation coefficients (one-sided, positive direction) between the betweenness centralities of RH and HPRD networks shown.

Mentions: The maximum overlap significance occurred at low correlation coefficient thresholds between 0 and 0.2 (Figure 1). To test whether this is simply because large thresholds (>0.2) yield too few edges and small thresholds (<0) give too many edges for significant overlap, we randomly permuted the elements of the adjacency matrix for the RH network and repeated the one-sided Fisher's exact and chi-square tests. The permuted network had the same size (number of edges) as the non-permuted RH network. As shown in Figures 2A (overlap with HPRD network) and 2B (overlap significance averaged over existing networks), the permuted networks did not show any significant overlap with the existing datasets (FDR-corrected ). These computational controls imply that the low correlation coefficient thresholds for maximum overlap significance are not simply a statistical artifact.


Directed mammalian gene regulatory networks using expression and comparative genomic hybridization microarray data from radiation hybrids.

Ahn S, Wang RT, Park CC, Lin A, Leahy RM, Lange K, Smith DJ - PLoS Comput. Biol. (2009)

Comparison of RH networks and existing datasets.(A) Overlap between 10 randomly permuted RH networks and HPRD network. The RH networks were constructed from right-tailed thresholding and one-sided Fisher's exact and chi-square tests used to assess significance. (B) Averaged  values for overlap between randomly permuted RH networks and different existing datasets (HPRD, KEGG, SymAtlas coexpression, GO, GO-molecular function, GO-cellular component and GO-biological process annotation networks). (C) Overlap between RH networks constructed from a subset of randomly selected RH clones and HPRD network. Mean of overlap significance (solid line) over 50 random subsets shown with standard errors calculated by bootstrapping (dash-dot line). (D) Same as (C) except averaged  values over different existing datasets. (E) Comparing different thresholding approaches. Maximum  over varying correlation coefficient thresholds shown. (F) Comparing betweenness centralities of RH and HPRD networks. P-values of Spearman correlation coefficients (one-sided, positive direction) between the betweenness centralities of RH and HPRD networks shown.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2690838&req=5

pcbi-1000407-g002: Comparison of RH networks and existing datasets.(A) Overlap between 10 randomly permuted RH networks and HPRD network. The RH networks were constructed from right-tailed thresholding and one-sided Fisher's exact and chi-square tests used to assess significance. (B) Averaged values for overlap between randomly permuted RH networks and different existing datasets (HPRD, KEGG, SymAtlas coexpression, GO, GO-molecular function, GO-cellular component and GO-biological process annotation networks). (C) Overlap between RH networks constructed from a subset of randomly selected RH clones and HPRD network. Mean of overlap significance (solid line) over 50 random subsets shown with standard errors calculated by bootstrapping (dash-dot line). (D) Same as (C) except averaged values over different existing datasets. (E) Comparing different thresholding approaches. Maximum over varying correlation coefficient thresholds shown. (F) Comparing betweenness centralities of RH and HPRD networks. P-values of Spearman correlation coefficients (one-sided, positive direction) between the betweenness centralities of RH and HPRD networks shown.
Mentions: The maximum overlap significance occurred at low correlation coefficient thresholds between 0 and 0.2 (Figure 1). To test whether this is simply because large thresholds (>0.2) yield too few edges and small thresholds (<0) give too many edges for significant overlap, we randomly permuted the elements of the adjacency matrix for the RH network and repeated the one-sided Fisher's exact and chi-square tests. The permuted network had the same size (number of edges) as the non-permuted RH network. As shown in Figures 2A (overlap with HPRD network) and 2B (overlap significance averaged over existing networks), the permuted networks did not show any significant overlap with the existing datasets (FDR-corrected ). These computational controls imply that the low correlation coefficient thresholds for maximum overlap significance are not simply a statistical artifact.

Bottom Line: Upregulated edges in the RH network had significantly more overlap than downregulated.Consistent with their regulatory role, transcription factors had significantly more outgoing edges (regulating) than incoming (regulated) in the RH network, a feature hidden by conventional undirected networks.Directed RH genetic networks thus showed concordance with pre-existing networks while also yielding information inaccessible to current undirected approaches.

View Article: PubMed Central - PubMed

Affiliation: Signal and Image Processing Institute, University of Southern California, Los Angeles, California, USA.

ABSTRACT
Meiotic mapping of quantitative trait loci regulating expression (eQTLs) has allowed the construction of gene networks. However, the limited mapping resolution of these studies has meant that genotype data are largely ignored, leading to undirected networks that fail to capture regulatory hierarchies. Here we use high resolution mapping of copy number eQTLs (ceQTLs) in a mouse-hamster radiation hybrid (RH) panel to construct directed genetic networks in the mammalian cell. The RH network covering 20,145 mouse genes had significant overlap with, and similar topological structures to, existing biological networks. Upregulated edges in the RH network had significantly more overlap than downregulated. This suggests repressive relationships between genes are missed by existing approaches, perhaps because the corresponding proteins are not present in the cell at the same time and therefore unlikely to interact. Gene essentiality was positively correlated with connectivity and betweenness centrality in the RH network, strengthening the centrality-lethality principle in mammals. Consistent with their regulatory role, transcription factors had significantly more outgoing edges (regulating) than incoming (regulated) in the RH network, a feature hidden by conventional undirected networks. Directed RH genetic networks thus showed concordance with pre-existing networks while also yielding information inaccessible to current undirected approaches.

Show MeSH