Limits...
Computing interaction probabilities in signaling networks.

Gabr H, Rivera-Mulia JC, Gilbert DM, Kahveci T - EURASIP J Bioinform Syst Biol (2015)

Bottom Line: The transcription levels define the signal reachability probability between membrane receptors and transcription factors.Our method computes the interaction probabilities that minimize the gap between the observed and the computed signal reachability probabilities.We evaluate our method on four signaling networks from the Kyoto Encyclopedia of Genes and Genomes (KEGG).

View Article: PubMed Central - PubMed

Affiliation: Department of Computer & Information Science & Engineering, University of Florida, Gainesville, Florida, USA.

ABSTRACT

Biological networks inherently have uncertain topologies. This arises from many factors. For instance, interactions between molecules may or may not take place under varying conditions. Genetic or epigenetic mutations may also alter biological processes like transcription or translation. This uncertainty is often modeled by associating each interaction with a probability value. Studying biological networks under this probabilistic model has already been shown to yield accurate and insightful analysis of interaction data. However, the problem of assigning accurate probability values to interactions remains unresolved. In this paper, we present a novel method for computing interaction probabilities in signaling networks based on transcription levels of genes. The transcription levels define the signal reachability probability between membrane receptors and transcription factors. Our method computes the interaction probabilities that minimize the gap between the observed and the computed signal reachability probabilities. We evaluate our method on four signaling networks from the Kyoto Encyclopedia of Genes and Genomes (KEGG). For each network, we compute its edge probabilities using the gene expression profiles for seven major leukemia subtypes. We use these values to analyze how the stress induced by different leukemia subtypes affects signaling interactions.

No MeSH data available.


Related in: MedlinePlus

Gene centrality values of the four KEGG networks in seven leukemia subtypes. a Apoptosis, b cell cycle, c complement and coagulation, and (d) chemokine. Rows represent leukemia subtypes, and columnsrepresent genes. Both rows and columns are hierarchically clustered
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4642599&req=5

Fig3: Gene centrality values of the four KEGG networks in seven leukemia subtypes. a Apoptosis, b cell cycle, c complement and coagulation, and (d) chemokine. Rows represent leukemia subtypes, and columnsrepresent genes. Both rows and columns are hierarchically clustered

Mentions: Next, we use the interaction probability values we computed in Section 3.2 to compute centrality values of the genes in each network. Briefly, we compute the centrality of a gene as its contribution to signal reachability probability between all pairs of genes (see Gabr et al. [14] for details). We compute centrality values for the genes in each of the four networks for each of the seven leukemia subtypes. For each network, we represent each leukemia subtype by a vector of the node centrality values computed for it. We then compute a hierarchical clustering of these vectors. Figure 3 presents the results.Fig. 3


Computing interaction probabilities in signaling networks.

Gabr H, Rivera-Mulia JC, Gilbert DM, Kahveci T - EURASIP J Bioinform Syst Biol (2015)

Gene centrality values of the four KEGG networks in seven leukemia subtypes. a Apoptosis, b cell cycle, c complement and coagulation, and (d) chemokine. Rows represent leukemia subtypes, and columnsrepresent genes. Both rows and columns are hierarchically clustered
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4642599&req=5

Fig3: Gene centrality values of the four KEGG networks in seven leukemia subtypes. a Apoptosis, b cell cycle, c complement and coagulation, and (d) chemokine. Rows represent leukemia subtypes, and columnsrepresent genes. Both rows and columns are hierarchically clustered
Mentions: Next, we use the interaction probability values we computed in Section 3.2 to compute centrality values of the genes in each network. Briefly, we compute the centrality of a gene as its contribution to signal reachability probability between all pairs of genes (see Gabr et al. [14] for details). We compute centrality values for the genes in each of the four networks for each of the seven leukemia subtypes. For each network, we represent each leukemia subtype by a vector of the node centrality values computed for it. We then compute a hierarchical clustering of these vectors. Figure 3 presents the results.Fig. 3

Bottom Line: The transcription levels define the signal reachability probability between membrane receptors and transcription factors.Our method computes the interaction probabilities that minimize the gap between the observed and the computed signal reachability probabilities.We evaluate our method on four signaling networks from the Kyoto Encyclopedia of Genes and Genomes (KEGG).

View Article: PubMed Central - PubMed

Affiliation: Department of Computer & Information Science & Engineering, University of Florida, Gainesville, Florida, USA.

ABSTRACT

Biological networks inherently have uncertain topologies. This arises from many factors. For instance, interactions between molecules may or may not take place under varying conditions. Genetic or epigenetic mutations may also alter biological processes like transcription or translation. This uncertainty is often modeled by associating each interaction with a probability value. Studying biological networks under this probabilistic model has already been shown to yield accurate and insightful analysis of interaction data. However, the problem of assigning accurate probability values to interactions remains unresolved. In this paper, we present a novel method for computing interaction probabilities in signaling networks based on transcription levels of genes. The transcription levels define the signal reachability probability between membrane receptors and transcription factors. Our method computes the interaction probabilities that minimize the gap between the observed and the computed signal reachability probabilities. We evaluate our method on four signaling networks from the Kyoto Encyclopedia of Genes and Genomes (KEGG). For each network, we compute its edge probabilities using the gene expression profiles for seven major leukemia subtypes. We use these values to analyze how the stress induced by different leukemia subtypes affects signaling interactions.

No MeSH data available.


Related in: MedlinePlus