Integrating Genetic and Network Analysis to Characterize Genes Related to Mouse Weight

Ghazalpour, Anatole; Doss, Sudheer; Zhang, Bin; Wang, Susanna; Plaisier, Christopher L.; Castellanos, Ruth; Brozell, Alec; Schadt, Eric E.; Drake, Thomas A.; Lusis, Aldons J.; Horvath, Steve

doi:10.1371/journal.pgen.0020130.eor

Cited by 61 publications

(101 citation statements)

References 41 publications

Supporting

Mentioning

100

Contrasting

Order By: Relevance

“…The method uses the correlation patterns among genes across microarray samples as opposed to the association between each probe set and outcome [27,28]. WGCNA has been previously used to analyse gene expression data for a range of applications including mouse genetics [28][29][30]. In this study, we used WGCNA as an unsupervised, hypothesis-free method to uncover gene modules correlated with aggressive behaviour across all three genetic models of aggression.…”

Section: Introductionmentioning

confidence: 99%

Genes and Gene Networks Implicated in Aggression Related Behaviour

et al. 2014

View full text Add to dashboard Cite

Aggressive behaviour is a major cause of mortality and morbidity. Despite of moderate heritability estimates, progress in identifying the genetic factors underlying aggressive behaviour has been limited. There are currently three genetic mouse models of high and low aggression created using selective breeding. This is the first study to offer a global transcriptomic characterization of the prefrontal cortex across all three genetic mouse models of aggression. A systems biology approach has been applied to transcriptomic data across the three pairs of selected inbred mouse strains (Turku Aggressive (TA) and Turku Non-Aggressive (TNA), Short Attack Latency (SAL) and Long Attack Latency (LAL) mice and North Carolina Aggressive (NC900) and North Carolina Non-Aggressive (NC100)), providing novel insight into the neurobiological mechanisms and genetics underlying aggression. First, weighted gene co-expression network analysis (WGCNA) was performed to identify modules of highly correlated genes associated with aggression. Probe sets belonging to gene modules uncovered by WGCNA were carried forward for network analysis using ingenuity pathway analysis (IPA). The RankProd non-parametric algorithm was then used to statistically evaluate expression differences across the genes belonging to modules significantly associated with aggression. IPA uncovered two pathways, involving NF-kB and MAPKs. The secondary RankProd analysis yielded 14 differentially expressed genes, some of which have previously been implicated in pathways associated with aggressive behaviour, such as Adrbk2. The results highlighted plausible candidate genes and gene networks implicated in aggression-related behaviour.

show abstract

Section: Introductionmentioning

confidence: 99%

Genes and Gene Networks Implicated in Aggression Related Behaviour

et al. 2014

View full text Add to dashboard Cite

show abstract

“…For example, an edge between two genes may indicate that the corresponding expression traits are correlated in a given population of interest , that the corresponding proteins interact (Kim et al 2005), or that changes in the activity of one gene lead to changes in the activity of the other gene . Interaction or association networks have recently gained more widespread use in the biological community, where networks are formed by considering only pair-wise relationships between genes, including protein interaction relationships (Han et al, 2004), coexpression relationships (Gargalovic et al, 2006;Ghazalpour et al, 2006), as well as other straightforward measures that may indicate association between two genes. Forming association networks from expression data based purely on correlations between genes in a set of experiments of interest can give rise to links in the network driven by correlated noise structures between array-based experiments or other such artifacts.…”

Section: Using Eqtl Data To Reconstruct Coexpression Networkmentioning

confidence: 99%

“…The p-value threshold for considering two genes linked in the coexpression networks is chosen such that the resulting network exhibits the scale-free property (Barabasi and Albert, 1999;Ghazalpour et al, 2006;Lum et al, 2006) and the FDR for the gene-gene Scale-free fitting Adipose_female Figure 9.3 Variation in key parameters over different p values provides an objective way to select p-value thresholds for reconstructing coexpression networks. The first box of this figure plots the percent of gene pairs connected in the network that share GO biological process category terms, as a function of − log of the p-value threshold for the correlation between the gene pairs used to construct the network.…”

Section: Using Eqtl Data To Reconstruct Coexpression Networkmentioning

confidence: 99%

“…Given the scale-free and hierarchical nature of coexpression networks (Barabasi and Oltvai, 2004;Ghazalpour et al, 2006;Lum et al, 2006), one of the problems is to identify the key network modules in the network, representing those hub nodes (nodes that are significantly correlated with many other nodes) that are highly interconnected with one another, but that are not as highly connected with other hub nodes. Plate 9(a) illustrates a topological connectivity map for the most highly connected genes in the adipose tissue of the BXH cross (Chen et al, 2007).…”

Section: Identifying Modules Of Highly Interconnected Genes In Coexprmentioning

confidence: 99%

“…It has been demonstrated that the types of modules depicted in Plate 9(a) are enriched for known biological pathways, enriched for genes that are associated with disease traits, and enriched for genes that are linked to common genetic loci (Ghazalpour et al, 2006;Lum et al, 2006).…”

Section: Identifying Modules Of Highly Interconnected Genes In Coexprmentioning

confidence: 99%

See 2 more Smart Citations

Inferring Causal Associations between Genes and Disease via the Mapping of Expression Quantitative Trait Loci

Sieberts

Schadt

2007

Handbook of Statistical Genetics

View full text Add to dashboard Cite

The ability to monitor transcript levels in a comprehensive fashion allows for a more general characterization of transcriptional networks and their relationship to disease and other complex physiological processes. Information on how variations in DNA impact complex physiological processes flows through the transcriptional networks. Therefore, integrating DNA variation, transcription, and phenotypic data has the potential to enhance identification of the associations between DNA variation and disease, as well as characterize those parts of the molecular networks that drive disease. Toward that end, we discuss mapping expression quantitative trait loci (eQTL) for gene expression traits and then detail a method for integrating eQTL and expression and clinical data to infer causal relationships among gene expression traits and between expression and clinical traits. We further describe methods to integrate these data in a more comprehensive manner by constructing coexpression gene networks, which leverage pair-wise gene interaction data to represent more general relationships. To infer gene networks that capture causal information, we describe a Bayesian algorithm that further integrates eQTL and expression and clinical phenotype data to reconstruct whole gene networks capable of representing causal relationships among genes and traits in the network. These emerging high-dimensional data analysis approaches, that integrate large-scale data from multiple sources, represent the first steps in statistical genetics moving away from considering one trait at a time and toward operating in a network context. Evolving statistical procedures that operate on networks will be critical to extracting information related to complex phenotypes like disease, as research goes beyond the single-gene focus. The early successes achieved with some of the methods described herein suggest that these more integrative genomics approaches to dissecting disease traits will significantly enhance the identification of key drivers of disease beyond what could be achieved by genetics alone. 296Handbook of Statistical Genetics, Third Edition . E dited by D . J. Balding, M . Bishop and C. Cannings.

show abstract

Systems biology analysis of sjögren's syndrome and mucosa‐associated lymphoid tissue lymphoma in parotid glands

Hu¹,

Zhou²,

Jiang³

et al. 2008

Arthritis & Rheumatism

Self Cite

View full text Add to dashboard Cite

Objective To identify key target genes and activated signaling pathways associated with the pathogenesis of Sjögren's syndrome (SS) by conducting a systems analysis of parotid glands manifesting primary SS or primary SS/mucosa‐associated lymphoid tissue (MALT) lymphoma phenotypes. Methods A systems biology approach was used to analyze parotid gland tissue samples obtained from patients with primary SS, patients with primary SS/MALT lymphoma, and subjects without primary SS (non–primary SS controls). The tissue samples were assessed concurrently by gene‐expression microarray profiling and proteomics analysis, followed by weighted gene‐coexpression network analysis. Results Gene‐coexpression modules related to primary SS and primary SS/MALT lymphoma were significantly enriched with genes known to be involved in the immune/defense response, apoptosis, cell signaling, gene regulation, and oxidative stress. Detailed functional pathway analyses indicated that primary SS–associated modules were enriched with genes involved in proteasome degradation, apoptosis, signal peptides of the class I major histocompatibility complex (MHC), complement activation, cell growth and death, and integrin‐mediated cell adhesion, while primary SS/MALT lymphoma–associated modules were enriched with genes involved in translation, ribosome biogenesis and assembly, proteasome degradation, class I MHC signal peptides, the G13 signaling pathway, complement activation, and integrin‐mediated cell adhesion. Combined analyses of gene expression and proteomics data implicated 6 highly connected “hub” genes for distinguishing primary SS from non–primary SS, and 8 hub genes for distinguishing primary SS/MALT lymphoma from primary SS. Conclusion Systems biology analyses of the parotid glands from patients with primary SS and those with primary SS/MALT lymphoma revealed pathways and molecular targets associated with disease pathogenesis. The identified gene modules/pathways provide further insights into the molecular mechanisms of primary SS and primary SS/MALT lymphoma. The identified disease‐hub genes represent promising targets for therapeutic intervention, diagnosis, and prognosis.

show abstract

Integrating Genetic and Network Analysis to Characterize Genes Related to Mouse Weight

Cited by 61 publications

References 41 publications

Genes and Gene Networks Implicated in Aggression Related Behaviour

Genes and Gene Networks Implicated in Aggression Related Behaviour

Inferring Causal Associations between Genes and Disease via the Mapping of Expression Quantitative Trait Loci

Systems biology analysis of sjögren's syndrome and mucosa‐associated lymphoid tissue lymphoma in parotid glands

Contact Info

Product

Resources

About