WGCNA: an R package for weighted correlation network analysis

Distinct neurogenomic states in basal ganglia subregions relate differently to singing behavior in songbirds

Both avian and mammalian basal ganglia are involved in voluntary motor control. In birds, such movements include hopping, perching and flying. Two organizational features that distinguish the songbird basal ganglia are that striatal and pallidal neurons are intermingled, and that neurons dedicated to vocal-motor function are clustered together in a dense cell group known as area X that sits within the surrounding striato-pallidum. This specification allowed us to perform molecular profiling of two striato-pallidal subregions, comparing transcriptional patterns in tissue dedicated to vocal-motor function (area X) to those in tissue that contains similar cell types but supports non-vocal behaviors: the striato-pallidum ventral to area X (VSP), our focus here. Since any behavior is likely underpinned by the coordinated actions of many molecules, we constructed gene co-expression networks from microarray data to study large-scale transcriptional patterns in both subregions. Our goal was to investigate any relationship between VSP network structure and singing and identify gene co-expression groups, or modules, found in the VSP but not area X. We observed mild, but surprising, relationships between VSP modules and song spectral features, and found a group of four VSP modules that were highly specific to the region. These modules were unrelated to singing, but were composed of genes involved in many of the same biological processes as those we previously observed in area X-specific singing-related modules. The VSP-specific modules were also enriched for processes disrupted in Parkinson's and Huntington's Diseases. Our results suggest that the activation/inhibition of a single pathway is not sufficient to functionally specify area X versus the VSP and support the notion that molecular processes are not in and of themselves specialized for behavior. Instead, unique interactions between molecular pathways create functional specificity in particular brain regions during distinct behavioral states.

Understanding how gene transcription relates to behavior is challenging. Learned vocal-motor behavior is a complex trait that represents the output of multiple converging genes, pathways, and patterns of neural activity. Here, we applied a systems analytical approach to determine how thousands of genes change their expression levels simultaneously in a region of the vertebrate brain important for vocal-motor function, the basal ganglia, during a specific vocal-motor behavior, singing. We used the zebra finch species of songbird based on similarities between song learning/production and speech, and because they possess a set of brain subregions dedicated to singing. Microarrays were used to measure gene expression levels in one such song-dedicated region and in an adjacent motor area that is not thought to play a role in vocal function. This allowed us to address the question of whether distinct gene co-expression patterns could be found in each area. We found that each area contained unique patterns of transcriptional co-activity, but there were also unexpected overlaps. We conclude that the particular behaviors (singing versus non-vocal behaviors) supported by these subregions depend on the particular sets of interactions between molecular pathways that occur in each subregion.

Maize source leaf adaptation to nitrogen deficiency affects not only nitrogen and carbon metabolism but also control of phosphate homeostasis

Crop plant development is strongly dependent on the availability of nitrogen (N) in the soil and the efficiency of N utilization for biomass production and yield. However, knowledge about molecular responses to N deprivation derives mainly from the study of model species. In this article, the metabolic adaptation of source leaves to low N was analyzed in maize (Zea mays) seedlings by parallel measurements of transcriptome and metabolome profiling. Inbred lines A188 and B73 were cultivated under sufficient (15 mM) or limiting (0.15 mM) nitrate supply for up to 30 d. Limited availability of N caused strong shifts in the metabolite profile of leaves. The transcriptome was less affected by the N stress but showed strong genotype- and age-dependent patterns. N starvation initiated the selective down-regulation of processes involved in nitrate reduction and amino acid assimilation; ammonium assimilation-related transcripts, on the other hand, were not influenced. Carbon assimilation-related transcripts were characterized by high transcriptional coordination and general down-regulation under low-N conditions. N deprivation caused a slight accumulation of starch but also directed increased amounts of carbohydrates into the cell wall and secondary metabolites. The decrease in N availability also resulted in accumulation of phosphate and strong down-regulation of genes usually involved in phosphate starvation response, underlining the great importance of phosphate homeostasis control under stress conditions.

Systems analysis of primary Sjögren's syndrome pathogenesis in salivary glands identifies shared pathways in human and a mouse model

Introduction

Primary Sjögren's syndrome (pSS) is a chronic autoimmune disease with complex etiopathogenesis. Despite extensive studies to understand the disease process utilizing human and mouse models, the intersection between these species remains elusive. To address this gap, we utilized a novel systems biology approach to identify disease-related gene modules and signaling pathways that overlap between humans and mice.

Methods

Parotid gland tissues were harvested from 24 pSS and 16 non-pSS sicca patients and 25 controls. For mouse studies, salivary glands were harvested from C57BL/6.NOD-Aec1Aec2 mice at various times during development of pSS-like disease. RNA was analyzed with Affymetrix HG U133+2.0 arrays for human samples and with MOE430+2.0 arrays for mouse samples. The images were processed with Affymetrix software. Weighted-gene co-expression network analysis was used to identify disease-related and functional pathways.

Results

Nineteen co-expression modules were identified in human parotid tissue, of which four were significantly upregulated and three were downregulated in pSS patients compared with non-pSS sicca patients and controls. Notably, one of the human disease-related modules was highly preserved in the mouse model, and was enriched with genes involved in immune and inflammatory responses. Further comparison between these two species led to the identification of genes associated with leukocyte recruitment and germinal center formation.

Conclusion

Our systems biology analysis of genome-wide expression data from salivary gland tissue of pSS patients and from a pSS mouse model identified common dysregulated biological pathways and molecular targets underlying critical molecular alterations in pSS pathogenesis.

The metabolic profile of long-lived Drosophila melanogaster

We investigated the age-related changes in the metabolic profile of male Drosophila melanogaster and compared the metabolic profile of flies selected for increased longevity to that of control flies of equal age. We found clear differences in metabolite composition between selection regimes and among age groups. Contrary to results found in a previous study of the transcriptome of these lines the metabolic profile did not show a younger pattern in longevity-selected (LS) flies than in same aged control (C) flies. Rather, many of the metabolites affected by age had levels common to older control individuals in the young LS flies. Furthermore, ageing affected the metabolome in a different LS specific direction. The selection induced difference increased with age. Some metabolites involved in oxidative phosphorylation changed with age highlighting the importance of mitochondrial function in the ageing process. However, these metabolites were not affected by selection for increased longevity, indicating that improvements of mitochondrial function were not involved in the increased lifespan of LS lines. Of the eight metabolites identified as having a significant difference in relative abundance between selection regimes in our study choline, lysine and glucose also show difference among lifespan phenotypes in C. elegans indicating that the correlation between the concentration of these metabolites and longevity was evolutionary conserved. Links between longevity and choline concentration is also found in mice making this metabolite an obvious target for further study.

A novel regulatory factor recruits the nucleosome remodeling complex to wingless integrated (Wnt) signaling gene promoters in mouse embryonic stem cells

The nucleosome remodeling and deacetylation (NuRD) complex is required for modulating the transcription of essential pluripotency genes in ESC self-renewal. MBD3 is considered a key player in the formation of a functional NuRD complex. The recruitment of MBD3 to methylated promoters may require other prerequisite factors. We show that cyclin-dependent kinase 2-associated protein 1 (CDK2AP1), an essential gene for early embryonic development, plays a role in pluripotency of ESC by engaging MBD3 to the promoter region of Wnt signaling genes. The occupancy of MBD3 on several promoters of Wnt genes was significantly lost in the absence of CDK2AP1, resulting in hyperactivation of Wnt. We propose that the transcriptional modulation of the Wnt pathway mediated by NuRD requires the presence of essential auxiliary components such as CDK2AP1, which may aid the association of the complex with specific focal regions of the target promoters.

"Good enough solutions" and the genetics of complex diseases

In this Emerging Science Review, we discuss a systems genetics strategy, which we call gene module association study (GMAS), as a novel approach complementing genome-wide association studies (GWAS), to understand complex diseases by focusing on how genes work together in groups rather than singly. The first step is to characterize phenotypic differences among a genetically diverse population. The second step is to use gene expression microarray (or other high-throughput) data from the population to construct gene coexpression networks. Coexpression analysis typically groups 20 000 genes into 20 to 30 modules containing tens to hundreds of genes, whose aggregate behavior can be represented by the module's "eigengene." The third step is to correlate expression patterns with phenotype, as in GWAS, only applied to eigengenes instead of single nucleotide polymorphisms. The goal of the GMAS approach is to identify groups of coregulated genes that explain complex traits from a systems perspective. From an evolutionary standpoint, we hypothesize that variability in eigengene patterns reflects the "good enough solution" concept, that biological systems are sufficiently complex so that many possible combinations of the same elements (in this case eigengenes) can produce an equivalent output, that is, a "good enough solution" to accomplish normal biological functions. However, when faced with environmental stresses, some "good enough solutions" adapt better than others, explaining individual variability to disease and drug susceptibility. If validated, GMAS may imply that common polygenic diseases are related as much to group interactions between normal genes, as to multiple gene mutations.

Systematic characterization of hypothetical proteins in Synechocystis sp. PCC 6803 reveals proteins functionally relevant to stress responses

We described here a global detection and functional inference of hypothetical proteins involved in stress response in Synechocystis sp. PCC 6803. In the study, we first applied an iTRAQ-LC-MS/MS based quantitative proteomics to the Synechocystis cells grown under five stress conditions. The analysis detected a total of 807 hypothetical proteins with high confidence. Among them, 480 were differentially regulated. We then applied a Weighted Gene Co-expression Network Analysis approach to construct transcriptional networks for Synechocystis under nutrient limitation and osmotic stress conditions using transcriptome datasets. The analysis showed that 305 and 467 coding genes of hypothetical proteins were functionally relevant to nutrient limitation and osmotic stress, respectively. A comparison of responsive hypothetical proteins to all stress conditions allowed identification of 22 hypothetical proteins commonly responsive to all stresses, suggesting they may be part of the core stress responses in Synechocystis. Finally, functional inference of these core stress responsive proteins using both sequence similarity and non-similarity approaches was conducted. The study provided new insights into the stress response networks in Synechocystis, and also demonstrated that a combination of experimental "OMICS" and bioinformatics methodologies could improve functional annotation for hypothetical proteins.

Systems biology approaches to finding novel pain mediators

Chronic pain represents a major health burden; this maladaptive pain state occurs as a consequence of hypersensitivity within the peripheral and central components of the somatosensory system. High throughput technologies (genomics, transciptomics, lipidomics, and proteomics) are now being applied to tissue derived from pain patients as well as experimental pain models to discover novel pain mediators. The use of clustering, meta-analysis and other techniques can help refine potential candidates. Of particular importance are systems biology methods, such as co-expression network generating algorithms, which infer potential associations/interactions between molecules and build networks based on these interactions. Protein-protein interaction networks allow the lists of potential targets generated by these different platforms to be analyzed in their biological context. Outputs from these different methods must also be related to the clinical pain phenotype. The improved and standardized phenotyping of pain symptoms and sensory signs enables much better subject stratification. Our hope is that, in the future, the use of computational approaches to integrate datasets including sensory phenotype as well as the outputs of high throughput technologies will help define novel pain mediators and provide insights into the pathogenesis of chronic pain.

Impact of experience-dependent and -independent factors on gene expression in songbird brain

Songbirds provide rich natural models for studying the relationships between brain anatomy, behavior, environmental signals, and gene expression. Under the Songbird Neurogenomics Initiative, investigators from 11 laboratories collected brain samples from six species of songbird under a range of experimental conditions, and 488 of these samples were analyzed systematically for gene expression by microarray. ANOVA was used to test 32 planned contrasts in the data, revealing the relative impact of different factors. The brain region from which tissue was taken had the greatest influence on gene expression profile, affecting the majority of signals measured by 18,848 cDNA spots on the microarray. Social and environmental manipulations had a highly variable impact, interpreted here as a manifestation of paradoxical "constitutive plasticity" (fewer inducible genes) during periods of enhanced behavioral responsiveness. Several specific genes were identified that may be important in the evolution of linkages between environmental signals and behavior. The data were also analyzed using weighted gene coexpression network analysis, followed by gene ontology analysis. This revealed modules of coexpressed genes that are also enriched for specific functional annotations, such as "ribosome" (expressed more highly in juvenile brain) and "dopamine metabolic process" (expressed more highly in striatal song control nucleus area X). These results underscore the complexity of influences on neural gene expression and provide a resource for studying how these influences are integrated during natural experience.

Aging effects on DNA methylation modules in human brain and blood tissue

Background

Several recent studies reported aging effects on DNA methylation levels of individual CpG dinucleotides. But it is not yet known whether aging-related consensus modules, in the form of clusters of correlated CpG markers, can be found that are present in multiple human tissues. Such a module could facilitate the understanding of aging effects on multiple tissues.

Results

We therefore employed weighted correlation network analysis of 2,442 Illumina DNA methylation arrays from brain and blood tissues, which enabled the identification of an age-related co-methylation module. Module preservation analysis confirmed that this module can also be found in diverse independent data sets. Biological evaluation showed that module membership is associated with Polycomb group target occupancy counts, CpG island status and autosomal chromosome location. Functional enrichment analysis revealed that the aging-related consensus module comprises genes that are involved in nervous system development, neuron differentiation and neurogenesis, and that it contains promoter CpGs of genes known to be down-regulated in early Alzheimer's disease. A comparison with a standard, non-module based meta-analysis revealed that selecting CpGs based on module membership leads to significantly increased gene ontology enrichment, thus demonstrating that studying aging effects via consensus network analysis enhances the biological insights gained.

Conclusions

Overall, our analysis revealed a robustly defined age-related co-methylation module that is present in multiple human tissues, including blood and brain. We conclude that blood is a promising surrogate for brain tissue when studying the effects of age on DNA methylation profiles.

Activity-dependent protein dynamics define interconnected cores of co-regulated postsynaptic proteins

Synapses are highly dynamic structures that mediate cell-cell communication in the central nervous system. Their molecular composition is altered in an activity-dependent fashion, which modulates the efficacy of subsequent synaptic transmission events. Whereas activity-dependent trafficking of individual key synaptic proteins into and out of the synapse has been characterized previously, global activity-dependent changes in the synaptic proteome have not been studied. To test the feasibility of carrying out an unbiased large-scale approach, we investigated alterations in the molecular composition of synaptic spines following mass stimulation of the central nervous system induced by pilocarpine. We observed widespread changes in relative synaptic abundances encompassing essentially all proteins, supporting the view that the molecular composition of the postsynaptic density is tightly regulated. In most cases, we observed that members of gene families displayed coordinate regulation even when they were not known to physically interact. Analysis of correlated synaptic localization revealed a tightly co-regulated cluster of proteins, consisting of mainly glutamate receptors and their adaptors. This cluster constitutes a functional core of the postsynaptic machinery, and changes in its size affect synaptic strength and synaptic size. Our data show that the unbiased investigation of activity-dependent signaling of the postsynaptic density proteome can offer valuable new information on synaptic plasticity.

Dynamic landscape of tandem 3' UTRs during zebrafish development

Tandem 3' untranslated regions (UTRs), produced by alternative polyadenylation (APA) in the terminal exon of a gene, could have critical roles in regulating gene networks. Here we profiled tandem poly(A) events on a genome-wide scale during the embryonic development of zebrafish (Danio rerio) using a recently developed SAPAS method. We showed that 43% of the expressed protein-coding genes have tandem 3' UTRs. The average 3' UTR length follows a V-shaped dynamic pattern during early embryogenesis, in which the 3' UTRs are first shortened at zygotic genome activation, and then quickly lengthened during gastrulation. Over 4000 genes are found to switch tandem APA sites, and the distinct functional roles of these genes are indicated by Gene Ontology analysis. Three families of cis-elements, including miR-430 seed, U-rich element, and canonical poly(A) signal, are enriched in 3' UTR-shortened/lengthened genes in a stage-specific manner, suggesting temporal regulation coordinated by APA and trans-acting factors. Our results highlight the regulatory role of tandem 3' UTR control in early embryogenesis and suggest that APA may represent a new epigenetic paradigm of physiological regulations.

Comparative co-expression analysis in plant biology

The analysis of gene expression data generated by high-throughput microarray transcript profiling experiments has shown that transcriptionally coordinated genes are often functionally related. Based on large-scale expression compendia grouping multiple experiments, this guilt-by-association principle has been applied to study modular gene programmes, identify cis-regulatory elements or predict functions for unknown genes in different model plants. Recently, several studies have demonstrated how, through the integration of gene homology and expression information, correlated gene expression patterns can be compared between species. The incorporation of detailed functional annotations as well as experimental data describing protein-protein interactions, phenotypes or tissue specific expression, provides an invaluable source of information to identify conserved gene modules and translate biological knowledge from model organisms to crops. In this review, we describe the different steps required to systematically compare expression data across species. Apart from the technical challenges to compute and display expression networks from multiple species, some future applications of plant comparative transcriptomics are highlighted.

Using networks to understand medical data: the case of Class III malocclusions

A system of elements that interact or regulate each other can be represented by a mathematical object called a network. While network analysis has been successfully applied to high-throughput biological systems, less has been done regarding their application in more applied fields of medicine; here we show an application based on standard medical diagnostic data. We apply network analysis to Class III malocclusion, one of the most difficult to understand and treat orofacial anomaly. We hypothesize that different interactions of the skeletal components can contribute to pathological disequilibrium; in order to test this hypothesis, we apply network analysis to 532 Class III young female patients. The topology of the Class III malocclusion obtained by network analysis shows a strong co-occurrence of abnormal skeletal features. The pattern of these occurrences influences the vertical and horizontal balance of disharmony in skeletal form and position. Patients with more unbalanced orthodontic phenotypes show preponderance of the pathological skeletal nodes and minor relevance of adaptive dentoalveolar equilibrating nodes. Furthermore, by applying Power Graphs analysis we identify some functional modules among orthodontic nodes. These modules correspond to groups of tightly inter-related features and presumably constitute the key regulators of plasticity and the sites of unbalance of the growing dentofacial Class III system. The data of the present study show that, in their most basic abstraction level, the orofacial characteristics can be represented as graphs using nodes to represent orthodontic characteristics, and edges to represent their various types of interactions. The applications of this mathematical model could improve the interpretation of the quantitative, patient-specific information, and help to better targeting therapy. Last but not least, the methodology we have applied in analyzing orthodontic features can be applied easily to other fields of the medical science.

CTen: a web-based platform for identifying enriched cell types from heterogeneous microarray data

Background

Interpreting in vivo sampled microarray data is often complicated by changes in the cell population demographics. To put gene expression into its proper biological context, it is necessary to distinguish differential gene transcription from artificial gene expression induced by changes in the cellular demographics.

Results

CTen (cell type enrichment) is a web-based analytical tool which uses our highly expressed, cell specific (HECS) gene database to identify enriched cell types in heterogeneous microarray data. The web interface is designed for differential expression and gene clustering studies, and the enrichment results are presented as heatmaps or downloadable text files.

Conclusions

In this work, we use an independent, cell-specific gene expression data set to assess CTen's performance in accurately identifying the appropriate cell type and provide insight into the suggested level of enrichment to optimally minimize the number of false discoveries. We show that CTen, when applied to microarray data developed from infected lung tissue, can correctly identify the cell signatures of key lymphocytes in a highly heterogeneous environment and compare its performance to another popular bioinformatics tool. Furthermore, we discuss the strong implications cell type enrichment has in the design of effective microarray workflow strategies and show that, by combining CTen with gene expression clustering, we may be able to determine the relative changes in the number of key cell types.

CTen is available at http://www.influenza-x.org/~jshoemaker/cten/

Systems genetics: a novel approach to dissect the genetic basis of osteoporosis

From the early 1990s to the middle of the last decade, the search for genes influencing osteoporosis proved difficult with few successes. However, over the last 5 years this has begun to change with the introduction of genome-wide association (GWA) studies. In this short period of time, GWA studies have significantly accelerated the pace of gene discovery, leading to the identification of nearly 100 independent associations for osteoporosis-related traits. However, GWA does not specifically pinpoint causal genes or provide functional context for associations. Thus, there is a need for approaches that provide systems-level insight on how associated variants influence cellular function, downstream gene networks, and ultimately disease. In this review we discuss the emerging field of "systems genetics" and how it is being used in combination with and independent of GWA to improve our understanding of the molecular mechanisms involved in bone fragility.

Application of the Gini correlation coefficient to infer regulatory relationships in transcriptome analysis

One of the computational challenges in plant systems biology is to accurately infer transcriptional regulation relationships based on correlation analyses of gene expression patterns. Despite several correlation methods that are applied in biology to analyze microarray data, concerns regarding the compatibility of these methods with the gene expression data profiled by high-throughput RNA transcriptome sequencing (RNA-Seq) technology have been raised. These concerns are mainly due to the fact that the distribution of read counts in RNA-Seq experiments is different from that of fluorescence intensities in microarray experiments. Therefore, a comprehensive evaluation of the existing correlation methods and, if necessary, introduction of novel methods into biology is appropriate. In this study, we compared four existing correlation methods used in microarray analysis and one novel method called the Gini correlation coefficient on previously published microarray-based and sequencing-based gene expression data in Arabidopsis (Arabidopsis thaliana) and maize (Zea mays). The comparisons were performed on more than 11,000 regulatory relationships in Arabidopsis, including 8,929 pairs of transcription factors and target genes. Our analyses pinpointed the strengths and weaknesses of each method and indicated that the Gini correlation can compensate for the shortcomings of the Pearson correlation, the Spearman correlation, the Kendall correlation, and the Tukey's biweight correlation. The Gini correlation method, with the other four evaluated methods in this study, was implemented as an R package named rsgcc that can be utilized as an alternative option for biologists to perform clustering analyses of gene expression patterns or transcriptional network analyses.

Expression profiling based on coexpressed modules in obese prepubertal children

The aim of this study was to identify related genes and the underlying molecular mechanisms in obese patients who show a series of clinical and metabolic abnormalities known as metabolic syndrome. We identified expression profiles through a coexpression network. In addition, a similarity matrix and expression modules were constructed based on domain and pathway enrichment analysis. The genes in module 1 were mainly involved in the metabolism of xenobiotics by cytochrome P450, aldosterone-regulated sodium reabsorption, and focal adhesion owing to the presence of aldo/ketoreductase, basic helix-loop-helix, von Willebrand factor, Frizzled-related domain, and other domains. The genes in module 3 may be involved in cell cycle (hsa04110) and DNA replication (hsa03030) pathways through mini-chromosome maintenance, serine/threonine protein kinase, the protein kinase domain, and other domains. We analyzed the published molecular mechanisms of obesity and found many genes and pathways that have not been given enough attention and require further confirmation.

Weighted frequent gene co-expression network mining to identify genes involved in genome stability

Gene co-expression network analysis is an effective method for predicting gene functions and disease biomarkers. However, few studies have systematically identified co-expressed genes involved in the molecular origin and development of various types of tumors. In this study, we used a network mining algorithm to identify tightly connected gene co-expression networks that are frequently present in microarray datasets from 33 types of cancer which were derived from 16 organs/tissues. We compared the results with networks found in multiple normal tissue types and discovered 18 tightly connected frequent networks in cancers, with highly enriched functions on cancer-related activities. Most networks identified also formed physically interacting networks. In contrast, only 6 networks were found in normal tissues, which were highly enriched for housekeeping functions. The largest cancer network contained many genes with genome stability maintenance functions. We tested 13 selected genes from this network for their involvement in genome maintenance using two cell-based assays. Among them, 10 were shown to be involved in either homology-directed DNA repair or centrosome duplication control including the well-known cancer marker MKI67. Our results suggest that the commonly recognized characteristics of cancers are supported by highly coordinated transcriptomic activities. This study also demonstrated that the co-expression network directed approach provides a powerful tool for understanding cancer physiology, predicting new gene functions, as well as providing new target candidates for cancer therapeutics.

Transcriptional and metabolic analysis of senescence induced by preventing pollination in maize

Transcriptional and metabolic changes were evaluated during senescence induced by preventing pollination in the B73 genotype of maize (Zea mays). Accumulation of free glucose and starch and loss of chlorophyll in leaf was manifested early at 12 d after anthesis (DAA), while global transcriptional and phenotypic changes were evident only at 24 DAA. Internodes exhibited major transcriptomic changes only at 30 DAA. Overlaying expression data onto metabolic pathways revealed involvement of many novel pathways, including those involved in cell wall biosynthesis. To investigate the overlap between induced and natural senescence, transcriptional data from induced senescence in maize was compared with that reported for Arabidopsis (Arabidopsis thaliana) undergoing natural and sugar-induced senescence. Notable similarities with natural senescence in Arabidopsis included up-regulation of senescence-associated genes (SAGs), ethylene and jasmonic acid biosynthetic genes, APETALA2, ethylene-responsive element binding protein, and no apical meristem transcription factors. However, differences from natural senescence were highlighted by unaltered expression of a subset of the SAGs, and cytokinin, abscisic acid, and salicylic acid biosynthesis genes. Key genes up-regulated during sugar-induced senescence in Arabidopsis, including a cysteine protease (SAG12) and three flavonoid biosynthesis genes (PRODUCTION OF ANTHOCYANIN PIGMENT1 (PAP1), PAP2, and LEUCOANTHOCYANIDIN DIOXYGENASE), were also induced, suggesting similarities in senescence induced by pollination prevention and sugar application. Coexpression analysis revealed networks involving known senescence-related genes and novel candidates; 82 of these were shared between leaf and internode networks, highlighting similarities in induced senescence in these tissues. Insights from this study will be valuable in systems biology of senescence in maize and other grasses.

Beyond differential expression: the quest for causal mutations and effector molecules

High throughput gene expression technologies are a popular choice for researchers seeking molecular or systems-level explanations of biological phenomena. Nevertheless, there has been a groundswell of opinion that these approaches have not lived up to the hype because the interpretation of the data has lagged behind its generation. In our view a major problem has been an over-reliance on isolated lists of differentially expressed (DE) genes which – by simply comparing genes to themselves – have the pitfall of taking molecular information out of context. Numerous scientists have emphasised the need for better context. This can be achieved through holistic measurements of differential connectivity in addition to, or in replacement, of DE. However, many scientists continue to use isolated lists of DE genes as the major source of input data for common readily available analytical tools. Focussing this opinion article on our own research in skeletal muscle, we outline our resolutions to these problems – particularly a universally powerful way of quantifying differential connectivity. With a well designed experiment, it is now possible to use gene expression to identify causal mutations and the other major effector molecules with whom they cooperate, irrespective of whether they themselves are DE. We explain why, for various reasons, no other currently available experimental techniques or quantitative analyses are capable of reaching these conclusions.

Integrative prescreening in analysis of multiple cancer genomic studies

BACKGROUND

In high throughput cancer genomic studies, results from the analysis of single datasets often suffer from a lack of reproducibility because of small sample sizes. Integrative analysis can effectively pool and analyze multiple datasets and provides a cost effective way to improve reproducibility. In integrative analysis, simultaneously analyzing all genes profiled may incur high computational cost. A computationally affordable remedy is prescreening, which fits marginal models, can be conducted in a parallel manner, and has low computational cost.

RESULTS

An integrative prescreening approach is developed for the analysis of multiple cancer genomic datasets. Simulation shows that the proposed integrative prescreening has better performance than alternatives, particularly including prescreening with individual datasets, an intensity approach and meta-analysis. We also analyze multiple microarray gene profiling studies on liver and pancreatic cancers using the proposed approach.

CONCLUSIONS

The proposed integrative prescreening provides an effective way to reduce the dimensionality in cancer genomic studies. It can be coupled with existing analysis methods to identify cancer markers.

Constructing endophenotypes of complex diseases using non-negative matrix factorization and adjusted rand index

Complex diseases are typically caused by combinations of molecular disturbances that vary widely among different patients. Endophenotypes, a combination of genetic factors associated with a disease, offer a simplified approach to dissect complex trait by reducing genetic heterogeneity. Because molecular dissimilarities often exist between patients with indistinguishable disease symptoms, these unique molecular features may reflect pathogenic heterogeneity. To detect molecular dissimilarities among patients and reduce the complexity of high-dimension data, we have explored an endophenotype-identification analytical procedure that combines non-negative matrix factorization (NMF) and adjusted rand index (ARI), a measure of the similarity of two clusterings of a data set. To evaluate this procedure, we compared it with a commonly used method, principal component analysis with k-means clustering (PCA-K). A simulation study with gene expression dataset and genotype information was conducted to examine the performance of our procedure and PCA-K. The results showed that NMF mostly outperformed PCA-K. Additionally, we applied our endophenotype-identification analytical procedure to a publicly available dataset containing data derived from patients with late-onset Alzheimer's disease (LOAD). NMF distilled information associated with 1,116 transcripts into three metagenes and three molecular subtypes (MS) for patients in the LOAD dataset: MS1 (n1=80), MS2 (n2=73), and MS3 (n3=23). ARI was then used to determine the most representative transcripts for each metagene; 123, 89, and 71 metagene-specific transcripts were identified for MS1, MS2, and MS3, respectively. These metagene-specific transcripts were identified as the endophenotypes. Our results showed that 14, 38, 0, and 28 candidate susceptibility genes listed in AlzGene database were found by all patients, MS1, MS2, and MS3, respectively. Moreover, we found that MS2 might be a normal-like subtype. Our proposed procedure provides an alternative approach to investigate the pathogenic mechanism of disease and better understand the relationship between phenotype and genotype.

Network organization of the huntingtin proteomic interactome in mammalian brain

We used affinity-purification mass spectrometry to identify 747 candidate proteins that are complexed with Huntingtin (Htt) in distinct brain regions and ages in Huntington's disease (HD) and wild-type mouse brains. To gain a systems-level view of the Htt interactome, we applied Weighted Correlation Network Analysis to the entire proteomic data set to unveil a verifiable rank of Htt-correlated proteins and a network of Htt-interacting protein modules, with each module highlighting distinct aspects of Htt biology. Importantly, the Htt-containing module is highly enriched with proteins involved in 14-3-3 signaling, microtubule-based transport, and proteostasis. Top-ranked proteins in this module were validated as Htt interactors and genetic modifiers in an HD Drosophila model. Our study provides a compendium of spatiotemporal Htt-interacting proteins in the mammalian brain and presents an approach for analyzing proteomic interactome data sets to build in vivo protein networks in complex tissues, such as the brain.

Inferring gene functions through dissection of relevance networks: interleaving the intra- and inter-species views

Inference of accurate gene annotations requires integration of existing biological knowledge, structured in a form of ontology, with data from transcriptomics high-throughput technologies. This undertaking requires developing algorithms that integrate genome-scale data, even for model organisms. Gene relevance networks have emerged as a powerful representative of the structure of the data. Such networks can be used for intra-species transfer of gene annotations following the guilt-by-association principle. An analogous principle can serve as a basis for inter-species transfer of gene annotations by comparing well-defined subnetworks. In this review, we compare and contrast the concepts of relevance and proximity networks and briefly review the concept of semantic similarity. We then provide a detailed account of quantitative guilt-by-association inference in the setting of genome-scale relevance networks. Moreover, we systematically survey the existing network-based approaches for automated gene function annotation and categorize them under one umbrella in terms of employed methodology. Furthermore, we discuss suitable data selection strategies required for deriving meaningful and unbiased genome-scale networks from large transcriptomics compendia. Lastly, by simulating gene function prediction with a classical network-based algorithm, we show how the number of genes of unknown function influences prediction within a species and pinpoint the need and the requirements for inter-species knowledge transfer.

Mathematical and statistical modeling in cancer systems biology

Cancer is a major health problem with high mortality rates. In the post-genome era, investigators have access to massive amounts of rapidly accumulating high-throughput data in publicly available databases, some of which are exclusively devoted to housing Cancer data. However, data interpretation efforts have not kept pace with data collection, and gained knowledge is not necessarily translating into better diagnoses and treatments. A fundamental problem is to integrate and interpret data to further our understanding in Cancer Systems Biology. Viewing cancer as a network provides insights into the complex mechanisms underlying the disease. Mathematical and statistical models provide an avenue for cancer network modeling. In this article, we review two widely used modeling paradigms: deterministic metabolic models and statistical graphical models. The strength of these approaches lies in their flexibility and predictive power. Once a model has been validated, it can be used to make predictions and generate hypotheses. We describe a number of diverse applications to Cancer Biology, including, the system-wide effects of drug-treatments, disease prognosis, tumor classification, forecasting treatment outcomes, and survival predictions.

A gene co-expression network in whole blood of schizophrenia patients is independent of antipsychotic-use and enriched for brain-expressed genes

Despite large-scale genome-wide association studies (GWAS), the underlying genes for schizophrenia are largely unknown. Additional approaches are therefore required to identify the genetic background of this disorder. Here we report findings from a large gene expression study in peripheral blood of schizophrenia patients and controls. We applied a systems biology approach to genome-wide expression data from whole blood of 92 medicated and 29 antipsychotic-free schizophrenia patients and 118 healthy controls. We show that gene expression profiling in whole blood can identify twelve large gene co-expression modules associated with schizophrenia. Several of these disease related modules are likely to reflect expression changes due to antipsychotic medication. However, two of the disease modules could be replicated in an independent second data set involving antipsychotic-free patients and controls. One of these robustly defined disease modules is significantly enriched with brain-expressed genes and with genetic variants that were implicated in a GWAS study, which could imply a causal role in schizophrenia etiology. The most highly connected intramodular hub gene in this module (ABCF1), is located in, and regulated by the major histocompatibility (MHC) complex, which is intriguing in light of the fact that common allelic variants from the MHC region have been implicated in schizophrenia. This suggests that the MHC increases schizophrenia susceptibility via altered gene expression of regulatory genes in this network.

Functional annotation of the human brain methylome identifies tissue-specific epigenetic variation across brain and blood

Background

Dynamic changes to the epigenome play a critical role in establishing and maintaining cellular phenotype during differentiation, but little is known about the normal methylomic differences that occur between functionally distinct areas of the brain. We characterized intra- and inter-individual methylomic variation across whole blood and multiple regions of the brain from multiple donors.

Results

Distinct tissue-specific patterns of DNA methylation were identified, with a highly significant over-representation of tissue-specific differentially methylated regions (TS-DMRs) observed at intragenic CpG islands and low CG density promoters. A large proportion of TS-DMRs were located near genes that are differentially expressed across brain regions. TS-DMRs were significantly enriched near genes involved in functional pathways related to neurodevelopment and neuronal differentiation, including BDNF, BMP4, CACNA1A, CACA1AF, EOMES, NGFR, NUMBL, PCDH9, SLIT1, SLITRK1 and SHANK3. Although between-tissue variation in DNA methylation was found to greatly exceed between-individual differences within any one tissue, we found that some inter-individual variation was reflected across brain and blood, indicating that peripheral tissues may have some utility in epidemiological studies of complex neurobiological phenotypes.

Conclusions

This study reinforces the importance of DNA methylation in regulating cellular phenotype across tissues, and highlights genomic patterns of epigenetic variation across functionally distinct regions of the brain, providing a resource for the epigenetics and neuroscience research communities.

Network methods for describing sample relationships in genomic datasets: application to Huntington's disease

BACKGROUND

Genomic datasets generated by new technologies are increasingly prevalent in disparate areas of biological research. While many studies have sought to characterize relationships among genomic features, commensurate efforts to characterize relationships among biological samples have been less common. Consequently, the full extent of sample variation in genomic studies is often under-appreciated, complicating downstream analytical tasks such as gene co-expression network analysis.

RESULTS

Here we demonstrate the use of network methods for characterizing sample relationships in microarray data generated from human brain tissue. We describe an approach for identifying outlying samples that does not depend on the choice or use of clustering algorithms. We introduce a battery of measures for quantifying the consistency and integrity of sample relationships, which can be compared across disparate studies, technology platforms, and biological systems. Among these measures, we provide evidence that the correlation between the connectivity and the clustering coefficient (two important network concepts) is a sensitive indicator of homogeneity among biological samples. We also show that this measure, which we refer to as cor(K,C), can distinguish biologically meaningful relationships among subgroups of samples. Specifically, we find that cor(K,C) reveals the profound effect of Huntington's disease on samples from the caudate nucleus relative to other brain regions. Furthermore, we find that this effect is concentrated in specific modules of genes that are naturally co-expressed in human caudate nucleus, highlighting a new strategy for exploring the effects of disease on sets of genes.

CONCLUSIONS

These results underscore the importance of systematically exploring sample relationships in large genomic datasets before seeking to analyze genomic feature activity. We introduce a standardized platform for this purpose using freely available R software that has been designed to enable iterative and interactive exploration of sample networks.

Human gene correlation analysis (HGCA): a tool for the identification of transcriptionally co-expressed genes

Background

Bioinformatics and high-throughput technologies such as microarray studies allow the measure of the expression levels of large numbers of genes simultaneously, thus helping us to understand the molecular mechanisms of various biological processes in a cell.

Findings

We calculate the Pearson Correlation Coefficient (r-value) between probe set signal values from Affymetrix Human Genome Microarray samples and cluster the human genes according to the r-value correlation matrix using the Neighbour Joining (NJ) clustering method. A hyper-geometric distribution is applied on the text annotations of the probe sets to quantify the term overrepresentations. The aim of the tool is the identification of closely correlated genes for a given gene of interest and/or the prediction of its biological function, which is based on the annotations of the respective gene cluster.

Conclusion

Human Gene Correlation Analysis (HGCA) is a tool to classify human genes according to their coexpression levels and to identify overrepresented annotation terms in correlated gene groups. It is available at: http://biobank-informatics.bioacademy.gr/coexpression/.

mrhl RNA, a long noncoding RNA, negatively regulates Wnt signaling through its protein partner Ddx5/p68 in mouse spermatogonial cells

Meiotic recombination hot spot locus (mrhl) RNA is a nuclear enriched long noncoding RNA encoded in the mouse genome and expressed in testis, liver, spleen, and kidney. mrhl RNA silencing in Gc1-Spg cells, derived from mouse spermatogonial cells, resulted in perturbation of expression of genes belonging to cell adhesion, cell signaling and development, and differentiation, among which many were of the Wnt signaling pathway. A weighted gene coexpression network generated nine coexpression modules, which included TCF4, a key transcription factor involved in Wnt signaling. Activation of Wnt signaling upon mrhl RNA downregulation was demonstrated by beta-catenin nuclear localization, beta-catenin-TCF4 interaction, occupancy of beta-catenin at the promoters of Wnt target genes, and TOP/FOP-luciferase assay. Northwestern blot and RNA pulldown experiments identified Ddx5/p68 as one of the interacting proteins of mrhl RNA. Downregulation of mrhl RNA resulted in the cytoplasmic translocation of tyrosine-phosphorylated p68. Concomitant downregulation of both mrhl RNA and p68 prevented the nuclear translocation of beta-catenin. mrhl RNA was downregulated on Wnt3a treatment in Gc1-Spg cells. This study shows that mrhl RNA plays a negative role in Wnt signaling in mouse spermatogonial cells through its interaction with p68.

Transcripts with in silico predicted RNA structure are enriched everywhere in the mouse brain

Background

Post-transcriptional control of gene expression is mostly conducted by specific elements in untranslated regions (UTRs) of mRNAs, in collaboration with specific binding proteins and RNAs. In several well characterized cases, these RNA elements are known to form stable secondary structures. RNA secondary structures also may have major functional implications for long noncoding RNAs (lncRNAs). Recent transcriptional data has indicated the importance of lncRNAs in brain development and function. However, no methodical efforts to investigate this have been undertaken. Here, we aim to systematically analyze the potential for RNA structure in brain-expressed transcripts.

Results

By comprehensive spatial expression analysis of the adult mouse in situ hybridization data of the Allen Mouse Brain Atlas, we show that transcripts (coding as well as non-coding) associated with in silico predicted structured probes are highly and significantly enriched in almost all analyzed brain regions. Functional implications of these RNA structures and their role in the brain are discussed in detail along with specific examples. We observe that mRNAs with a structure prediction in their UTRs are enriched for binding, transport and localization gene ontology categories. In addition, after manual examination we observe agreement between RNA binding protein interaction sites near the 3’ UTR structures and correlated expression patterns.

Conclusions

Our results show a potential use for RNA structures in expressed coding as well as noncoding transcripts in the adult mouse brain, and describe the role of structured RNAs in the context of intracellular signaling pathways and regulatory networks. Based on this data we hypothesize that RNA structure is widely involved in transcriptional and translational regulatory mechanisms in the brain and ultimately plays a role in brain function.

Molecular ecological network analyses

BACKGROUND

Understanding the interaction among different species within a community and their responses to environmental changes is a central goal in ecology. However, defining the network structure in a microbial community is very challenging due to their extremely high diversity and as-yet uncultivated status. Although recent advance of metagenomic technologies, such as high throughout sequencing and functional gene arrays, provide revolutionary tools for analyzing microbial community structure, it is still difficult to examine network interactions in a microbial community based on high-throughput metagenomics data.

RESULTS

Here, we describe a novel mathematical and bioinformatics framework to construct ecological association networks named molecular ecological networks (MENs) through Random Matrix Theory (RMT)-based methods. Compared to other network construction methods, this approach is remarkable in that the network is automatically defined and robust to noise, thus providing excellent solutions to several common issues associated with high-throughput metagenomics data. We applied it to determine the network structure of microbial communities subjected to long-term experimental warming based on pyrosequencing data of 16 S rRNA genes. We showed that the constructed MENs under both warming and unwarming conditions exhibited topological features of scale free, small world and modularity, which were consistent with previously described molecular ecological networks. Eigengene analysis indicated that the eigengenes represented the module profiles relatively well. In consistency with many other studies, several major environmental traits including temperature and soil pH were found to be important in determining network interactions in the microbial communities examined. To facilitate its application by the scientific community, all these methods and statistical tools have been integrated into a comprehensive Molecular Ecological Network Analysis Pipeline (MENAP), which is open-accessible now (http://ieg2.ou.edu/MENA).

CONCLUSIONS

The RMT-based molecular ecological network analysis provides powerful tools to elucidate network interactions in microbial communities and their responses to environmental changes, which are fundamentally important for research in microbial ecology and environmental microbiology.

Molecular Mechanisms of Same TCM Syndrome for Different Diseases and Different TCM Syndrome for Same Disease in Chronic Hepatitis B and Liver Cirrhosis

Traditional Chinese medicine (TCM) treatment is based on the traditional diagnose method to distinguish the TCM syndrome, not the disease. So there is a phenomenon in the relationship between TCM syndrome and disease, called Same TCM Syndrome for Different Diseases and Different TCM Syndrome for Same Disease. In this study, we demonstrated the molecular mechanisms of this phenomenon using the microarray samples of liver-gallbladder dampness-heat syndrome (LGDHS) and liver depression and spleen deficiency syndrome (LDSDS) in the chronic hepatitis B (CHB) and liver cirrhosis (LC). The results showed that the difference between CHB and LC was gene expression level and the difference between LGDHS and LDSDS was gene coexpression in the G-protein-coupled receptor protein-signaling pathway. Therein genes GPER, PTHR1, GPR173, and SSTR1 were coexpressed in LDSDS, but not in LGDHS. Either CHB or LC was divided into the alternative LGDHS and LDSDS by the gene correlation, which reveals the molecular feature of Different TCM Syndrome for Same Disease. The alternatives LGDHS and LDSDS were divided into either CHB or LC by the gene expression level, which reveals the molecular feature of Same TCM Syndrome for Different Diseases.

Genetic locus on mouse chromosome 7 controls elevated heart rate

Elevated heart rate (HR) is a risk factor for cardiovascular diseases. The goal of the study was to map HR trait in mice using quantitative trait locus (QTL) analysis followed by genome-wide association (GWA) analysis. The first approach provides mapping power and the second increases genome resolution. QTL analyses were performed in a C3HeB×SJL backcross. HR and systolic blood pressure (SBP) were measured by the tail-cuff plethysmography. HR was ∼80 beats/min higher in SJL compared with C3HeB. There was a wide distribution of the HR (536-763 beats/min) in N2 mice. We discovered a highly significant QTL (logarithm of odds = 6.7, P < 0.001) on chromosome 7 (41 cM) for HR in the C3HeB×SJL backcross. In the Hybrid Mouse Diversity Panel (58 strains, n = 5-6/strain) we found that HR (beats/min) ranged from 546 ± 12 in C58/J to 717 ± 7 in MA/MyJ mice. SBP (mmHg) ranged from 99 ± 6 in strain I/LnJ to 151 ± 4 in strain BXA4/PgnJ. GWA analyses were done using the HMDP, which revealed a locus (64.2-65.1 Mb) on chromosome 7 that colocalized with the QTL for elevated HR found in the C3HeB×SJL backcross. The peak association was observed for 17 SNPs that are localized within three GABA(A) receptor genes. In summary, we used a combined genetic approach to fine map a novel elevated HR locus on mouse chromosome 7.

Utilizing RNA-Seq data for de novo coexpression network inference

MOTIVATION

RNA-Seq experiments have shown great potential for transcriptome profiling. While sequencing increases the level of biological detail, integrative data analysis is also important. One avenue is the construction of coexpression networks. Because the capacity of RNA-Seq data for network construction has not been previously evaluated, we constructed a coexpression network using striatal samples, derived its network properties and compared it with microarray-based networks.

RESULTS

The RNA-Seq coexpression network displayed scale-free, hierarchical network structure. We detected transcripts groups (modules) with correlated profiles; modules overlap distinct ontology categories. Neuroanatomical data from the Allen Brain Atlas reveal several modules with spatial colocalization. The network was compared with microarray-derived networks; correlations from RNA-Seq data were higher, likely because greater sensitivity and dynamic range. Higher correlations result in higher network connectivity, heterogeneity and centrality. For transcripts present across platforms, network structure appeared largely preserved. From this study, we present the first RNA-Seq data de novo network inference.

Association between SNPs and gene expression in multiple regions of the human brain

Identifying the genetic cis associations between DNA variants (single-nucleotide polymorphisms (SNPs)) and gene expression in brain tissue may be a promising approach to find functionally relevant pathways that contribute to the etiology of psychiatric disorders. In this study, we examined the association between genetic variations and gene expression in prefrontal cortex, hippocampus, temporal cortex, thalamus and cerebellum in subjects with psychiatric disorders and in normal controls. We identified cis associations between 648 transcripts and 6725 SNPs in the various brain regions. Several SNPs showed brain regional-specific associations. The expression level of only one gene, PDE4DIP, was associated with a SNP, rs12124527, in all the brain regions tested here. From our data, we generated a list of brain cis expression quantitative trait loci (eQTL) genes that we compared with a list of schizophrenia candidate genes downloaded from the Schizophrenia Forum (SZgene) database (http://www.szgene.org/). Of the SZgene candidate genes, we found that the expression levels of four genes, HTR2A, PLXNA2, SRR and TCF4, were significantly associated with cis SNPs in at least one brain region tested. One gene, SRR, was also involved in a coexpression module that we found to be associated with disease status. In addition, a substantial number of cis eQTL genes were also involved in the module, suggesting eQTL analysis of brain tissue may identify more reliable susceptibility genes for schizophrenia than case-control genetic association analyses. In an attempt to facilitate the identification of genetic variations that may underlie the etiology of major psychiatric disorders, we have integrated the brain eQTL results into a public and online database, Stanley Neuropathology Consortium Integrative Database (SNCID; http://sncid.stanleyresearch.org).

Gene regulatory network inference: evaluation and application to ovarian cancer allows the prioritization of drug targets

Background

Altered networks of gene regulation underlie many complex conditions, including cancer. Inferring gene regulatory networks from high-throughput microarray expression data is a fundamental but challenging task in computational systems biology and its translation to genomic medicine. Although diverse computational and statistical approaches have been brought to bear on the gene regulatory network inference problem, their relative strengths and disadvantages remain poorly understood, largely because comparative analyses usually consider only small subsets of methods, use only synthetic data, and/or fail to adopt a common measure of inference quality.

Methods

We report a comprehensive comparative evaluation of nine state-of-the art gene regulatory network inference methods encompassing the main algorithmic approaches (mutual information, correlation, partial correlation, random forests, support vector machines) using 38 simulated datasets and empirical serous papillary ovarian adenocarcinoma expression-microarray data. We then apply the best-performing method to infer normal and cancer networks. We assess the druggability of the proteins encoded by our predicted target genes using the CancerResource and PharmGKB webtools and databases.

Results

We observe large differences in the accuracy with which these methods predict the underlying gene regulatory network depending on features of the data, network size, topology, experiment type, and parameter settings. Applying the best-performing method (the supervised method SIRENE) to the serous papillary ovarian adenocarcinoma dataset, we infer and rank regulatory interactions, some previously reported and others novel. For selected novel interactions we propose testable mechanistic models linking gene regulation to cancer. Using network analysis and visualization, we uncover cross-regulation of angiogenesis-specific genes through three key transcription factors in normal and cancer conditions. Druggabilty analysis of proteins encoded by the 10 highest-confidence target genes, and by 15 genes with differential regulation in normal and cancer conditions, reveals 75% to be potential drug targets.

Conclusions

Our study represents a concrete application of gene regulatory network inference to ovarian cancer, demonstrating the complete cycle of computational systems biology research, from genome-scale data analysis via network inference, evaluation of methods, to the generation of novel testable hypotheses, their prioritization for experimental validation, and discovery of potential drug targets.

Exploring the transcriptome of ciliated cells using in silico dissection of human tissues

Cilia are cell organelles that play important roles in cell motility, sensory and developmental functions and are involved in a range of human diseases, known as ciliopathies. Here, we search for novel human genes related to cilia using a strategy that exploits the previously reported tendency of cell type-specific genes to be coexpressed in the transcriptome of complex tissues. Gene coexpression networks were constructed using the noise-resistant WGCNA algorithm in 12 publicly available microarray datasets from human tissues rich in motile cilia: airways, fallopian tubes and brain. A cilia-related coexpression module was detected in 10 out of the 12 datasets. A consensus analysis of this module's gene composition recapitulated 297 known and predicted 74 novel cilia-related genes. 82% of the novel candidates were supported by tissue-specificity expression data from GEO and/or proteomic data from the Human Protein Atlas. The novel findings included a set of genes (DCDC2, DYX1C1, KIAA0319) related to a neurological disease dyslexia suggesting their potential involvement in ciliary functions. Furthermore, we searched for differences in gene composition of the ciliary module between the tissues. A multidrug-and-toxin extrusion transporter MATE2 (SLC47A2) was found as a brain-specific central gene in the ciliary module. We confirm the localization of MATE2 in cilia by immunofluorescence staining using MDCK cells as a model. While MATE2 has previously gained attention as a pharmacologically relevant transporter, its potential relation to cilia is suggested for the first time. Taken together, our large-scale analysis of gene coexpression networks identifies novel genes related to human cell cilia.

Network analysis of oyster transcriptome revealed a cascade of cellular responses during recovery after heat shock

Oysters, as a major group of marine bivalves, can tolerate a wide range of natural and anthropogenic stressors including heat stress. Recent studies have shown that oysters pretreated with heat shock can result in induced heat tolerance. A systematic study of cellular recovery from heat shock may provide insights into the mechanism of acquired thermal tolerance. In this study, we performed the first network analysis of oyster transcriptome by reanalyzing microarray data from a previous study. Network analysis revealed a cascade of cellular responses during oyster recovery after heat shock and identified responsive gene modules and key genes. Our study demonstrates the power of network analysis in a non-model organism with poor gene annotations, which can lead to new discoveries that go beyond the focus on individual genes.

Using bioinformatics and systems genetics to dissect HDL-cholesterol genetics in an MRL/MpJ x SM/J intercross

A higher incidence of coronary artery disease is associated with a lower level of HDL-cholesterol. We searched for genetic loci influencing HDL-cholesterol in F2 mice from a cross between MRL/MpJ and SM/J mice. Quantitative trait loci (QTL) mapping revealed one significant HDL QTL (Apoa2 locus), four suggestive QTL on chromosomes 10, 11, 13, and 18 and four additional QTL on chromosomes 1 proximal, 3, 4, and 7 after adjusting HDL for the strong Apoa2 locus. A novel nonsynonymous polymorphism supports Lipg as the QTL gene for the chromosome 18 QTL, and a difference in Abca1 expression in liver tissue supports it as the QTL gene for the chromosome 4 QTL. Using weighted gene co-expression network analysis, we identified a module that after adjustment for Apoa2, correlated with HDL, was genetically determined by a QTL on chromosome 11, and overlapped with the HDL QTL. A combination of bioinformatics tools and systems genetics helped identify several candidate genes for both the chromosome 11 HDL and module QTL based on differential expression between the parental strains, cis regulation of expression, and causality modeling. We conclude that integrating systems genetics to a more-traditional genetics approach improves the power of complex trait gene identification.

Molecular microcircuitry underlies functional specification in a basal ganglia circuit dedicated to vocal learning

Similarities between speech and birdsong make songbirds advantageous for investigating the neurogenetics of learned vocal communication--a complex phenotype probably supported by ensembles of interacting genes in cortico-basal ganglia pathways of both species. To date, only FoxP2 has been identified as critical to both speech and birdsong. We performed weighted gene coexpression network analysis on microarray data from singing zebra finches to discover gene ensembles regulated during vocal behavior. We found ∼2,000 singing-regulated genes comprising three coexpression groups unique to area X, the basal ganglia subregion dedicated to learned vocalizations. These contained known targets of human FOXP2 and potential avian targets. We validated biological pathways not previously implicated in vocalization. Higher-order gene coexpression patterns, rather than expression levels, molecularly distinguish area X from the ventral striato-pallidum during singing. The previously unknown structure of singing-driven networks enables prioritization of molecular interactors that probably bear on human motor disorders, especially those affecting speech.

A progressive mouse model of Parkinson's disease: the Thy1-aSyn ("Line 61") mice

Identification of mutations that cause rare familial forms of Parkinson's disease (PD) and subsequent studies of genetic risk factors for sporadic PD have led to an improved understanding of the pathological mechanisms that may cause nonfamilial PD. In particular, genetic and pathological studies strongly suggest that alpha-synuclein, albeit very rarely mutated in PD patients, plays a critical role in the vast majority of individuals with the sporadic form of the disease. We have extensively characterized a mouse model over-expressing full-length, human, wild-type alpha-synuclein under the Thy-1 promoter. We have also shown that this model reproduces many features of sporadic PD, including progressive changes in dopamine release and striatal content, alpha-synuclein pathology, deficits in motor and nonmotor functions that are affected in pre-manifest and manifest phases of PD, inflammation, and biochemical and molecular changes similar to those observed in PD. Preclinical studies have already demonstrated improvement with promising new drugs in this model, which provides an opportunity to test novel neuroprotective strategies during different phases of the disorder using endpoint measures with high power to detect drug effects.

Integrative genomics strategies to elucidate the complexity of drug response

Pharmacogenomic investigation from both genome-wide association studies and experiments focused on candidate loci involved in drug mechanism and metabolism has yielded a substantial and increasing list of robust genetic effects on drug therapy in humans. At the same time, reasonably comprehensive molecular data such as gene expression, proteomic and metabolomic data are now available for collections of hundreds to thousands of individuals. If these data are structured in a statistically robust and computationally tractable way, such as a network model, they can aid in the analysis of new pharmacogenomics studies by suggesting novel hypotheses for the regulation of genes involved in drug metabolism and response. Similarly, hypotheses taken from these same models can direct genome-wide association studies by focusing the genome-wide association studies analysis on a number of specific hypotheses informed by the relationships customarily seen between a gene's expression or protein activity and genetic variation at a particular locus. Network models based on other sorts of systematic biological data such as cell-based surveys of drug effect on gene expression and mining of literature and electronic medical records for associations between clinical and molecular phenotypes also promise similar utility. Although surely primitive in comparison with what will be developed, these model-based approaches to leveraging the increasing volume of data generated in the course of patient care and medical research nevertheless suggest a huge opportunity to improve our understanding of biological systems involved in pharmacogenomics and apply them to questions of medical relevance.

Predicting glioblastoma prognosis networks using weighted gene co-expression network analysis on TCGA data

Background

Using gene co-expression analysis, researchers were able to predict clusters of genes with consistent functions that are relevant to cancer development and prognosis. We applied a weighted gene co-expression network (WGCN) analysis algorithm on glioblastoma multiforme (GBM) data obtained from the TCGA project and predicted a set of gene co-expression networks which are related to GBM prognosis.

Methods

We modified the Quasi-Clique Merger algorithm (QCM algorithm) into edge-covering Quasi-Clique Merger algorithm (eQCM) for mining weighted sub-network in WGCN. Each sub-network is considered a set of features to separate patients into two groups using K-means algorithm. Survival times of the two groups are compared using log-rank test and Kaplan-Meier curves. Simulations using random sets of genes are carried out to determine the thresholds for log-rank test p-values for network selection. Sub-networks with p-values less than their corresponding thresholds were further merged into clusters based on overlap ratios (>50%). The functions for each cluster are analyzed using gene ontology enrichment analysis.

Results

Using the eQCM algorithm, we identified 8,124 sub-networks in the WGCN, out of which 170 sub-networks show p-values less than their corresponding thresholds. They were then merged into 16 clusters.

Conclusions

We identified 16 gene clusters associated with GBM prognosis using the eQCM algorithm. Our results not only confirmed previous findings including the importance of cell cycle and immune response in GBM, but also suggested important epigenetic events in GBM development and prognosis.

Networks of neuronal genes affected by common and rare variants in autism spectrum disorders

Autism spectrum disorders (ASD) are neurodevelopmental disorders with phenotypic and genetic heterogeneity. Recent studies have reported rare and de novo mutations in ASD, but the allelic architecture of ASD remains unclear. To assess the role of common and rare variations in ASD, we constructed a gene co-expression network based on a widespread survey of gene expression in the human brain. We identified modules associated with specific cell types and processes. By integrating known rare mutations and the results of an ASD genome-wide association study (GWAS), we identified two neuronal modules that are perturbed by both rare and common variations. These modules contain highly connected genes that are involved in synaptic and neuronal plasticity and that are expressed in areas associated with learning and memory and sensory perception. The enrichment of common risk variants was replicated in two additional samples which include both simplex and multiplex families. An analysis of the combined contribution of common variants in the neuronal modules revealed a polygenic component to the risk of ASD. The results of this study point toward contribution of minor and major perturbations in the two sub-networks of neuronal genes to ASD risk.

Fast R Functions for Robust Correlations and Hierarchical Clustering

Many high-throughput biological data analyses require the calculation of large correlation matrices and/or clustering of a large number of objects. The standard R function for calculating Pearson correlation can handle calculations without missing values efficiently, but is inefficient when applied to data sets with a relatively small number of missing data. We present an implementation of Pearson correlation calculation that can lead to substantial speedup on data with relatively small number of missing entries. Further, we parallelize all calculations and thus achieve further speedup on systems where parallel processing is available. A robust correlation measure, the biweight midcorrelation, is implemented in a similar manner and provides comparable speed. The functions cor and bicor for fast Pearson and biweight midcorrelation, respectively, are part of the updated, freely available R package WGCNA.The hierarchical clustering algorithm implemented in R function hclust is an order n(3) (n is the number of clustered objects) version of a publicly available clustering algorithm (Murtagh 2012). We present the package flashClust that implements the original algorithm which in practice achieves order approximately n(2), leading to substantial time savings when clustering large data sets.