Regulatory networks: Linking microarray data to systems biology

Systems genetics of the nuclear factor-κB signal transduction network. I. Detection of several quantitative trait loci potentially relevant to aging

A theory of aging holds that senescence is caused by a dysregulated nuclear factor kappa B (NF-κB) signal transduction network (STN). We adopted a systems genetics approach in our study of the NF-κB STN. Ingenuity Pathways Analysis (IPA) was used to identify gene/gene product interactions between NF-κB and the genes in our transcriptional profiling array. Principal components factor analysis (PCFA) was performed on a sub-network of 19 genes, including two initiators of the toll-like receptor (TLR) pathway, myeloid differentiation primary response gene (88) (MyD88) and TIR (Toll/interleukin-1 receptor)-domain-containing adapter-inducing interferon-β (TRIF). TLR pathways are either MyD88-dependent or TRIF-dependent. Therefore, we also performed PCFA on a subset excluding the MyD88 transcript, and on another subset excluding two TRIF transcripts. Using linkage analysis we found that each set gave rise to at least one factor with a logarithm of the odds (LOD) score greater than 3, two on chromosome 15 at 15q12 and 15q22.2, and another two on chromosome 17 at 17p13.3 and 17q25.3. We also found several suggestive signals (2<LOD score<3) at 1q32.1, 1q41, 2q34, 3q23, and 7p15.3. We are currently examining potential associations with single nucleotide polymorphisms within the 1-LOD intervals of our linkage signals.

Elevated osteonectin/SPARC expression in primary prostate cancer predicts metastatic progression

BACKGROUND

The majority of prostate cancers (CaP) are detected in early stages with uncertain prognosis. Therefore, an intensive effort is underway to define early predictive markers of CaP with aggressive progression characteristics.

METHODS

In order to define such prognostic markers, we performed comparative analyses of transcriptomes of well- and poorly differentiated (PD) tumor cells from primary tumors of patients (N=40) with 78 months of mean follow-up after radical prostatectomy. Validation experiments were carried out at transcript level by quantitative real-time reverse transcriptase-PCR (RT-PCR) (N=110) and at protein level by immunohistochemistry (N=53) in primary tumors from an independent patient cohort.

RESULTS

Association of a biochemical network of 12 genes with SPARC gene as a central node was highlighted with PD phenotype. Of note, there was remarkable enrichment of NKXH_NKXH_HOX composite regulatory elements in the promoter of the genes in this network suggesting a biological significance of this gene-expression regulatory mechanism in CaP progression. Further, quantitative expression analyses of SPARC mRNA in primary prostate tumor cells of 110 patients validated the association of SPARC expression with poor differentiation and higher Gleason score. Most significantly, higher SPARC protein expression at the time of prostatectomy was associated with the subsequent development of metastasis (P=0.0006, AUC=0.803).

CONCLUSIONS

In summary, we propose that evaluation of SPARC in primary CaP has potential as a prognostic marker of metastatic progression.

[Next generation sequencing technologies (NGST) -- development and applications]

In the past ten years the development of next generation sequencing technologies brought a new era in the field of quick and efficient DNA sequencing. In our study we give an overview of the methodological achievements from Sanger's chain-termination sequencing in 1975 to those allowing real-time DNA sequencing today. Sequencing methods that utilize clonal amplicons for parallel multistrand sequencing comprise the basics of currently available next generation sequencing techniques. Nowadays next generation sequencing is mainly used for basic research in functional genomics, providing quintessential information in the meta-analyses of data from signal transduction pathways, onthologies, proteomics and metabolomics. Although next generation sequencing is yet sparsely used in clinical practice, cardiology, oncology and epidemiology already show an immense need for the additional knowledge obtained by this new technology. The main barrier of its spread is the lack of standardization of analysis evaluation methods, which obscure objective assessment of the results.

BiologicalNetworks 2.0--an integrative view of genome biology data

Background

A significant problem in the study of mechanisms of an organism's development is the elucidation of interrelated factors which are making an impact on the different levels of the organism, such as genes, biological molecules, cells, and cell systems. Numerous sources of heterogeneous data which exist for these subsystems are still not integrated sufficiently enough to give researchers a straightforward opportunity to analyze them together in the same frame of study. Systematic application of data integration methods is also hampered by a multitude of such factors as the orthogonal nature of the integrated data and naming problems.

Results

Here we report on a new version of BiologicalNetworks, a research environment for the integral visualization and analysis of heterogeneous biological data. BiologicalNetworks can be queried for properties of thousands of different types of biological entities (genes/proteins, promoters, COGs, pathways, binding sites, and other) and their relations (interactions, co-expression, co-citations, and other). The system includes the build-pathways infrastructure for molecular interactions/relations and module discovery in high-throughput experiments. Also implemented in BiologicalNetworks are the Integrated Genome Viewer and Comparative Genomics Browser applications, which allow for the search and analysis of gene regulatory regions and their conservation in multiple species in conjunction with molecular pathways/networks, experimental data and functional annotations.

Conclusions

The new release of BiologicalNetworks together with its back-end database introduces extensive functionality for a more efficient integrated multi-level analysis of microarray, sequence, regulatory, and other data. BiologicalNetworks is freely available at http://www.biologicalnetworks.org.

Untangling the intracellular signalling network in cancer--a strategy for data integration in acute myeloid leukaemia

Protein and gene networks centred on the regulatory tumour suppressor proteins may be of crucial importance both in carcinogenesis and in the response to chemotherapy. Tumour suppressor protein p53 integrates intracellular data in stress responses, receiving signals and translating these into differential gene expression. Interpretation of the data integrated on p53 may therefore reveal the response to therapy in cancer. Proteomics offers more specific data - closer to "the real action" - than the hitherto more frequently used gene expression profiling. Integrated data analysis may reveal pathways disrupted at several regulatory levels. Ultimately, integrated data analysis may also contribute to finding key underlying cancer genes. We here proposes a Partial Least Squares Regression (PLSR)-based data integration strategy, which allows simultaneous analysis of proteomic data, gene expression data and classical clinical parameters. PLSR collapses multidimensional data into fewer relevant dimensions for data interpretation. PLSR can also aid identification of functionally important modules by also performing comparison to databases on known biological interactions. Further, PLSR allows meaningful visualization of complex datasets, aiding interpretation of the underlying biology. Extracting the true biological causal mechanisms from heterogeneous patient populations is the key to discovery of new therapeutic options in cancer.

Next generation sequencing in functional genomics

Genome-wide sequencing has enabled modern biomedical research to relate more and more events in healthy as well as disease-affected cells and tissues to the genomic sequence. Now next generation sequencing (NGS) extends that reach into multiple almost complete genomes of the same species, revealing more and more details about how individual genomes as well as individual aspects of their regulation differ from each other. The inclusion of NGS-based transcriptome sequencing, chromatin-immunoprecipitation (ChIP) of transcription factor binding and epigenetic analyses (usually based on DNA methylation or histone modification ChIP) completes the picture with unprecedented resolution enabling the detection of even subtle differences such as alternative splicing of individual exons. Functional genomics aims at the elucidation of the molecular basis of biological functions and requires analyses that go far beyond the primary analysis of the reads such as mapping to a genome template sequence. The various and complex interactions between the genome, gene products and metabolites define biological function, which necessitates inclusion of results other than sequence tags in quite elaborative approaches. However, the extra efforts pay off in revealing mechanisms as well as providing the foundation for new strategies in systems biology and personalized medicine. This review emphasizes the particular contribution NGS-based technologies make to functional genomics research with a special focus on gene regulation by transcription factor binding sites.

The systems view in addiction research

The wealth of information accessible on the molecular level after completion of sequencing of the human and other genomes and in conjunction with new high-throughput technologies like microarrays has paved the way for research on whole systems rather than on single components. Here we describe exemplarily the construction of a rather complex molecular network involved in alcohol addiction by using information from DNA-microarray studies in alcohol-dependent animals. In this network, haemoglobin downregulation in different parts of the brain reward system plays a central role in affecting synaptic plasticity, circadian rhythmicity and opioid receptors. Furthermore, we discuss the dynamic aspect of biological systems with respect to repeated and intermittent drug intake. This aspect can best be captured by the allostatic model on the molecular level. Using a molecular oscillator model where levels of oscillations are changed by repetitive drug administration, changes in set point adjustment are described that underlay allostatic shifts in drug reinforcement processes.

A modular approach for integrative analysis of large-scale gene-expression and drug-response data

High-throughput technologies are now used to generate more than one type of data from the same biological samples. To properly integrate such data, we propose using co-modules, which describe coherent patterns across paired data sets, and conceive several modular methods for their identification. We first test these methods using in silico data, demonstrating that the integrative scheme of our Ping-Pong Algorithm uncovers drug-gene associations more accurately when considering noisy or complex data. Second, we provide an extensive comparative study using the gene-expression and drug-response data from the NCI-60 cell lines. Using information from the DrugBank and the Connectivity Map databases we show that the Ping-Pong Algorithm predicts drug-gene associations significantly better than other methods. Co-modules provide insights into possible mechanisms of action for a wide range of drugs and suggest new targets for therapy.

Motif discovery in tissue-specific regulatory sequences using directed information

Motif discovery for the identification of functional regulatory elements underlying gene expression is a challenging problem. Sequence inspection often leads to discovery of novel motifs (including transcription factor sites) with previously uncharacterized function in gene expression. Coupled with the complexity underlying tissue-specific gene expression, there are several motifs that are putatively responsible for expression in a certain cell type. This has important implications in understanding fundamental biological processes such as development and disease progression. In this work, we present an approach to the identification of motifs (not necessarily transcription factor sites) and examine its application to some questions in current bioinformatics research. These motifs are seen to discriminate tissue-specific gene promoter or regulatory regions from those that are not tissue-specific. There are two main contributions of this work. Firstly, we propose the use of directed information for such classification constrained motif discovery, and then use the selected features with a support vector machine (SVM) classifier to find the tissue specificity of any sequence of interest. Such analysis yields several novel interesting motifs that merit further experimental characterization. Furthermore, this approach leads to a principled framework for the prospective examination of any chosen motif to be discriminatory motif for a group of coexpressed/coregulated genes, thereby integrating sequence and expression perspectives. We hypothesize that the discovery of these motifs would enable the large-scale investigation for the tissue-specific regulatory role of any conserved sequence element identified from genome-wide studies.

A framework to identify physiological responses in microarray-based gene expression studies: selection and interpretation of biologically relevant genes

In whole genome microarray studies major gene expression changes are easily identified, but it is a challenge to capture small, but biologically important, changes. Pathway-based programs can capture small effects but may have the disadvantage of being restricted to functionally annotated genes. A structured approach toward the identification of major and small changes for interpretation of biological effects is needed. We present a structured approach, a framework, that addresses different considerations in 1) the identification of informative genes in microarray data sets and 2) the interpretation of their biological relevance. The steps of this framework include gene ranking, gene selection, gene grouping, and biological interpretation. Random forests (RF), which takes gene-gene interactions into account, is examined to rank and select genes. For human, mouse, and rat whole genome arrays, less than half of the probes on the array are annotated. Consequently, pathway analysis tools ignore half of the information present in the microarray data set. The framework described takes all genes into account. RF is a useful tool to rank genes by taking interactions into account. Applying a permutation approach, we were able to define an objective threshold for gene selection. RF combined with self-organizing maps identified genes with coordinated but small gene expression responses that were not fully annotated but corresponded to the same biological process. The presented approach provides a flexible framework for biological interpretation of microarray data sets. It includes all genes in the data set, takes gene-gene interactions into account, and provides an objective threshold for gene selection.

Opportunities and challenges in applying genomics to the study of oogenesis and folliculogenesis in farm animals

Ovarian oogenesis and folliculogenesis are complex and coordinated biological processes which require a series of events that induce morphological and functional changes within the follicle, leading to cell differentiation and oocyte development. In this context, the challenge of the researchers is to describe the dynamics of gene expression in the different compartments and their interactions during the follicular programme. In recent years, high-throughput arrays have become a powerful tool with which to compare the whole population of transcripts in a single experiment. Here, we review the challenges of applying genomics to this model in farm animal species. The first limitation lies in limited the availability of biological material, which makes the study of the follicle compartments (oocyte, granulosa cells and thecal cells) or early embryo much more difficult. The concept of observing all transcripts at once is very attractive but despite progress in sequencing, the genome annotation remains very incomplete in non-model species. Particularly, oogenesis and early embryo development relate to the high proportion of unknown expressed sequence tags. Then, it is important to consider post-transcriptional and translational regulation to understand the role of these genes. Ultimately, these new inferred insights will still have to be validated by functional approaches. In addition toin vitroorex vivofunctional approaches, both ‘natural mutant’ ewe models and RNA interference represent, at the moment, the best hope for functional genomics. Advances in our understanding of reproductive physiology should be facilitated by gene expression data exchange and translation into a better understanding of the underlying biological phenomena.

Bioinformatics applications for pathway analysis of microarray data

Changes in transcript levels are assessed by microarray analysis on an individual basis, essentially resulting in long lists of genes that were found to have significantly changed transcript levels. However, in biology these changes do not occur as independent events as such lists suggest, but in a highly coordinated and interdependent manner. Understanding the biological meaning of the observed changes requires elucidating such biological interdependencies. The most common way to achieve this is to project the gene lists onto distinct biological processes often represented in the form of gene-ontology (GO) categories or metabolic and regulatory pathways as derived from literature analysis. This review focuses on different approaches and tools employed for this task, starting form GO-ranking methods, covering pathway mappings, finally converging on biological network analysis. A brief outlook of the application of such approaches to the newest microarray-based technologies (Chromatin-ImmunoPrecipitation, ChIP-on-chip) concludes the review.

Toward a control theory analysis of aging

Aging is due to the accumulation of damage over time that affects the function and survival of the organism; however, it has proven difficult to infer the relative importance of the many processes that contribute to aging. To address this, here we outline an approach that may prove useful in analyzing aging. In this approach, the function of the organism is described as a set of interacting physiological systems. Degradation of their outputs leads to functional decline and death as a result of aging. In turn, degradation of the system outputs is attributable to changes at the next hierarchical level down, the cell, through changes in cell number or function, which are in turn a consequence of the metabolic history of the cell. Within this framework, we then adapt the methods of metabolic control analysis (MCA) to determine which modifications are important for aging. This combination of a hierarchical framework and the methodologies of MCA may prove useful both for thinking about aging and for analyzing it experimentally.