Integration Of Biological Data Research Articles

Methods for biological data integration: perspectives and challenges.

Rapid technological advances have led to the production of different types of biological data and enabled construction of complex networks with various types of interactions between diverse biological entities. Standard network data analysis methods were shown to be limited in dealing with such heterogeneous networked data and consequently, new methods for integrative data analyses have been proposed. The integrative methods can collectively mine multiple types of biological data and produce more holistic, systems-level biological insights. We survey recent methods for collective mining (integration) of various types of networked biological data. We compare different state-of-the-art methods for data integration and highlight their advantages and disadvantages in addressing important biological problems. We identify the important computational challenges of these methods and provide a general guideline for which methods are suited for specific biological problems, or specific data types. Moreover, we propose that recent non-negative matrix factorization-based approaches may become the integration methodology of choice, as they are well suited and accurate in dealing with heterogeneous data and have many opportunities for further development.

Data integration in biological research: an overview.

Data sharing, integration and annotation are essential to ensure the reproducibility of the analysis and interpretation of the experimental findings. Often these activities are perceived as a role that bioinformaticians and computer scientists have to take with no or little input from the experimental biologist. On the contrary, biological researchers, being the producers and often the end users of such data, have a big role in enabling biological data integration. The quality and usefulness of data integration depend on the existence and adoption of standards, shared formats, and mechanisms that are suitable for biological researchers to submit and annotate the data, so it can be easily searchable, conveniently linked and consequently used for further biological analysis and discovery. Here, we provide background on what is data integration from a computational science point of view, how it has been applied to biological research, which key aspects contributed to its success and future directions.

An unsupervised learning approach to find ovarian cancer genes through integration of biological data.

Cancer is a disease characterized largely by the accumulation of out-of-control somatic mutations during the lifetime of a patient. Distinguishing driver mutations from passenger mutations has posed a challenge in modern cancer research. With the advanced development of microarray experiments and clinical studies, a large numbers of candidate cancer genes have been extracted and distinguishing informative genes out of them is essential. As a matter of fact, we proposed to find the informative genes for cancer by using mutation data from ovarian cancers in our framework. In our model we utilized the patient gene mutation profile, gene expression data and gene gene interactions network to construct a graphical representation of genes and patients. Markov processes for mutation and patients are triggered separately. After this process, cancer genes are prioritized automatically by examining their scores at their stationary distributions in the eigenvector. Extensive experiments demonstrate that the integration of heterogeneous sources of information is essential in finding important cancer genes.

A systematic investigation based on microRNA-mediated gene regulatory network reveals that dysregulation of microRNA-19a/Cyclin D1 axis confers an oncogenic potential and a worse prognosis in human hepatocellular carcinoma

MicroRNAs (miRNAs) contribute to a wide variety of human diseases by regulating gene expression, leading to imbalances in gene regulatory networks. To discover novel hepatocellular carcinoma (HCC)-related miRNA-target axes and to elucidate their functions, we here performed a systematic investigation combining biological data acquisition and integration, miRNA-target prediction, network construction, functional assay and clinical validation. As a result, a total of 117 HCC differentially expressed miRNAs were identified, and 728 high confident target genes of these miRNAs were collected. Then, the interaction network of target genes was constructed and 221 key nodes with topological importance in the network were identified according to their topological features including degree, node-betweenness, closeness and K-coreness. Among these key nodes, Cyclin D1 had the highest node-betweenness, implying its bottleneck role in the network. Luciferase reporter assay confirmed that miRNA-19a, which was one of HCC downregulated miRNAs, directly targeted Cyclin D1 in HCC cells. Moreover, miR-19a might play inhibitory roles in HCC malignancy via regulating Cyclin D1 expression. Further clinical evidence also highlighted the prognostic potential of miR-19a/Cyclin D1 axis in HCC. In conclusion, this systematic investigation provides a framework to identify featured miRNAs and their target genes which are potent effectors in the occurrence and development of HCC. More importantly, miR-19a/Cyclin D1 axis might have promising applications as a therapeutic target and a prognostic marker for patients with HCC.

Dynamic ocean management: Defining and conceptualizing real-time management of the ocean

Most spatial marine management techniques (e.g., marine protected areas) draw stationary boundaries around often mobile marine features, animals, or resource users. While these approaches can work for relatively stationary marine resources, to be most effective marine management must be as fluid in space and time as the resources and users we aim to manage. Instead, a shift towards dynamic ocean management is suggested, defined as management that rapidly changes in space and time in response to changes in the ocean and its users through the integration of near real-time biological, oceanographic, social and/or economic data. Dynamic management can refine the temporal and spatial scale of managed areas, thereby better balancing ecological and economic objectives. Temperature dependent habitat of a hypothetical mobile marine species was simulated to show the efficiency of dynamic management, finding that 82.0 to 34.2 percent less area needed to be managed using a dynamic approach. Dynamic management further complements existing management by increasing the speed at which decisions are implemented using predefined protocols. With advances in data collection and sharing, particularly in remote sensing, animal tracking, and mobile technology, managers are poised to apply dynamic management across numerous marine sectors. Existing examples demonstrate that dynamic management can successfully allow managers to respond rapidly to changes on-the-water, however to implement dynamic ocean management widely, several gaps must be filled. These include enhancing legal instruments, incorporating ecological and socioeconomic considerations simultaneously, developing ‘out-of-the-box’ platforms to serve dynamic management data to users, and developing applications broadly across additional marine resource sectors.

A reproducible approach to high-throughput biological data acquisition and integration.

Modern biological research requires rapid, complex, and reproducible integration of multiple experimental results generated both internally and externally (e.g., from public repositories). Although large systematic meta-analyses are among the most effective approaches both for clinical biomarker discovery and for computational inference of biomolecular mechanisms, identifying, acquiring, and integrating relevant experimental results from multiple sources for a given study can be time-consuming and error-prone. To enable efficient and reproducible integration of diverse experimental results, we developed a novel approach for standardized acquisition and analysis of high-throughput and heterogeneous biological data. This allowed, first, novel biomolecular network reconstruction in human prostate cancer, which correctly recovered and extended the NFκB signaling pathway. Next, we investigated host-microbiome interactions. In less than an hour of analysis time, the system retrieved data and integrated six germ-free murine intestinal gene expression datasets to identify the genes most influenced by the gut microbiota, which comprised a set of immune-response and carbohydrate metabolism processes. Finally, we constructed integrated functional interaction networks to compare connectivity of peptide secretion pathways in the model organisms Escherichia coli, Bacillus subtilis, and Pseudomonas aeruginosa.

Identification of Tumor Subtypes of Endometrial Carcinoma by Integration of Heterogeneous Datasets

The Cancer Genome Atlas (TCGA) project has made available multiple heterogeneous datasets. Although several methodological approaches have been proposed for the heterogeneous data integration, there is no framework of sparse non-negative matrix factorization (NMF) for handling heterogeneous biological data integration. Here, we propose the block-weighted sparse NMF bwsNMF) to identify tumor subtypes of endometrial carcinoma by integrating gene expression, mutations, a protein-protein interaction network and a transcription factor target network.

Inflammatory bowel disease-the role of cross-sectional imaging techniques in the investigation of the small bowel.

Background: The diagnosis of inflammatory bowel disease (IBD) in children and adolescents is based on the integration of clinical, biological, endoscopic, histological and radiological data. Methods: The most important part of the diagnosis is the histology, which is acquired by endoscopy. Imaging of the small bowel has changed in recent years, but the imaging goals are primarily to determine the extent of small bowel involvement, assess complications and define candidates for surgery. Imaging techniques are divided into conventional and cross-sectional ones. Results: The spectrum of imaging findings of cross-sectional techniques is discussed, emphasising the advantages and limitations of each technique, acknowledging the specificities of the paediatric population. Cross-sectional techniques have advanced the ability to diagnose and monitor inflammatory disease of the small bowel. Conclusion: MR enterography is the technique of choice in children with known IBD, for the investigation of the small bowel and the whole GI tract. US should be the first choice examination in children with suspected IBD, while CT should be reserved for cases in which MRI is contraindicated or in acute emergency situations when US is inadequate.Teaching Points• Cross-sectional imaging of the small bowel is essential in paediatric IBD.• Endoscopy is unable to assess extramural disease and examine the entire small bowel.• US should be the first choice examination in children with suspected IBD.• MR enterography is the technique of choice in children with known IBD.• There are still controversies regarding the prediction of disease activity or fibrosis.

Graphical algorithm for integration of genetic and biological data: proof of principle using psoriasis as a model

Pathway analysis to reveal biological mechanisms for results from genetic association studies have great potential to better understand complex traits with major human disease impact. However, current approaches have not been optimized to maximize statistical power to identify enriched functions/pathways, especially when the genetic data derives from studies using platforms (e.g. Immunochip and Metabochip) customized to have pre-selected markers from previously identified top-rank loci. We present here a novel approach, called Minimum distance-based Enrichment Analysis for Genetic Association (MEAGA), with the potential to address both of these important concerns. MEAGA performs enrichment analysis using graphical algorithms to identify sub-graphs among genes and measure their closeness in interaction database. It also incorporates a statistic summarizing the numbers and total distances of the sub-graphs, depicting the overlap between observed genetic signals and defined function/pathway gene-sets. MEAGA uses sampling technique to approximate empirical and multiple testing-corrected P-values. We show in simulation studies that MEAGA is more powerful compared to count-based strategies in identifying disease-associated functions/pathways, and the increase in power is influenced by the shortest distances among associated genes in the interactome. We applied MEAGA to the results of a meta-analysis of psoriasis using Immunochip datasets, and showed that associated genes are significantly enriched in immune-related functions and closer with each other in the protein-protein interaction network. http://genome.sph.umich.edu/wiki/MEAGA CONTACT: : tsoi.teen@gmail.com or goncalo@umich.edu Supplementary data are available at Bioinformatics online.

Toward a better integration of biological data from precipitation manipulation experiments into Earth system models

The biological responses to precipitation within the terrestrial components of Earth system models, or land surface models (LSMs), are mechanistically simple and poorly constrained, leaving projections of terrestrial ecosystem functioning and feedbacks to climate change uncertain. A number of field experiments have been conducted or are underway to test how changing precipitation will affect terrestrial ecosystems. Results from these experiments have the potential to vastly improve modeled processes. However, the transformation of experimental results into model improvements still represents a grand challenge. Here we review the current state of precipitation manipulation experiments and the precipitation responses of biological processes in LSMs to explore how these experiments can help improve model realism. First, we discuss contemporary precipitation projections and then review the structure and function of current-generation LSMs. We then examine different experimental designs and discuss basic variables that, if measured, would increase a field experiment's usefulness in a modeling context. Next, we compare biological processes commonly measured in the field with their model analogs and find that, in many cases, the way these processes are measured in the field is not compatible with the way they are represented in LSMs, an effect that hinders model development. We then discuss the challenge of scaling from the plot to the globe. Finally, we provide a series of recommendations aimed to improve the connectivity between experiments and LSMs and conclude that studies designed from the perspective of researchers in both communities will provide the greatest benefit to the broader global change community.

A dedicated database system for handling multi-level data in systems biology.

BackgroundAdvances in high-throughput technologies have enabled extensive generation of multi-level omics data. These data are crucial for systems biology research, though they are complex, heterogeneous, highly dynamic, incomplete and distributed among public databases. This leads to difficulties in data accessibility and often results in errors when data are merged and integrated from varied resources. Therefore, integration and management of systems biological data remain very challenging.MethodsTo overcome this, we designed and developed a dedicated database system that can serve and solve the vital issues in data management and hereby facilitate data integration, modeling and analysis in systems biology within a sole database. In addition, a yeast data repository was implemented as an integrated database environment which is operated by the database system. Two applications were implemented to demonstrate extensibility and utilization of the system. Both illustrate how the user can access the database via the web query function and implemented scripts. These scripts are specific for two sample cases: 1) Detecting the pheromone pathway in protein interaction networks; and 2) Finding metabolic reactions regulated by Snf1 kinase.Results and conclusionIn this study we present the design of database system which offers an extensible environment to efficiently capture the majority of biological entities and relations encountered in systems biology. Critical functions and control processes were designed and implemented to ensure consistent, efficient, secure and reliable transactions. The two sample cases on the yeast integrated data clearly demonstrate the value of a sole database environment for systems biology research.

Integration of biological data by kernels on graph nodes allows prediction of new genes involved in mitotic chromosome condensation.

The advent of genome-wide RNA interference (RNAi)-based screens puts us in the position to identify genes for all functions human cells carry out. However, for many functions, assay complexity and cost make genome-scale knockdown experiments impossible. Methods to predict genes required for cell functions are therefore needed to focus RNAi screens from the whole genome on the most likely candidates. Although different bioinformatics tools for gene function prediction exist, they lack experimental validation and are therefore rarely used by experimentalists. To address this, we developed an effective computational gene selection strategy that represents public data about genes as graphs and then analyzes these graphs using kernels on graph nodes to predict functional relationships. To demonstrate its performance, we predicted human genes required for a poorly understood cellular function-mitotic chromosome condensation-and experimentally validated the top 100 candidates with a focused RNAi screen by automated microscopy. Quantitative analysis of the images demonstrated that the candidates were indeed strongly enriched in condensation genes, including the discovery of several new factors. By combining bioinformatics prediction with experimental validation, our study shows that kernels on graph nodes are powerful tools to integrate public biological data and predict genes involved in cellular functions of interest.

DINIES: drug–target interaction network inference engine based on supervised analysis

DINIES (drug–target interaction network inference engine based on supervised analysis) is a web server for predicting unknown drug–target interaction networks from various types of biological data (e.g. chemical structures, drug side effects, amino acid sequences and protein domains) in the framework of supervised network inference. The originality of DINIES lies in prediction with state-of-the-art machine learning methods, in the integration of heterogeneous biological data and in compatibility with the KEGG database. The DINIES server accepts any ‘profiles’ or precalculated similarity matrices (or ‘kernels’) of drugs and target proteins in tab-delimited file format. When a training data set is submitted to learn a predictive model, users can select either known interaction information in the KEGG DRUG database or their own interaction data. The user can also select an algorithm for supervised network inference, select various parameters in the method and specify weights for heterogeneous data integration. The server can provide integrative analyses with useful components in KEGG, such as biological pathways, functional hierarchy and human diseases. DINIES (http://www.genome.jp/tools/dinies/) is publicly available as one of the genome analysis tools in GenomeNet.

Applied and Translational Genomics for Human Genetics and Clinical Science

Applied and Translational Genomics for Human Genetics and Clinical Science

A Methodology and Architecture Embedding Quality Assessment in Data Integration

Data integration aims to combine heterogeneous information sources and to provide interfaces for accessing the integrated resource. Data integration is a collaborative task that may involve many people with different degrees of experience, knowledge of the application domain, and expectations relating to the integrated resource. It may be difficult to determine and control the quality of an integrated resource due to these factors. In this article, we propose a data integration methodology that has embedded within it iterative quality assessment and improvement of the integrated resource. We also propose an architecture for the realisation of this methodology. The quality assessment is based on an ontology representation of different users’ quality requirements and of the main elements of the integrated resource. We use description logic as the formal basis for reasoning about users’ quality requirements and for validating that an integrated resource satisfies these requirements. We define quality factors and associated metrics which enable the quality of alternative global schemas for an integrated resource to be assessed quantitively, and hence the improvement which results from the refinement of a global schema following our methodology to be measured. We evaluate our approach through a large-scale real-life case study in biological data integration in which an integrated resource is constructed from three autononous proteomics data sources.

Identifying potential therapeutic targets of a natural product Jujuboside B for insomnia through network pharmacology

Knowledge of the interactome improves the understanding of disease metabolism. Biological information about interactions among genes and their protein products, computationally extracted in the context of SysBiomics, can hint at molecular causes of diseases, be essential for understanding biological systems, and provide clues for new therapeutic approaches. Quick and efficient access to this data have become critical issues for biologists. We have implemented a computational platform that integrates pathway, protein–protein interaction, differentially expressed genome and literature mining data to result in comprehensive networks for insomnia and intervention effects of Jujuboside B (JuB). The interaction data were imported into Cytoscape software, a popular bioinformatics package for biological network visualization and data integration, for screening the central nodes of the network, exploiting functional study of the central node genes, exploring the mechanism of insomnia. Results showed that seven differentially expressed genes confirmed by Cytoscape as the central nodes of the network in insomnia had interactions, forming a complicated interaction network (77 nodes, 96 edges). Among gene nodes, HBA1, LEP, MAOA, PRNP, GHRL, CLOCK and SLC6A4 were verified as the genes with maximal differential expressions. Of note, we further observed that the HBA1, LEP, SLC6A4 and MAOA were JuB target genes. The interaction network of the differentially expressed genes, especially the central nodes of this network, can provide clues to the insomnia, early diagnosis and molecular targeted therapy. Our findings demonstrate that the integration of interaction network in genomic space can not only speed the genome-wide identification of drug targets but also find new applications for the existing drugs.

Model-Based Analysis for Qualitative Data: An Application in Drosophila Germline Stem Cell Regulation

Discovery in developmental biology is often driven by intuition that relies on the integration of multiple types of data such as fluorescent images, phenotypes, and the outcomes of biochemical assays. Mathematical modeling helps elucidate the biological mechanisms at play as the networks become increasingly large and complex. However, the available data is frequently under-utilized due to incompatibility with quantitative model tuning techniques. This is the case for stem cell regulation mechanisms explored in the Drosophila germarium through fluorescent immunohistochemistry. To enable better integration of biological data with modeling in this and similar situations, we have developed a general parameter estimation process to quantitatively optimize models with qualitative data. The process employs a modified version of the Optimal Scaling method from social and behavioral sciences, and multi-objective optimization to evaluate the trade-off between fitting different datasets (e.g. wild type vs. mutant). Using only published imaging data in the germarium, we first evaluated support for a published intracellular regulatory network by considering alternative connections of the same regulatory players. Simply screening networks against wild type data identified hundreds of feasible alternatives. Of these, five parsimonious variants were found and compared by multi-objective analysis including mutant data and dynamic constraints. With these data, the current model is supported over the alternatives, but support for a biochemically observed feedback element is weak (i.e. these data do not measure the feedback effect well). When also comparing new hypothetical models, the available data do not discriminate. To begin addressing the limitations in data, we performed a model-based experiment design and provide recommendations for experiments to refine model parameters and discriminate increasingly complex hypotheses.

The research on molecular network models of complex disease

Complex diseases such as cardiovascular disease, cancer and diabetes are major killers of human health. Unlike single gene defect diseases, complex diseases are usually believed to be associated with the interactions among genes and the interactions among genes and environmental factors. The underlying complex pathogeneses make both the early diagnosis and treatment difficult. Therefore the research of complex diseases is one of the major challenges of biomedical research in this century. Recently, the rapid accumulation of biological knowledge and multi-level high-throughput omics data, revolutionarily changed the research paradigm on complex disease. Instead of only focusing on single molecular, researchers are gradually extending their research to systematically analyzing genome-wide biomolecular interactions, i.e., biomolecular networks. In this context, bio-molecular network, which is a powerful tool to study complex diseases, enables systematically integration of high throughput biological data and plenty of biological knowledge. Recently, a large number of biological networks have been constructed, including protein interaction network, transcriptional regulation network, signal transduction network, metabolic network and so on. These studies have played a significant role in revealing the pathogeneses of complex diseases and promoting treatment or prevention of diseases. However, the biomolecular network research on complex diseases is still in its infancy. Due to the complexity of mechanisms and disease-related biological data, many challenges remain to be further explored and resolved. In this thesis, we use mathematical models of biomolecular networks to address some main challenges in the research of complex diseases. Specifically, we study molecular networks step by step from static, node dynamic, edge dynamic, to module dynamic of network construction and network analysis, and build several mathematical models to deepen the complex disease study by biomolecular networks concept.

Heterogeneous biological data integration with declarative query language

The requirements for scalable integration systems for modern biology are indisputable, due to the very large, heterogeneous, and complex datasets available in public databases. The management and fusion of this big data with local databases represents a major challenge, since it underlies the computational inferences and models that will be subsequently generated and validated experimentally. In this paper, we present an alternative conception for local integration, called BIRD (Biological Integration and Retrieval Data), based on four concepts: (i) a hybrid flat file and relational database architecture permits the rapid management of large volumes of heterogeneous datasets; (ii) a generic model allows the simultaneous organization and classification of local databases according to real-world requirements; (iii) configuration rules are used to divide and map each resource into several model entities; and (iv) a simple, declarative query language (BIRD-QL) facilitates information extraction from heterogeneous datasets. This flexible, generic design allows the integration of diverse formats in a searchable database with high-level functionalities depending on the specific scientific context. It has been validated in the context of real world projects, notably the SM2PH (Structural Mutation to the Phenotypes of Human Pathologies) project.

GUILDify: a web server for phenotypic characterization of genes through biological data integration and network-based prioritization algorithms

Determining genetic factors underlying various phenotypes is hindered by the involvement of multiple genes acting cooperatively. Over the past years, disease-gene prioritization has been central to identify genes implicated in human disorders. Special attention has been paid on using physical interactions between the proteins encoded by the genes to link them with diseases. Such methods exploit the guilt-by-association principle in the protein interaction network to uncover novel disease-gene associations. These methods rely on the proximity of a gene in the network to the genes associated with a phenotype and require a set of initial associations. Here, we present GUILDify, an easy-to-use web server for the phenotypic characterization of genes. GUILDify offers a prioritization approach based on the protein-protein interaction network where the initial phenotype-gene associations are retrieved via free text search on biological databases. GUILDify web server does not restrict the prioritization to any predefined phenotype, supports multiple species and accepts user-specified genes. It also prioritizes drugs based on the ranking of their targets, unleashing opportunities for repurposing drugs for novel therapies. Available online at http://sbi.imim.es/GUILDify.php