In this paper we present SysBio-Gateway, a framework of standard solution for data integration in bioinformatics and Systems Biology. In the context of several research projects, we developed a set of databases and related web interfaces to cover many levels of the biological system complexity. Furthermore several analysis tools have been developed and integrated in the web resources in order to cope with problems of data integration and mining in a systems perspective. The SysBio-Gateway, freely accessible at the URL http://www.itb.cnr.it/sysbio-gateway, offers access to all the presented resources, which concerns different levels of organization of biological living systems, from genes to organs, passing through proteins, protein families, cellular processes, tissues, and pathologies.
Data integration is nowadays an essential task to accomplish in order to achieve a view of the biological knowledge as much complete as possible. This is very important in a discipline, such as the bioinformatics one, in which data are growing at lightning speed thanks to novel bio-molecular high-throughput techniques. In particular, considering the systems biology field, the integration of biological knowledge related to different levels such as genomics, transcriptomics, proteomics, and network interactions – is crucial in order to support the mathematical modelling and the computer simulation of biological pathways.
Data integration can be defined as the process of combining information, residing at different sources, to provide the user with a unified view of these data for enabling the possibility to achieve real knowledge. Experimental researchers and computer scientists can discover through data integration new and interesting relationships that allow better and faster experimental decisions, for example about protein targets and drug molecules. Moreover, the achievement of interesting results in most bioinformatics and *
To whom correspondence should be addressed. systems biology-related activities, from functional characterization of genomic and proteomic data to the development of mathematical models of biological processes, requires an integrated view of all relevant information useful to accomplish those tasks.
Data integration problem can be described by three levels of complexity. The first layer is the integration of information from heterogeneous resources by collecting data between different databases to allow a unified query schema. The second level consists in identifying correlative associations across different datasets, generally using an ontology support, to provide a comprehensive view of the same objects in light of different data sources. The third layer is mapping the information gained about interacting objects into networks and pathways that may be used as basic models for the underlying cellular systems.
In this paper we show an overview of solutions, developed by our bioinformatics laboratory, for data integration in the context of bioinformatics and Systems Biology, where many levels of complexity are covered: from proteins to cellular processes, from tissues to disease and organs. A set of databases and web resources oriented to different biological topics are presented here in a unified framework, to provide some practical examples of our experience in this field. In particular we present resources related to specific a) biological processes, such as cell cycle, b) pathologies, such as breast cancer, c) organs, such as brain, d) protein families, such as protein kinases, e) protein mutations, and f) tissues. The here proposed solutions are based on a common methodology of data integration suitable we adoptedto support different biological projects. Several tools have been developed ad hoc to cope with different problems arising both from the increasing number of experimental data and from the need to improve the knowledge in a systems perspective.
Web resources are presented here in a unified framework, to provide some practical examples of our experience in this field. We present resources related to specific knowledge about a) biological processes, such as cell cycle, b) pathology, such as breast cancer, c) organs, such as brain, d) protein family, such as protein kinases, e) protein mutations, and f) tissue.
All the presented databases rely on a data warehouse approach, which requires collection and transformation of heterogeneous data coming from different sources to make them accessible by the scientific community though a unified query schema. This model is typical of data integration and differs from the normalized databases, designed to support data integrity, which are widely used to maintain primary resources. While many data warehouses solutions used in bioinformatics provide generic query interfaces applicable to all the data they contain, our system allows the construction of queries that extract and filter information derived from the original resources. This integration improves the usability of the information, but data must be fitted into unique format that takes into account the relationships between the different sources.
The implementation of our relational databases is always managed by a MySQL server. The primary data are collected with a series of Perl scripts which retrieve data from external resources, transform them in a compliant format and load them into the warehouse data model. The developed resources are all freely accessible through web interfaces which are made up of a set of HTML pages dynamically generated from PHP scripts, in order to provide information in specific reports created for responding to specific requirements related to the biological problem of interest. Besides the integration problem, the analysis of large quantity of experimental data in a Systems Biology perspective has been tackled by developing a number of specific tools integrated in the web resources. Some examples available through the SysBioGateway are: • • • • • a high performance tool for the simulation of Ordinary Differential Equation relying on mathematical models based on Xppaut a tool for the visualization of the Protein Data Bank protein structure and correlated Connolly surfaces a set of tools for the analysis of protein-protein interactions network (search for the first neighbourhood, search for shortest path and common annotations) a tool for modelling the protein mutant starting from Single Nucleotide Polymorphism data which relies on Modeller a tool for pathological image processing oriented to support tissue microarray analysis.
In order to guarantee data sharing and structuring, developed databases are enriched by a underlying crucial feature: the ontological support. Exploited ontologies concern all levels of molecular biology, from genes to proteins to pathways, even covering tissues and diseases aspects. As example, we used Gene Ontology for genes annotation and KEGG Pathway Ontology (derived from the hierarchical organization of KEGG pathways) for biological networks. For their intrinsic structure, that is more complex than a simple recognized vocabulary, ontologies are used in the developed resources to enrich to the informative content. Ontologies provides not only the availability of a commonly accepted vocabulary, which facilitates data sharing and information querying, but also increases the performance of statistical and analytical studies. The hierarchical graphs, which represent ontologies backbone, can in fact support the generation of novel knowledge, leading to the creation of new relationships among biological entities or to the deduction of new associations.
In the context of Systems Biology, another service provided by our infrastructure is a high performance framework for the numerical simulation of molecular models and an associated parameters values estimation tool. In fact, uncertainty of parameters values is one of the greatest problem in the development of new cellular models and in silico parameters estimation is one of the most common adopted solutions. Our system provides a global optimization algorithm, which relies on an evolution strategy, that can be employed on the top of the simulation engine to accomplish parameter estimation of complex mathematical models. Due to the high computational load needed for the parameter estimation of large networks, we implement this system using a distributed paradigm which can also be used in the context of grid computing technologies. Relying on a data parallel approach it is possible to handle the parameters estimation of complex models using large computational facilities, which is very useful for example in the case of cell cycle related models. 3
We propose here a standard solution for the implementation of bioinformatics framework, by collecting our experiences in the SysBio-Gateway, which embraces different levels of organization of biological living systems by linking, integrating and analysing heterogeneous data that vertically lead from genes to organs, passing through proteins, protein families, cellular processes, tissues, and pathologies. This approach has been tested for different databases implementations, such as: • • • • • •
Cell Cycle G2S Breast Cancer Gene Nerve Cell Kinweb ProCMD
TMA Rep SysBio-Gateway is freely accessible at the URL, http://www.itb.cnr.it/sysbio-gateway. In this page (Fig.1) the user can directly access the resources held in the gateway web page, handle and analyse data according to specific demands coming from the bioinformatics and the Systems Biology communities. 3.1
The cell cycle is one of the biological processes most frequently investigated in systems biology studies and it involves the knowledge of a large number of genes and networks of protein interactions. A deep knowledge of the molecular aspect of this biological process can contribute to making cancer research more accurate and innovative. In this context the mathematical modelling of the cell cycle has a relevant role to quantify the behaviour of each components of the systems. The mathematical modelling of a biological process such as the cell cycle allows a systemic description that helps to highlight some features such as emergent properties which could be hidden when the analysis is performed only from a reductionist point of view. Moreover, in modelling complex systems, a complete annotation of all the components is equally important to understand the interaction mechanism inside the network: for this reason data integration of the model components has high relevance in systems biology studies.
The Cell Cycle Database [
This integrated system aims to become a useful resource for collecting all the information related to actual and future models of this network. The flexibility of the database allows the addition of mathematical data which are used for simulating the behaviour of the cell cycle components in the different models. Coupling structure and dynamical information about models Cell Cycle Database allows to achieve system-level properties, such as stable steady states and oscillations. 3.2
The study of breast cancer and its development involves the
knowledge of a large number of genes and molecular interactions
and thus the systems biology approach is essential to describe the
processes related to the pathology and to perform useful
predictions. For the effective application of the systemic approach it is
essential to arrange information about genes, cellular pathways and
interactions that they undertake. These annotations are publicly
available in bioinformatics resources and their integration produces
an information enrichment, allowing, for example, the clustering of
breast cancer genes by common signatures or the suggestion of
possible annotations for not yet annotated genes. The
Genes-toSystems Breast Cancer (G2SBC) Database [
In the past few years the new research field of neuroinformatics has strongly emerged. Two main aspects must be highlighted in his context: the interplay of structural, chemical and electrical signals in nervous tissue and the importance of modelling such signals. The great amount of qualitative experimental data in neuroscience represents the starting point to expand the nerve cell modelling in new directions, especially in the development of gene and protein interaction networks. The aim of this new discipline is thus to gather the application of computational models and analytical tools and the improvement of neuroscience knowledge. Moreover in the nervous system a great importance must be given to the study of the molecular processes, and the knowledge about the key players (genes and proteins) involved in such processes should be as much complete as possible. The study of nerve cells, neurons and, more in general, brain and its development involves information regarding a large number of genes and molecular interactions. Thus the systems biology approach is essential. Starting from the integration of genes and proteins interaction data with experimental data it is possible to develop new discovery strategies in brain studies.
In this context we propose a new data integration system,
the “Gene Nerve Cell Database” [
Protein kinases are a well defined family of proteins, characterized by the presence of a common kinase catalytic domain and playing a significant role in many important cellular processes, such as proliferation, maintenance of cell shape, apoptosys. In many members of the family, additional non-kinase domains contribute to further specialization, resulting in subcellular localization, protein binding and regulation of activity, among others.
About 500 genes encode members of the kinase family in the human genome, and although many of them represent well known genes, a larger number of genes code for proteins of more recent identification, or for unknown proteins identified as kinase only after computational studies.
A systematic in silico study performed on the human genome, led to the identification of 5 genes, on chromosome 1, 11, 13, 15 and 16 respectively, and 1 pseudogene on chromosome X; some of these genes are reported as kinases from NCBI but are absent in other databases, such as KinBase. Comparative analysis of 483 gene regions and subsequent computational analysis, aimed at identifying unannotated exons, indicate that a large number of kinase may code for alternately spliced forms or be incorrectly annotated. An InterProScan automated analysis was perfomed to study domains distribution and combination in the various families. At the same time, other structural features were also added to the annotation process, including the putative presence of transmembrane alpha helices, and the cystein propensity to participate into a disulfide bridge.
The predicted human kinome was extended by identifying
both additional genes and potential splice variants, resulting in a
varied panorama where functionality may be searched at the gene
and protein level. Structural analysis of kinase proteins domains as
defined in multiple sources together with transmembrane alpha
helices and signal peptide prediction provides hints to function
assignment. The results of the human kinome analysis are collected
in the KinWeb database [
Activated Protein C (ProC) is an anticoagulant plasma serine protease which also plays an important role in controlling inflammation and cell proliferation. Several mutations of the gene are associated with phenotypic functional deficiency of protein C, and with the risk of developing venous thrombosis. Structure prediction and computational analysis of the mutants have proven to be a valuable aid in understanding the molecular aspects of clinical thrombophilia. A specialized relational database and a search tool for natural mutants of protein C have been built. The database contains 195 entries that include 182 missense and 13 stop mutations. A menu driven search engine allows the user to retrieve stored information for each variant, that include genetic as well as structural data and a multiple alignment highlighting the substituted position. Molecular models of variants can be visualized with interactive tools; PDB coordinates of the models are also available for further analysis. Furthermore, an automatic modelling interface allows the user to generate multiple alignments and 3D models of new variants.
ProCMD [
TMA Rep
Tissue MicroArray technique is becoming increasingly important in pathology for the validation of experimental data from transcriptomics analysis. This approach produces many images which need to be properly managed, if possible exploiting an infrastructure able to support tissue sharing between institutes. Moreover, the nowadays available frameworks oriented to Tissue MicroArray provide good storage for clinical patients, sample treatments and block constructions information, but their utility is limited by the lack of data integration with bioinformatic approaches.
We propose a Tissue MicroArray web oriented system
[
Through this system, users associate an ontology-based description to each image uploaded into the database and also integrate results with the ontological descriptions of biosequences identified in each tissue. It is even possible to integrate the ontological description provided by the user with a fully compliant gene ontology definition, enabling statistical studies about correlation between the analyzed pathology and the most commonly related biological processes. Finally, the web site embeds a tool oriented to pre-array tissue image analysis, specific for tubular breast cancer affected tissues. 4
In this paper we present an integrated solution to explore part of the information gained in the field of life science oriented to systems biology. In order to achieve a systemic perspective of a set interesting topics for our group, we come to this integrated portal, SysBio-Gateway, which combines a bioinformatics approach, i.e. data integration using data warehouse approach, application of tools for the data analysis, study of structural modifications - both for genome and proteins -, and systems biology approach, that is the study of protein-protein interaction networks, molecular mathematical models, pathological states, under a systemic point of view.
This work has been supported by the NET2DRUG, EGEE-III, BBMRI, EDGE European projects and by the MIUR FIRB LITBIO (RBLA0332RH), ITALBIONET (RBPR05ZK2Z), BIOPOPGEN (RBIN064YAT), CNR-BIOINFORMATICS initiatives. We also acknowledge the support of the e-Science Institute in Edinburgh.