Lena Strömbäck

Representations of molecular pathways: an evaluation of SBML, PSI MI and BioPAX

MOTIVATION Analysis and simulation of pathway data is of high importance in bioinformatics. Standards for representation of information about pathways are necessary for integration and analysis of data from various sources. Recently, a number of representation formats for pathway data, SBML, PSI MI and BioPAX, have been proposed. RESULTS In this paper we compare these formats and evaluate them with respect to their underlying models, information content and possibilities for easy creation of tools. The evaluation shows that the main structure of the formats is similar. However, SBML is tuned towards simulation models of molecular pathways while PSI MI is more suitable for representing details about particular interactions and experiments. BioPAX is the most general and expressive of the formats. These differences are apparent in allowed information and the structure for representation of interactions. We discuss the impact of these differences both with respect to information content in existing databases and computational properties for import and analysis of data.

Challenges for modeling and simulation methods in systems biology

Systems biology is aimed at analyzing the behavior and interrelationships of biological systems and is characterized by combining experimentation, theory, and computation. Dedicated to exploring current challenges, the panel brings together people from a variety of disciplines whose perspectives illuminate diverse facets of systems biology and the challenges for modeling and simulation methods

Information integration in bioinformatics with ontologies and standards

New experimental methods allow researchers within molecular and systems biology to rapidly generate larger and larger amounts of data. This data is often made publicly available on the Internet and although this data is extremely useful, we are not using its full capacity. One important reason is that we still lack good ways to connect or integrate information from different resources. One kind of resource is the over 1000 data sources freely available on the Web. As most data sources are developed and maintained independently, they are highly heterogeneous. Information is also updated frequently. Other kinds of resources that are not so well-known or commonly used yet are the ontologies and the standards. Ontologies aim to define a common terminology for a domain of interest. Standards provide a way to exchange data between data sources and tools, even if the internal representations of the data in the resources and tools are different. In this chapter we argue that ontological knowledge and standards should be used for integration of data. We describe properties of the different types of data sources, ontological knowledge and standards that are available on the Web and discuss how this knowledge can be used to support integrated access to multiple biological data sources. Further, we present an integration approach that combines the identified ontological knowledge and standards with traditional information integration techniques. Current integration approaches only cover parts of the suggested approach. We also discuss the components in the model on which much recent work has been done in more detail: ontology-based data source integration, ontology alignment and integration using standards. Although many of our discussions in this chapter are general we exemplify mainly using work done within the REWERSE working group on Adding Semantics to the Bioinformatics Web.

Using Mediation to Achieve Provenance Interoperability

Provenance is essential in scienti¿c experiments. It contains information that is key to preserving the data, to determining their quality and authorship, and to reproduce as well as validate the results. In complex experiments and analyses, where multiple tools are used to derive data products, provenance captured by these tools must be combined in order to determine the complete lineage of the derived products. In this paper we describe a mediator-based architecture for integrating provenance information from multiple sources. This architecture contains two key components: a global mediated schema that is general and capable of representing provenance information represented in different model; and describe a new system-independent query API that is general and able to express complex queries over provenance information from different sources. We also present a case study where we show how this model was applied to integrate provenance from three provenance-enabled systems and discuss the issues involved in this integration process.

An evaluation of the use of XML for representation, querying, and analysis of molecular interactions

Currently, biology researchers rapidly generate new information on how genes, proteins and other molecules interact in living organisms. To completely understand the machinery underlying life it is necessary to integrate and analyze these large quantities of data. As one step in this direction, new standards for describing molecular interactions have been defined based on XML. This work evaluates the usage of the XML Query language XQuery for molecular interactions, as it would be of great benefit to the user to work directly on data represented in the new standards. We use and compare a set of available XQuery implementations, eXist, X-Hive, Sedna and QizX/open for querying and analysis on data exported from available databases. Our conclusion is that XQuery can easily be used for the most common queries in this domain but is not feasible for more complex analyses. In particular, for queries containing path analysis the available XQuery implementations have poor performance and an extension of the GTL package clearly outperforms XQuery. The paper ends with a discussion regarding the usability of XQuery in this domain. In particular we point out the need for more efficient graph handling and that XQuery also requires the user to understand the exact XML format of each dataset.

Possibilities and Challenges Using XML Technology for Storage and Integration of Molecular Interactions

The amount of data emerging from experiments on molecular and protein interactions is rapidly growing. It is of high importance to understand these datasets, putting them together to meaningful pathways describing some aspects of the cell physiology. Currently, there is a big interest in XML representations of protein interaction and molecular pathways. New standards, SBML, PSI MI and BioPAX are under development and it is of high importance to build tools for easy integration and analysis of data represented in these standards. There are currently two main approaches for storage of data represented within XML, one is to base the tools directly on XML while the other is to translate the information into a relational database. In this paper we report on experiments with these approaches for storage and analysis of molecular interaction data represented in the standards. The results give important pointers to the possibility of building easy to use tools

An Extension of XQuery for Graph Analysis of Biological Pathways

The vast quantity of scientific data produced in life sciences demands the use of sophisticated storage and analysis techniques. In particular, for biological pathways graph analysis plays an important role and data is commonly available in XML-based formats. Thus, there is a growing need to make analysis capabilities available through query languages for XML. This paper presents an approach to extend XQuery for graph analysis with focus on data for biological pathways. A graph model is introduced within the XQuery environment. New built-in functions define the available operations on the graph model. XQuery expressions can be utilized to populate graphs with data and execute graph algorithms. Graph data and results of algorithms can be accessed in an XML representation for further processing. In addition, a reference mechanism can be used to preserve associations from graph data to the original XML data. The approach has been implemented as an extension to exist. First evaluations of the implementation show that the introduced approach is practical and efficient for reaction networks with several thousand vertices and edges.

A first study on strategies for generating workflow snippets

Workflows are increasingly being used to specify computational tasks, from simulations and data analysis to the creation of Web mashups. Recently, a number of public workflow repositories have become available, for example, myExperiment for scientific workflows, and Yahoo! Pipes. Workflow collections are also commonplace in many scientific projects. Having such collections opens up new opportunities for knowledge sharing and re-use. But for this to become a reality, mechanisms are needed that help users explore these collections and locate useful workflows. Although there has been work on querying workflows, not much attention has been given to presenting query results. In this paper, we take a first look at the requirements for workflow snippets and study alternative techniques for deriving concise, yet informative snippets.

HShreX - A Tool for Design and Evaluation of Hybrid XML Storage

XML is a commonly used data representation format for web applications. One of the reasons for the attractiveness of XML is its flexibility to store unstructured, semi-structured and structured data.However, supporting this flexibility is challenging from a technical perspective and several approaches have been proposed for storage of XML. The focus of this paper is hybrid storage, combining relational and native solutions as it allows many alternatives for designing the data model. The paper presents our tool HShreX that enables quick design and evaluation of alternative choices by annotating the XML schema.The main benefit of the tool is that it is easy to work with and provides quick design of storage model and import of data. The paper describes the main features of the tool and a use case where we discuss how the tool can be used to design, compare and evaluate storage alternatives.

A method for comparison of standardized information within systems biology

Standards and standardized data representation to allow efficient exchange of information is an important topic within systems biology. Within this area there is currently a rapid development of new standards as well as a need for import of datasets into various computer tools for further analysis. As the number of available standards within systems biology is large, tools for comparison and translation of standards are of high interest. In this paper we present a method for comparison of standards. We illustrate how the method works by providing an analysis of the three standards SBML, PSI MI and BioPAX. The analysis gives information on how similar the three standards are and it also gives pointers on how to build tools to aid a user in the analysis of a standard.