LibKiSAO: a Java library for Querying KiSAO
© Zhukova et al.; licensee BioMed Central Ltd. 2012
Received: 10 November 2011
Accepted: 17 September 2012
Published: 24 September 2012
The Kinetic Simulation Algorithm Ontology (KiSAO) supplies information about existing algorithms available for the simulation of Systems Biology models, their characteristics, parameters and inter-relationships. KiSAO enables the unambiguous identification of algorithms from simulation descriptions. Information about analogous methods having similar characteristics and about algorithm parameters incorporated into KiSAO is desirable for simulation tools. To retrieve this information programmatically an application programming interface (API) for KiSAO is needed.
We developed libKiSAO, a Java library to enable querying of the KiSA Ontology. It implements methods to retrieve information about simulation algorithms stored in KiSAO, their characteristics and parameters, and methods to query the algorithm hierarchy and search for similar algorithms providing comparable results for the same simulation set-up. Using libKiSAO, simulation tools can make logical inferences based on this knowledge and choose the most appropriate algorithm to perform a simulation. LibKiSAO also enables simulation tools to handle a wider range of simulation descriptions by determining which of the available methods are similar and can be used instead of the one indicated in the simulation description if that one is not implemented.
LibKiSAO enables Java applications to easily access information about simulation algorithms, their characteristics and parameters stored in the OWL-encoded Kinetic Simulation Algorithm Ontology. LibKiSAO can be used by simulation description editors and simulation tools to improve reproducibility of computational simulation tasks and facilitate model re-use.
KeywordsJava library Simulation Algorithm search Ontology
Since KiSAO is implemented in OWL 2 , developers of libraries and tools processing simulation description languages could potentially work with KiSAO using general programming interfaces for manipulating ontologies, such as OWL API , and ontology reasoners, such as HermiT . Richer programming interfaces for interacting with ontology resources are also available, e.g., OntoCAT , an application suite enabling the querying of several resources simultaneously, looking for synonyms and definitions (which is not supported directly by OWL API). Nevertheless, due to the complex structure and variety of relationships used in KiSAO, none of these interfaces provide an easy way to retrieve all the information from KiSAO. Moreover, some operations require the knowledge of how KiSAO is organised. For instance, to find similar simulation algorithms both the hierarchical structure of the algorithm sub-tree of KiSAO, and algorithm-characteristic relationships should be taken into account. It is more efficient and convenient for developers to use a higher-level, KiSAO-specific API. We therefore developed libKiSAO to address this need.
LibKiSAO is a Java library which uses the OWL API  and the HermiT reasoner  to implement an intuitive interface for querying KiSAO. It is available under the LGPL license for anyone to use and extend. We chose the Java language for the implementation because of its platform independence and wide adoption. The compatibility requirements of the Hermit reasoner restricted us to Java version 1.5 or higher.
LIbKiSAO consists of two parts. The first part is a general purpose ontology querying library libOnto, wrapping some of the OWL API low-level functionality to implement shorter methods, which are more clear for non-ontology specialists. Examles are getSynonyms(), getPropertyValues(), searchByName(), which are not implemented natively in the OWL API but are often used while working with ontologies. LibOnto can be used separately from libKiSAO for various ontology-querying purposes and is independent of any knowledge about the KiSA Ontology. An Ant build script which packs libOnto into a separate jar archive is available for download from the project website .
Taxonomy level querying methods implemented in libKiSAO
Lists simulation algorithms stored in KiSAO
Lists simulation algorithm characteristics stored in KiSAO
Lists simulation algorithm parameters stored in KiSAO
Returns type (algorithm, characteristic or parameter) of the
Checks if the specified entry represents simulation algorithm
Checks if the specified entry represents simulation algorithm
characteristic in KiSAO
Checks if the specified entry represents simulation algorithm
parameter in KiSAO
Looks for the entry with the specified identifier or miriam urn
Looks for the entries with the specified name or synonym
Retrieving Element Information
Returns name of the specified entry
Returns synonyms of the specified entry
Returns synonyms of the specified type (exact, related,
broad or narrow)
Returns definition of the specified entry
Returns links to articles/books that describe the specified entry
Returns id (e.g. kisao:0000001) of the specified entry
Returns MIRIAM URN (e.g. urn: miriam: biomodels.kisao:
KISAO_0000001) of the specified entry
Returns identifiers.org URL of the specified entry
Checks whether the specified entry is deprecated
Returns ancestors of the specified entry
Returns descendants of the specified entry
Checks if a descendantCandidate entry is a descendant
(concerning rdf:subClassOf relationship) of the
ancestorCandidate entry in KiSAO
Complex/application level querying methods implemented in libKiSAO
Querying for Algorithm Characteristics
getCharacteristics(algorithm, optional type(s))
Returns characteristics of the specified algorithm. An optional
type(s) parameter filters the returned characteristic set by
Checks if the specified algorithm possesses the specified
Looking for Similar Algorithms
Returns algorithms which possess specified characteristic(s)
(algorithm, optional type(s))
Returns algorithms having the same characteristics as the
specified one. An optional type(s) parameter specifies which
types of characteristics should be considered. By default,
all the characteristics are considered
Returns n algorithms which are most similar to the specified
one, sorted by the distance in the hierarchy tree. An
optional type(s) parameter specifies which types of
characteristics should be considered when looking for
similar algorithms. By default, all the characteristics
Returns algorithms described by the specified query
Querying for Algorithm Parameters
Returns parameters that are used by the specified algorithm
Returns parameters that are used by the algorithms with the
Checks if the algorithm described by its characteristics uses
the specified parameter(s)
Returns a set of parameters that are used by the algorithms
with the specified characteristic(s) and ancestor
Checks if the specified algorithm uses the specified
Returns the type of the specified parameter
Querying for Inter-algorithm Relationships
Checks whether the specified algorithm is a hybrid one
Returns collection of the algorithms the specified one is a
KiSAO-specific library part implements such methods as looking for algorithms having particular characteristics or algorithm grouping methods with similarity distance based on algorithm characteristics and hierarchy relationships.
A user can specify which characteristics should be considered while looking for similar methods. For example, getNMostSimilarAlgorithms(poissonTauLIri, 4, KiSAOIRI.TYPE_OF_SYSTEM_BEHAVIOUR_IRI, KiSAOIRI.TYPE_OF_VARIABLE_IRI) returns 4 algorithms having the same type of system behaviour (stochastic) and of variables (discrete) as the Poisson tau-leaping method, and sorted by the hierarchical distance to it: tau-leaping method, multinomial tau-leaping method, binomial tau-leaping method, and implicit tau-leaping method. Other characteristics are not taken into account (e.g. method type can be implicit as well as explicit).
If the distance in the hierarchy tree is not important, the method getAlgorithmsWithSameCharacteristics(IRI.. characteristicType) can be used. By default (if no characteristic types are specified) all the characteristics are looked at.
Currently all the characteristics are equally important for the algorithms finding similar simulation methods. In the future they might be prioritized to improve the similarity measurement.
Code examples and Java documentation are available both for download and on-line from the project website .
Results and discussion
LibKiSAO supports working with KiSAO at taxonomic and complex/application levels . The taxonomic layer represents a hierarchy of simulation algorithms, annotated with the information about their names, definitions, and references to the articles they are described in. For instance, as shown in Figure 1, the tau-leaping method is an ancestor of the binomial tau-leaping method and is annotated with the synonym tauL. The complex and application levels of KiSAO enable algorithm grouping based on characteristics (e.g. stochastic/deterministic system behaviour), link algorithms to the parameters they use (e.g. tau-leaping method ’has parameter’ tau-leaping epsilon), and describe parameter types (e.g. tau-leaping epsilon ’has type’ xsd:double).
LibKiSAO provides Java applications with the mechanisms to retrieve this information programmatically from the OWL-encoded KiSAO. LibKiSAO can be used by simulation description language manipulating libraries, such as jlibSEDML , and simulation tools, such as COPASI .
Taxonomy level querying methods
LibKiSAO provides methods for listing simulation algorithms stored in KiSAO and searching for them by names, synonymous names, MIRIAM URIs  and identifiers. Retrieving algorithm descriptions, synonyms and references to the articles describing the algorithm is also implemented, as well as methods for querying algorithm hierarchy, such as searching for descendant (derived from, more specific) and ancestor (more general) algorithms. The supported taxonomy querying methods are listed in Table 1. Use case 1 describes how querying KiSAO taxonomy level capabilities of libKiSAO can be used by simulation description language editors, such as SED-ED , a tool for editing, viewing and running SED-ML files.
Use Case 1: Using libKiSAO for querying the taxonomy level of KiSAO
Complex/application level querying capabilities
LibKiSAO implements characteristic-based grouping of the simulation algorithms, stored in KiSAO, by providing methods for searching for the algorithms which possess/lack given characteristics, as well as querying for the characteristics of a given algorithm. LibKiSAO also includes methods for retrieving information about simulation algorithm parameters and their types. The complex/application level querying methods available are described in more details in Table 2. These features enable simulation tools dealing with SED-ML descriptions to automatically handle a wider range of simulation tasks, as it is discussed in use case 2.
Use Case 2: Using libKiSAO for querying the complex/application level of KiSAO
In the future versions of SED-ML not only defining an identifier of a simulation algorithm, but also its parameters and their values might be supported . In this case libKiSAO methods for extracting KiSAO knowledge about parameters common for different algorithms, will help simulation tools to switch between similar algorithms easier.
LibKiSAO therefore provides simulation tools with additional knowledge about available simulation methods, allowing them to automatically handle even the simulation descriptions referring to algorithms they cannot access.
The ability to choose related algorithms is also desirable in a web-service based environment. Since a particular solver service may not always be available, it is important to determine ones which can be used instead. LibKiSAO functionality to indicate similar algorithms can help to search for alternatives from a list of simulation providers, such as Biocatalogue , the life science web services registry.
We have developed a Java library for querying the Kinetic Simulation Algorithm Ontology. LibKiSAO helps modellers to encode their simulation tasks in unambiguous languages, such as SED-ML, which can be understood by other modellers and also by simulation tools. It enables accurate and repeatable execution of computational simulation tasks and facilitates the re-usability of models in Systems Biology.
Availability and requirements
Project name: libKiSAOProject home page:http://biomodels.net/kisao/libkisao.htmlOperating system(s): Platform independentProgramming language: JavaOther requirements: Java 1.5 or higherLicense: GNU LGPL version 3 or higherAny restrictions to use by non-academics: no restrictions
Application programming interface
Kinetic Simulation Algorithm Ontology
Livermore Solver for Ordinary Differential Equations
Minimal Information About a Simulation Experiment
Web Ontology Language
Simulation Experiment Description Markup Language.
This work was supported by the European Molecular Biology Laboratory.
- Waltemath D, Adams R, Beard D, Bergmann F, Bhalla U, Britten R, Chelliah V, Cooling M, Cooper J, Crampin E, Garny A, Hoops S, Hucka M, Hunter P, Klipp E, Laibe C, Miller A, Moraru I, Nickerson D, Nielsen P, Nikolski M, Sahle S, Sauro H, Schmidt H, Snoep J, Tolle D, Wolkenhauer O, Le Novère N: Minimum Information About a Simulation Experiment (MIASE). PLoS Comput Biol. 2011, 7: e1001122-10.1371/journal.pcbi.1001122.PubMedPubMed CentralView ArticleGoogle Scholar
- Courtot M, Juty N, Knüpfer C, Waltemath D, Zhukova A, Dräger A, Dumontier M, Finney A, Golebiewski M, Hastings J, Hoops S, Keating S, Kell D, Kerrien S, Lawson J, Lister A, Lu J, Machne R, Mendes P, Pocock M, Rodriguez N, Villeger A, Wilkinson D, Wimalaratne S, Laibe C, Hucka M, Le Novère N: Controlled vocabularies and semantics in Systems Biology. Mol Syst Biol. 2011, 7: 543-PubMedPubMed CentralView ArticleGoogle Scholar
- Waltemath D, Adams R, Bergmann F, Hucka M, Kolpakov F, Miller A, Moraru I, Nickerson D, Sahle S, Snoep J, Le Novere N: Reproducible computational biology experiments with SED-ML - The Simulation Experiment Description Markup Language. BMC Syst Biol. 2011, 5: 198-10.1186/1752-0509-5-198. [http://www.biomedcentral.com/1752-0509/5/198]PubMedPubMed CentralView ArticleGoogle Scholar
- W3C: OWL 2 Web Ontology Language: Structural Specification and Functional-Style Syntax. [http://www.w3.org/TR/owl2-syntax/]
- Horridge M, Bechhofer S: The OWL API: a Java API for OWL Ontologies. Semant Web J Spec Issue on Semant Web Tools Syst. 2011, 2 (1): 11-21.Google Scholar
- Motik B, Shearer R, Horrocks I: Optimized reasoning in description logics using hypertableaux. Proceedings of the 21st Int. Conf. on Automated Deduction (CADE-21), Lecture Notes in Artificial Intelligence. 2007,, Springer Berlin Heidelberg, 4603: 67-83.Google Scholar
- LibKiSAO web page. [http://www.biomodels.net/kisao/libkisao.html]
- KiSAO URI. [http://www.biomodels.net/kisao/KISAO]
- Dumontier M, Villanueva-Rosales N: Three-Layer OWL Ontology Design. 2nd International Workshop on Modular Ontologies (WOMO07). Co-located with 4th International Conference on Knowledge Capture (K-CAP2007.). 2007, Whistler: Canada,Google Scholar
- Adams R, Moraru I, Lakshminaryana A: jlibSEDML — a Java library for working with SED-ML. 2010, 10.1038/npre.2010.4965.1.Google Scholar
- Hoops S, Sahle S, Gauges R, Lee C, Pahle J, Simus N, Singhal M, Xu L, Mendes P, Kummer U: COPASI — a COmplex PAthway SImulator. Bioinformatics. 2006, 22: 3067-3074. 10.1093/bioinformatics/btl485.PubMedView ArticleGoogle Scholar
- Laibe C, Le Novère N: MIRIAM Resources: tools to generate and resolve robust cross-references in Systems Biology. BMC Syst Biol. 2007, 1: 58-10.1186/1752-0509-1-58.PubMedPubMed CentralView ArticleGoogle Scholar
- Adams R: SED-ED, a workflow editor for computational biology experiments written in SED-ML. Bioinformatics. 2012, 28 (8): 1180-1181. 10.1093/bioinformatics/bts101.PubMedView ArticleGoogle Scholar
- Adams R: Algorithm class proposal, version 1. 2011, [http://identifiers.org/combine.specifications/sed-ml.proposal.kisao.RA.version-1]Google Scholar
- Bhagat J, Tanoh F, Nzuobontane E, Laurent T, Orlowski J, Roos M, Wolstencroft K, Aleksejevs S, Stevens R, Pettifer S, Lopez R, Goble AC: BioCatalogue: a universal catalogue of web services for the life sciences. Nucleic Acids Res. 2010, 38 (suppl 2): W689-W694.PubMedPubMed CentralView ArticleGoogle Scholar