Limits...
Dynamic integration of biological data sources using the data concierge.

Gong P - Health Inf Sci Syst (2013)

Bottom Line: Experimental results demonstrate that for obtaining dynamic features, the Data Concierge sacrifices reasonable performance on reasoning knowledge models and dynamically doing data source API invocations.The overall costs to integrate new biological data sources are significantly lower when using the Data Concierge.The Data Concierge facilitates the rapid integration of new biological data sources in existing applications with no repetitive software development required, and hence, this mechanism would provide a cost-effective solution to the labor-intensive software engineering tasks.

View Article: PubMed Central - PubMed

Affiliation: Biomedical and Multimedia Information Technology (BMIT) Research Group, School of Information Technologies, the University of Sydney, Sydney, NSW 2006 Australia ; Department of PET and Nuclear Medicine, RPA Hospital, Camperdown, NSW 2050 Australia.

ABSTRACT

Background: The ever-changing landscape of large-scale network environments and innovative biology technologies require dynamic mechanisms to rapidly integrate previously unknown bioinformatics sources at runtime. However, existing integration technologies lack sufficient flexibility to adapt to these changes, because the techniques used for integration are static, and sensitive to new or changing bioinformatics source implementations and evolutionary biologist requirements.

Methods: To address this challenge, in this paper we propose a new semantics-based adaptive middleware, the Data Concierge, which is able to dynamically integrate heterogeneous biological data sources without the need for wrappers. Along with the architecture necessary to facilitate dynamic integration, API description mechanism is proposed to dynamically classify, recognize, locate, and invoke newly added biological data source functionalities. Based on the unified semantic metadata, XML-based state machines are able to provide flexible configurations to execute biologist's abstract and complex operations.

Results and discussion: Experimental results demonstrate that for obtaining dynamic features, the Data Concierge sacrifices reasonable performance on reasoning knowledge models and dynamically doing data source API invocations. The overall costs to integrate new biological data sources are significantly lower when using the Data Concierge.

Conclusions: The Data Concierge facilitates the rapid integration of new biological data sources in existing applications with no repetitive software development required, and hence, this mechanism would provide a cost-effective solution to the labor-intensive software engineering tasks.

No MeSH data available.


Tests on getDSMetadata.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4340781&req=5

Fig18: Tests on getDSMetadata.

Mentions: In Figure 18, the performance of getDSMetadata varies for different data sources. It is decided by how many meta attributes in Generic API Ontology are associated with each data source instance. If one classified data source instance has more associated meta attributes such as data elements and related user operations than the others, Data Concierge will spend a much longer time on querying these metadata from Generic API Ontology. For example, the performance of getting metadata of an FTP data source, which includes data elements File and Directory and their user operations (such as Read, Write, Delete, Up_Navigation, Down_Navigation) and other attributes (such as name, size, date, type, userID, groupID, permissions, numberofLinks, etc.) is 0.0034 ms. While querying metadata for a simple SMTP mail server that has Mail element, Write operation, and some simple attributes, is only 0.0014 ms.Figure 18


Dynamic integration of biological data sources using the data concierge.

Gong P - Health Inf Sci Syst (2013)

Tests on getDSMetadata.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4340781&req=5

Fig18: Tests on getDSMetadata.
Mentions: In Figure 18, the performance of getDSMetadata varies for different data sources. It is decided by how many meta attributes in Generic API Ontology are associated with each data source instance. If one classified data source instance has more associated meta attributes such as data elements and related user operations than the others, Data Concierge will spend a much longer time on querying these metadata from Generic API Ontology. For example, the performance of getting metadata of an FTP data source, which includes data elements File and Directory and their user operations (such as Read, Write, Delete, Up_Navigation, Down_Navigation) and other attributes (such as name, size, date, type, userID, groupID, permissions, numberofLinks, etc.) is 0.0034 ms. While querying metadata for a simple SMTP mail server that has Mail element, Write operation, and some simple attributes, is only 0.0014 ms.Figure 18

Bottom Line: Experimental results demonstrate that for obtaining dynamic features, the Data Concierge sacrifices reasonable performance on reasoning knowledge models and dynamically doing data source API invocations.The overall costs to integrate new biological data sources are significantly lower when using the Data Concierge.The Data Concierge facilitates the rapid integration of new biological data sources in existing applications with no repetitive software development required, and hence, this mechanism would provide a cost-effective solution to the labor-intensive software engineering tasks.

View Article: PubMed Central - PubMed

Affiliation: Biomedical and Multimedia Information Technology (BMIT) Research Group, School of Information Technologies, the University of Sydney, Sydney, NSW 2006 Australia ; Department of PET and Nuclear Medicine, RPA Hospital, Camperdown, NSW 2050 Australia.

ABSTRACT

Background: The ever-changing landscape of large-scale network environments and innovative biology technologies require dynamic mechanisms to rapidly integrate previously unknown bioinformatics sources at runtime. However, existing integration technologies lack sufficient flexibility to adapt to these changes, because the techniques used for integration are static, and sensitive to new or changing bioinformatics source implementations and evolutionary biologist requirements.

Methods: To address this challenge, in this paper we propose a new semantics-based adaptive middleware, the Data Concierge, which is able to dynamically integrate heterogeneous biological data sources without the need for wrappers. Along with the architecture necessary to facilitate dynamic integration, API description mechanism is proposed to dynamically classify, recognize, locate, and invoke newly added biological data source functionalities. Based on the unified semantic metadata, XML-based state machines are able to provide flexible configurations to execute biologist's abstract and complex operations.

Results and discussion: Experimental results demonstrate that for obtaining dynamic features, the Data Concierge sacrifices reasonable performance on reasoning knowledge models and dynamically doing data source API invocations. The overall costs to integrate new biological data sources are significantly lower when using the Data Concierge.

Conclusions: The Data Concierge facilitates the rapid integration of new biological data sources in existing applications with no repetitive software development required, and hence, this mechanism would provide a cost-effective solution to the labor-intensive software engineering tasks.

No MeSH data available.